Spir2018 PDF

Copyright © Common Cause & CSDS 2018
29 Rajpur Road, Civil Lines, Delhi -110054

Phone: +91-11-23942199 Fax : +91-11-23943450
Email: [email protected]; [email protected]
Website: www.csds.in; www.lokniti.org
Common Cause House, 5, Institutional Area,

Nelson Mandela Road, Vasant Kunj, New Delhi –110 070. Phone: 26131313 www.commoncauseindia.in;
e-mail: [email protected]; [email protected]
Published by
Common Cause & Lokniti - Centre for the Study Developing Societies (CSDS)
Design & Print by GENESIS | 9810033682
Disclaimer
The material in this publication is copyrighted. No part of this report can be reproduced either on paper or
electronic media without permission in writing from CSDS. Request for permission to reproduce any part of
the report may be sent to CSDS.
Usage of information
Forwarding/copy/using in publications without approval from Common Cause and CSDS will be considered
as infringement of intellectual property rights.
Cover Photo:
In the wake of school students dying in road accident, policemen help school children cross the road. (Credits: Praful
Gangurde, Hindustan Times, 19 January 2010, Mumbai)
Team
Advisory Committee
Sandeep Shastri
Sanjay Kumar
Suhas Palshikar
Vipul Mudgal
Analysis and report writing Research Support

Ananya Singh Ankita Barthwal Dhananjay Kumar Singh
Asmita Aasaavari Himanshu Bhattacharya
Arushi Gupta Jyoti Mishra
Radhika Jha Vibha Attri
Shakeb Ayaz
Shreyas Sardesai
Vijay Prakash
State Coordinators State Supervisors

A. K. Verma Anita Agarwal
Annapurna Nautiyal Kamal Srivastava
Anupama Saxena Nidhi Seth
Bhanu Parmar Nirmal Singh
Biswajit Mohanty Nurul Hasan
Dhruba Pratim Sharma Reetika Syal
E. Venkatesu Shamshad Ansari
Gyanaranjan Swain Umesh Bujji
Harishwar Dayal Veena Devi
Jagrup Singh Sekhon
K.M. Sajad Ibrahim
Kushal Pal
M. Asmer Beg
Mahashweta Jani
Nitin Birmal
P. Ramajayam
Rakesh Ranjan
Sanjay Lodha
Sudhir Kumar
Suprio Basu
Yatindra Singh Sisodia
Contents
List of Tables 1
List of Figures 3
List of Abbreviations 6
Surveyed States 8
Acknowledgement 9
Introduction 10
Chapter 1: Let the Numbers Speak: Police Performance Review 14
Chapter 2: Experience with the Police 33
Chapter 3: Trust in Police 54
Chapter 4: People’s Perception of Discrimination by the Police 70
Chapter 5: Fear of Police and Attitudes towards its Excesses 90
Chapter 6: People’s Perception on Different Aspects of Policing 105
Chapter 7: Analysing CAG Audit Reports 124
Chapter 8: Conclusion 133
Appendix 1: Technical details of study design and sample 141
Appendix 2: Questionnaire 144
Appendix 3: Details of how the indices were constructed 154
Appendix 4: Details of state rankings 160
Appendix 5: State wise findings from the survey 171
Appendix 6a: Police performance review using official data 185
Appendix 6b: Police performance review using objective data 192
Appendix 7: States’ compliance with Supreme Court directives 201
Appendix 8: CAG Report on Modernisation of Police 205

List of Tables
Table no. Title of the tables Page no.

Table 1.1 Crime rate index 18
Table 1.2 Disposal of cases by police and courts index 22
Table 1.3 Police diversity Index 23
Table 1.4 Police infrastructure Index 24
Table 1.5 Prison data index 26
Table 1.6 Disposal of cases of crimes against SCs, STs, women and children Index 28
Table 1.7 Comparison of state performance 30
Table 2.1 Mode of police contact by caste 37
Table 2.2 Form of FIR registration by locality and gender 40
Table 2.3 Receipt of complaint/ FIR across localities and gender 41
Table 2.4 Poor Muslims and OBCs are most likely to have paid bribe on contacting the police 42
Table 2.5 State-wise ranking of responses about satisfaction with police help after having contacted it 43
Table 2.6 State-wise ranking of responses on incidence of crime 45
Table 2.7 State-wise perception on police’s investigation of crime 48
Table 2.8 Sense of safety among people at different times of the day 50
Table 2.9 Upper castes most vocal about greater police presence 51
Table 3.1 Trust in local police by state 56
Table 3.2 Trust in senior police by state 57
Table 3.3 Respondents in West Bengal, Assam and Jharkhand most willing to send their child to the 58
police station if need arises
Table 3.4 Ranking of states in terms of trust in police 62
Table 3.5 Satisfaction with police performance and its impact on levels of trust in it 63
Table 3.6 Trust in police and its impact on satisfaction with its performance 63
Table 3.7 Ranking of states in terms of satisfaction with police performance 64
Table 3.8 Distrust levels highest for those who believe police intentionally implicates 65
Table 3.9 Perception of corruption and level of trust for local police 65
Table 3.10 Perception of corruption and level of trust for senior police officer 65
Table 3.11 Perception of corruption and level of trust for traffic police 65
Table 3.12 Satisfaction with police helps aid trust level 66
Table 3.13 Satisfaction with the help provided by police is positively related to willingness to approach 67
it again
Table 3.14 Dissatisfaction in Muslims most in states with ‘very bad’ Muslim representation 68
Table 3.15 Poor representation of women in police leads to dissatisfaction with police among women 68
Table 4.1 State-wise opinion on caste-based discrimination by the police 74
Table 4.2 Perception of castes and communities on police impartiality in the event of a caste strife 75
Table 4.3 State-wise opinion on religious discrimination by the police 77
Table 4.4 State-wise opinion on gender discrimination by the police 80
Status of Policing in India Report 2018 | 1

Table 4.5 State-wise opinion on class-based discrimination 82
Table 4.6 State-wise opinion on state-based discrimination 84
Table 4.7 State-wise ranking based on overall perception of discrimination 86
Table 4.8 Opinion on false implication of marginalized communities by the police 86
Table 4.9 State-wise opinion on false implication of SCs in petty crimes 87
Table 4.10 State-wise opinion on false implication of tribals on Maoist charges 87
Table 4.11 State-wise opinion on false implication of Muslims in terrorism related cases 88
Table 5.1 People most fearful of being beaten up by the police 91
Table 5.2 State-wise responses about fear of police: Punjab is the most afraid 92
Table 5.3 Muslims most fearful where their numbers are low 94
Table 5.4 Fear highest when one believes police is highly discriminatory 96
Table 5.5 Relation between awareness and fear of police: fear increases with awareness 97
Table 5.6 SCs very fearful of police if 97
Table 5.7 Muslims who believe police falsely implicates their community in terror related cases are 97
most fearful of it
Table 5.8 STs who believe police falsely implicates them are three times more likely to be highly 98
fearful of it than those who don’t
Table 5.9 Highly fearful respondents most likely to have a negative perception of the police 98
Table 5.10 Fear increases the level of partiality people perceive in the police 98
Table 5.11 People afraid of the police are less likely to approach them 99
Table 5.12 Fear leads to a demand for lesser police presence 99
Table 5.13 State-wise arrangement of scores showing attitudes towards police violence 100
Table 5.14 Sympathy for police likely to influence attitudes towards police brutality 102
Table 5.15 People in Jharkhand are most aware about quite a few cases involving police violence 103
Table 6.1 Trust levels lowest for traffic police 106
Table 6.2 Senior police thought to be least corrupt 107
Table 6.3 Scheduled tribes most likely to hold a negative perception of the police 110
Table 6.4 Haryana and Himachal Pradesh have the most positive perception of police 111
Table 6.5 People with a negative perception of the police least likely to seek police help 111
Table 6.6 Extent of interference in the police’s functioning in one’s area 112
Table 6.7 State-rankings on the question of police independence 113
Table 6.8 Negative perception of police greatest among those who don’t see police as independent 113
Table 6.9 Sympathy for the police increases with improvement in the literacy levels of respondents 116
Table 6.10 Ranking of states on sympathy for police’s working conditions 117
Table 6.11 Perception depends on sympathy levels for police 118
Table 6.12 Responses to questions on women working in the police force 120
Table 6.13 Gender based responses on women working in the police force 120
Table 6.14 Overall favorability towards women in the police force 121
Table 6.15 Young women more favorable towards having women in police than older women 121
Table 6.16 State-wise favorability towards women in the police force 122
Table 6.17 Fear of social stigma and harassment biggest reason why women don’t approach the police 123
2 | Status of Policing in India Report 2018

List of Figures
Figure no. Title of the figures Page no.

Figure 1.1 Rate of total cognizable crime and rate of violent crime indices 20
Figure 1.2 Disposal of cases: Police and Court Indices 21
Figure 1.3 Percentage of Muslim prisoners vis-a-vis percentage Muslim population in the states 27
Figure 1.4 Disposal Indices 29
Figure 2.1 Over four in five respondents did not have any contact the police in recent past 34
Figure 2.2 Men are more likely to have contacted the police 35
Figure 2.3 The well-to-do reported greatest police contact 35
Figure 2.4 Those who are most educated report highest police contact 36
Figure 2.5 Muslims reported highest police contact 36
Figure 2.6 Mode of police contact 37
Figure 2.7 The poorest are nearly twice as likely to have been contacted by the police as the rich 37
Figure 2.8 Reasons for police contact 38
Figure 2.9 A large plurality of people continue to visit the police station for their work 38
Figure 2.10 About two-fifth people sought a family member’s help in contacting the police 39
Figure 2.11 Registration of complaints/ FIR has a direct impact on people’s satisfaction levels 39
Figure 2.12 Reasons for non-registration of FIR 40
Figure 2.13 Copy of FIR/ Complaint 41
Figure 2.14 Profile of those who paid bribe 41
Figure 2.15 A large proportion of persons were moderately satisfied with police’s help 42
Figure 2.16 Satisfaction with the police help across localities 42
Figure 2.17 Reasons for dissatisfaction 43
Figure 2.18 Almost three-fourth respondents said that they will contact the police if the need arises 44
Figure 2.19 Satisfaction with the police likely to encourage police contact if needed 44
Figure 2.20 Incidence of crime by locality 45
Figure 2.21 Over one-third believe that crime in their locality has reduced 46
Figure 2.22 Change in the incidence of crime across localities 46
Figure 2.23 Opinion on police’s investigation by locality 47
Figure 2.24 Respondents’ overall satisfaction with police 48
Figure 2.25 Over one-fifth respondents have witnessed a police officer violating the law 49
Figure 2.26 Urban dwellers are more likely to have witnessed a police officer violating the law 49
Figure 2.27 Feeling safe has bearing on levels of satisfaction with the police 50
Figure 2.28 A large plurality of citizens want greater police presence in their locality 51
Figure 2.29 Respondents who used the 100-number service 52
Figure 2.30 Improvement in the access of ‘100’ number 52
Figure 2.31 7 out of ten women did not call on the helpline 52

Figure 2.32 Awareness on police related services 53
Figure 3.1 Trust in police vis-à-vis other institutions 55
Figure 3.2 Majority unlikely to let their wards visit police station alone 58
Figure 3.3 Overall trust levels in the police based on the trust Index 59
Figure 3.4 Distrust is inversely proportional to class hierarchy 59
Figure 3.5 Hindu STs distrust the police more than any other caste groups 60
Figure 3.6 Lower literacy levels correspond to high levels of distrust 60
Figure 3.7 Semi-skilled and agricultural workers more distrustful of the police 61
Figure 3.8 Women more distrustful of police 62
Figure 3.9 Negative perception of police indicates high levels of distrust 65
Figure 3.10 Paying a bribe and trust levels 66
Figure 3.11 Those having paid a bribe to the police, relatively more hesitant to approach it in future 67
Figure 4.1 Over a quarter think police discriminates on the basis of caste 72
Figure 4.2 OBCs among Hindus most likely to believe that police engages in caste discrimination 72
Muslims see the police as discriminating on caste basis more than other religious
Figure 4.3 72
communities
Muslim OBCs more vocal than Muslim upper castes in their perception of discrimination
Figure 4.4 73
on caste basis
Economically well-off among communities more likely to perceive caste-based
Figure 4.5 73
discrimination
Figure 4.6 People’s opinion on police impartiality in the event of a caste strife 75
Figure 4.7 Three in every five deny occurrence of discrimination by police on religious grounds 76
Figure 4.8 Muslims most likely to see police as discriminating on grounds of religion 76
Two in every three view police as being impartial when faced with an inter-religious
Figure 4.9 77
conflict
Figure 4.10 Perception of religious communities on police impartiality in the event of a religious strife 78
Figure 4.11 Perception on gender-based discrimination by the police 78
Urban men and women more likely to see gender bias among police than rural men and
Figure 4.12 79
women
Figure 4.13 Highly educated women and men more likely to express gender discrimination by police 79
Figure 4.14 Opinion on class- based discrimination by the police 80
Figure 4.15 Both rich and poor equally likely to believe that police discriminates on class lines 81
Figure 4.16 Urban poor more likely to view police as being discriminatory than rural poor 81
Figure 4.17 Opinion on discrimination by the police against people from another state 83
People in cities more likely to believe police discriminates against people from another
Figure 4.18 83
state
Non-migrants in big cities more likely to see police as discriminating against people from
Figure 4.19 84
another state than recent migrants
Figure 4.20 Index on perception of discrimination 85
Figure 5.1 Index of fear of police 92

Figure 5.2 Religion-wise distribution of police fear: Sikh fear higher than other communities 93
Figure 5.3 Muslims in the South are more likely to fear the police 93
Figure 5.4 Fear within the Hindu community: Upper Castes least fearful of the police 95
Figure 5.5 Region-wise distribution: Villages most fearful of the police 95
Figure 5.6 Experience has a significant effect on fear- bad experience leads to more fear. 96
Figure 5.7 Half the respondents condone the use of violence on criminals in police custody 99
Figure 5.8 STs least likely to agree with police violence 101
Figure 5.9 People in rural areas less likely to accept police violence towards criminals 101
Few people have reported awareness of cases of police violence: Maximum know victims
Figure 5.10 102
of police torture
Figure 5.11 Awareness of police excesses: Majority report being unaware of police violence 102
Figure 6.1 Nearly 1 in 6 feel police implicates under false charges 107
Figure 6.2 54% feel police is blamed unnecessarily despite doing its job well 107
Figure 6.3 55% feel that the police is corrupt and does not do its job well 108
Figure 6.4 Index of perception about the police-majority display a positive perception 108
Figure 6.5 Female perception of police: not very different from male perception 109
Figure 6.6 Non-literates most likely to have a negative perception of the police 109
Figure 6.7 Poorer individuals more likely to have a negative perception of the police 109
Figure 6.8 Sikhs most likely to hold a negative perception of the police 110
Figure 6.9 Nearly half the respondents likely to perceive police as independent 112
Figure 6.10 Urban dwellers more likely to acknowledge that working hours of the police are greater 114
Figure 6.11 60% believe police personnel work under high stress level 114
Figure 6.12 Half the respondents believe that police is lazy and unmotivated to serve the people 115
Figure 6.13 4 out of 5 believe that police work under tough conditions 115
Figure 6.14 Urban respondents more likely to be fully sympathetic than rural 116
Figure 6.15 OBC respondents most likely to have a fully sympathetic attitude towards police 116
Figure 6.16 Christians most likely to hold fully sympathetic attitude towards police 117
Policewomen thought to be more honest, but respondents more likely to approach
Figure 6.17 118
policemen
Across demographic variables, preference for police officers of both the genders is visibly
Figure 6.18 119
high
Figure 6.19 Women more likely to believe that policewomen are more honest 119
Figure 6.20 Women more likely to think of policewomen as more hard-working 119

List of Abbreviations
AAP: Annual Action Plan GDP: Gross Domestic Product
AIDS: Acquired Immunodeficiency Syndrome GoI: Government of India
ACCIL: Assam Cyber Crime Investigation GPI: Global Peace Index

Laboratory GSPHCL: Gujarat State Police Housing
AFSPA: Armed Forces (Special Powers) Act Corporation Limited
AP: Andhra Pradesh HC: Head Constable
ASI: Assistant Sub-Inspector HP: Himachal Pradesh
BPRD: Bureau of Police Research & IGP: Inspector General of Police

Development IPS: Indian Police Service
CAC: Crimes against children IPC: Indian Penal Code
CAG: Comptroller and Auditor General of INSAS: Indian Small Arms System
India
ITPA: Immoral Trafficking (Prevention) Act
CAW: Crimes against women
JJA: Juvenile Justice (Care and Protection
CII: Crime in India of Children) Act
CASCs: Crimes against Scheduled Castes LoP: Leader of Opposition
CASTs: Crimes against Scheduled Tribes MFSU: Mobile Forensic Science Unit
CAG: Comptroller & Auditor General of MHA: Ministry of Home Affairs
India
MO: Medical Officer
CCTV: Close Circuit Television
MoS: Minister of State
CID: Criminal Investigation Department
MP: Madhya Pradesh
CCTNS: Crime and Criminal Tracking
Networks and Systems MPF: Modernisation of Police Force
CTS: Constable Training School MPM: Modern Prison Manual
CFSL: Central Forensic Science Laboratories MPV: Medium Police Vehicles
CLAA: Criminal Law Amendment Act NCRB: National Crime Records Bureau
CPO: Central Police Organisations NFHS: National Family Health Survey
CrPC: Criminal Procedure Code NPC: National Police Commission
CTS: Constable Training School NSC: National Security Commission
DFSS: Directorate of Forensic Sciences OBCs: Other Backward Castes

Services PAC: Provincial Armed Constabulary
DGP: Director General of Police PCA: Police Complaints Authority
DMFU: District Mobile Forensic Unit PCR Act: Protection of Civil Rights Act
DPO: Data on Police Organisations PCMA: Prohibition of Child Marriage Act
DySP: Deputy Superintendent of Police PEB: Police Establishment Board
FSL: Forensic Science Laboratory PoC Act: Prevention of Corruption Act

POCSO: Protection of Children from Sexual SLEC: State Level Empowered Committee
Offences Act
SP: Superintendent of Police
PTC: Police Training College
SSC: State Security Commission
PS: Police Stations
STs: Scheduled Tribes
PSI: Prison Statistics India
SVSPA: Swami Vivekananda State Police
PWDVA: Protection of Women from Domestic Academy
Violence Act
TN: Tamil Nadu
RFSL: Regional Forensic Science Laboratories
TB: Tuberculosis
SC: Supreme Court
UNODC: United Nations Office on Drugs and
SCs: Scheduled Castes Crimes
SC/ST Scheduled Castes and Scheduled Tribes UP: Uttar Pradesh
PoA Act: (Prevention of Atrocities) Act
UT: Union Territory
SI: Sub-Inspector
UTPs: Undertrial Prisoners
SFSL: State Forensic Science Laboratory
UTRC: Under-Trial Review Committees
SHO: Station House Officer
WB: West Bengal
SLL: Special and Local Laws
WP(C): Writ Petition (Civil)
SLR: Self Loading Rifles

Surveyed States

Acknowlegement
Status of Policing in India Report 2018: A Study of Performance and Perceptions was conceived as
a collaborative effort. It is the teamwork of Common Cause and the Lokniti- Programme
for Comparative Democracy at the Centre for the Study of Developing Societies (CSDS)
along with their partners in many universities and research institutes across India. Even
before this report was conceptualised in its present form, successive Common Cause
Directors and board members have been brainstorming with subject experts and like-
minded organisations about the need for collating fact-based knowledge and time-series
data on policing in India. Mr. Kamal Kant Jaswal, Mr. Prakash Singh, Mr .Vikram Lal and
Dr. B P Mathur deserve a very special mention.
The study would not have been possible without the commitment of our philanthropic
partners, the Tata Trusts and the Lal Family Foundation. We are grateful to the National
Crime Records Bureau (NCRB), the Bureau of Police Research and Development (BPRD),
the Indian Police Foundation (IPF) and the Comptroller and Auditor General (CAG) of
India whose data and insights have been used in the study. A very special gratitude is owed
to Mr. Roshan Kishore, Senior Editor and data journalist; Mr. Shambhu Ghatak, Inclusive
Media for Change and Himanshu Bhattacharya, chief statistician, Lokniti-CSDS for their
guidance and support in statistical analysis of diverse data sets. Our special thanks to Dr.
Arvind Varma, Associate Professor, Indiana University who took the time to review the
questionnaire and offered constructive suggestions. The study gained a lot from the ideas
and insights of those who attended the brainstorming sessions in the run up to the survey,
particularly Professors Rajeev Bhargava, Abhay Dubey, Ujjwal Kumar, Awadhendra
Sharan, Hilal Ahmed, Sanjeer Alam; Maj. Gen. (Rtd) Anil Verma, Mr. Ajay K. Mehra,
Ms. Anuradha Bhattacharjee, Ms Devika Prasad and Ms. Devyani Srivastava. We would
also like to thank Ms. Maja Daruwala and Mr. Sanjoy Hazarika from the Commonwealth
Human Rights Initiative (CHRI), Mr. Harish Narasappa from Dakhsa, Mr. Jagdeep
Chhokar from the Association for Democratic Reforms (ADR) and Mr. Nikhil Dey from
the National Campaign for Peoples’ Right to Information (NCPRI) who contributed ideas
and encouraged us to take up the project. The study benefited immensely from the kind
support of a large number of respondents who were part of our survey and spared their
valuable time, and the field investigators whose dedication and efforts in data collection in
rural and urban locations contributed to the successful completion of the survey.
Lastly, our special thanks are extended to the interns at Common Cause and CSDS, Ms.
Tanya Krishnoo, Mr. Rajat Ahlawat, Ms. Aradhana Sanil, Harshit Pande, Amogh Sharma
and Ms. Antara Banerjee. Equally important is the contribution of the members of staff at
Common Cause and CSDS who have contributed in many areas between data collection
and logistics.

INTRODUCTION
Policemen retaliated with batons after a protest by lawyers turned violent.

(Credits: Prabhakar Sharma, Hindustan Times, 7 March 2013, Jaipur

Status of Policing in India Report: The Context
It is a matter of pride for Common Cause to launch Indians. Police establishments across India continue
the first Status of Policing in India Report 2018: A Study to function largely in the same casual and repressive
of Performance and Perceptions. True to its mission fashion as the colonial masters had envisioned.
statement, ‘A Romance with Public Causes’, They follow a rigid, unilinear hierarchy, with all the
Common Cause works on the rule of law, probity trappings of pomp and power, often behaving as the
in public life and accountability in governance since private armies of the new masters. This is hardly
its inception in 1980. Working for people-centric suited to the changing needs of the world’s largest
policing has always been an integral part of this democracy.
mission.
We have to accept that India’s performance in
The report is a collaboration between Common this area has been dismal and is worsening on
Cause and Lokniti Programme of the Centre for the many parameters. The Rule of Law Index under
Study of Developing Societies (CSDS), a research the World Justice Project ranks India at the 62nd
institution which works in partnership with a wide position out of 113 countries. In terms of criminal
network of researchers and academic institutions justice, it stands at the 66th position, but in the civil
all over India. The report draws on earlier efforts justice component it slides to the 97th position. In
by civil society, academia, think tanks and research the Corruption Index 2017 of the Transparency
institutions. The work has been supported by grants International, India is ranked at the 81st position
from Tata Trusts and Lal Family Foundation. out of 180 countries. The 2017 progress report on
Sustainable Development Goals (SDGs) has found
For Common Cause and its patrons, the report is
that globally, the proportion of people held in
not a one-off project, but a long-term commitment.
detention without being sentenced for a crime has
We see it as a firm step towards generating time-
remained almost unchanged — from 32 percent of
series data on the performance of the police and the
total prisoners in 2003-2005 to 31 percent in 2013-
levels of citizens’ trust and satisfaction in their day
2015. The percentage of undertrials in India is more
to day working. We hope that the data presented
than double the global standards, at 67.2 percent in
here will illuminate the nature of the relationship
2015.
between the police and the communities they are
meant to serve. We believe that a long-term study Changing roles and outlooks
will help, apart from common citizens, all those
policemen, scholars and activists who work on the Every country has its unique set of threats and
ground to humanise the law enforcing machinery opportunities, even though there is always scope for
and build trust by promoting people-centric policing learning from others. It is important to note that the
and community participation. traditional policing has undergone a sea change in the
more successful democracies of the world. Feedback
For the common person, the police are the most
mechanisms like citizen’s satisfaction surveys have
visible face of the state. A sovereign government is
resulted in better understanding of crimes and higher
called a ‘failed state’ if it is unable to control law-and-
levels of community policing. This transformation
order, but when it uses repression as the instrument
has also been shaped by the legitimate demands of
of control, it is condemned as a ‘police state.’ There
the citizens and their participation in governance.
is a ‘legitimacy deficit’ in both situations. Good
In India, we could have also done that, in our own
governance demands a balance between fair and
unique way, by taking full advantage of the landmark
effective enactment of the rule of law. A society
73rd and 74th Constitutional Amendments which
cannot become just or democratic if the police
have empowered the elected Village Panchayats
are not responsive to the needs of the community,
and Urban Local Bodies to take decisions at the
particularly its weaker and vulnerable sections. So,
grassroots levels. A systemic cooperation between
the obligation of the police is not only to control
the panchayats or urban wards and the police
crime but to do so in an unbiased way while treating
stations – which happen to be the citizens’ first point
people with dignity and respect.
of contact with law – will provide much-needed
Law and order is a state subject under the monitoring by the community. An important part
Constitution of India. It is a pity that while the of law enforcement is to identify gross misconduct
Constitution empowers the states to enact their and to take action against erring officials which
own laws, most of them still retain the essential can only be done with inbuilt mechanisms and the
characteristics of the colonial Indian Police Act, involvement of the community.
1861, which was framed to subjugate the ‘native’

The New York-based Vera Institute of Justice, experts, academics and serving and retired police
which has been working on fair policing since 1961, officers. The data for the objective analysis has been
sees community policing as democracy in action. It accessed from the police establishments like the
requires active participation of local governments, National Crime Records Bureau (NCRB), Bureau
civic and business leaders, public and private of Police Research and Development (BPRD) and
agencies, residents, religious organisations such other institutions like the Comptroller and Auditor
as temples, churches, mosques, and educational General (CAG) of India.
institutions, among others. The institute advocates
the use of citizen surveys to make sense of what is Harmony, peace and public order
and what ought to be. One of the most fundamental
It is obvious that multi-dimensional efforts are
issues for the police is to be able to protect citizens
required to promote peaceful societies and provide
without giving up its own accountability. After all,
access to justice for all as stated in SDG 16 of the
the police cannot expect people to be law-abiding
United Nations. After all, the purpose of law is to
when they themselves disobey law. It is a rational
create social harmony and cohesion in society rather
expectation, therefore, that citizens’ compliance of
than to mechanically apply the rules. And that is
law will increase if the policing apparatus is seen as
why the police have to be reoriented to keeping
fair, law-abiding and even-handed.
peace and maintaining public order against trying
The citizens’ satisfaction surveys are also in the circumstances, rather than just fighting crime. A
nature of citizens’ feedbacks on routine policing beginning has to be made by first grasping the
matters and have been carried out in the US, UK, issues of an antiquated policing system, and its
New Zealand, Canada and Australia, among many old-fashioned interface with public, before trying to
other countries. They capture citizens’ perception reform it. From that perspective also, we believe that
of police performance and competence; citizens’ this report would be a valuable tool for the citizens
perception of their attitudes and behaviour; to monitor the impact of policing on the ground.
community concerns over safety and security and A comparison between the states is also vital to be
their recommendations for service improvement. It able to examine what works and what is counter-
is a fairly common practice for the European Union productive from the point of view of governance
(EU) to commission surveys which compare public and public policies. The states can always take
trust in the police and the justice system across proactive steps to change the status quo; they can
member countries in order to observe cross-national even repeal archaic laws and enact progressive ones,
variations. Their studies have found that rather given the political will or public pressure. And that is
than from mechanisms of deterrence, compliance why the individual state has been treated as the unit
with law comes more easily from legitimacy of of performance in the report.
the authority and public’s trust and confidence in
The analysis of performance and perception about
the system (Hough, 2012). Many such surveys are
policing in 22 states in this report is arranged primarily
increasingly being done online. However, in India
in terms of best or worst-performing states. The
online surveys are problematic for many reasons,
information is also given according to age, gender,
particularly because they run the risk of excluding
caste, community, urban/ rural or economic/
disadvantaged groups who happen to be on the
educational status, among other parameters. The
wrong side of both, law enforcement and the digital
performance indicators have been developed on the
divide.
basis of official statistics for five years until 2016.
In today’s data-driven world, it is tough to bring Forty-two variables have been categorised into six
about an organisational transformation or to achieve main themes (i.e. crime rate, disposal of cases by
a perceptible improvement in the behaviour of the police and courts, diversity in the police force, police
police, or in their service delivery mechanisms, infrastructure, prison data and disposal of cases of
without identifying the existing need gaps. This crimes against SCs/STs/ women and children.)
report seeks to do precisely that by presenting a The survey provides snapshots of police-citizen
combination of fact-based markers of the capacities relations, levels of fairness and responsiveness of the
and competence of the forces, derived from the criminal justice system to distress and crime, and the
official data, followed by a comprehensive survey levels of accessibility and impartiality with respect
of peoples’ perception of the police on the ground. to the society’s vulnerable sections. The report also
The objective of this report is to provide accurate puts together critical deficiencies flagged by the
diagnostics for better understanding and meaningful Comptroller and Auditor General of India (CAG)
reforms. The surveys have been conducted face to in its audit reports of 11 states spread over a decade.
face by surveyors especially trained in handling
For those who would like to locate the performance
the questionnaire prepared after field-based pilots
of a particular state or study the specific details of
and several brainstorming exercises involving
sub-themes, the study contains detailed annexures.

These are tabular representations of select objective sections. And that is why this study attempts to
data, state-wise compliances to Supreme Court tap into the nuances of the public perception based
guidelines and technical details of survey indices. on the experiences of all citizens, irrespective of
For reasons of logistics and resources, we were not caste, class and gender, though more particularly of
able to include seven smaller states which constitute traditionally disadvantaged sections.
around one per cent of the country’s population.
The questions and indicators in the study have been
The North-East of India is represented by Assam
kept simple and comparable while the methodology
and Nagaland while the latter happens to be one
is transparent yet nuanced and rigorous. For policy
of best performing states on many parameters.
makers, media persons, scholars and activists,
Except Jammu and Kashmir, all the other states
the data presented here will hopefully provide
to be excluded have a population between 6 lakhs
important insights into policing in India. It will also
and 36 lakhs (Arunachal Pradesh, Goa, Manipur,
answer some old questions, raise some new ones,
Meghalaya, Mizoram, Sikkim and Tripura). J&K
and work as a building block for more research. The
has not been included because routine policing
indicators tell lay readers something concrete about
operations in the state have been limited by the
problem areas in the rule of law in India and the
ongoing conflict and the presence of the Army
direction in which we are headed. The study offers
and paramilitary forces in large areas of the state.
a good opportunity to the leaders of states which
Each chapter has its own methodology while more
perform poorly, or where the peoples’ trust in law
specific details like the questionnaire and the coding
enforcement is really low, or where the fear of the
sheet have been given in the end.
police is unusually high, to use the data to introspect
Legitimacy despite cynicism or to take lessons from their better performing
neighbours. We hope that those in power will use the
Contradictory as it may sound, the average Indian empirical evidence to take decisive steps to improve
reports abuse of authority and corruption in the things in their jurisdictions or areas of influence and
police in the same breath as an overall satisfaction those in the opposition would demand meaningful
with their day-to-day functioning. This is broadly reforms. Common Cause and CSDS will be happy
consistent with a global trend about peoples’ to receive any feedback on the report.
perception of their police. This is so, perhaps, because
Vipul Mudgal
while being estranged with law for whatever reason,
Director,
an individual can also simultaneously believe in the
Common Cause
legitimate authority of law. This, however, does not
diminish the individual’s fear from the police. The
survey tries to resolve this enigma by analysing the
citizens’ fear of the police according to the state they
live in or the caste, community or religious groups
they belong to. A significant part of the survey has References
also been devoted to examining difference between Hough, Mike. “Researching Trust in Police and Trust in
the rich and the poor as well as between rural and Justice: A UK Perspective.” Policing and Society 22, no.3
urban folks. On many parameters, the results of (2012): 332-345. http://eprints.bbk.ac.uk/5039/1/5039.
the study may not be shocking or dramatic but they pdf (accessed March 29, 2018)
provide definite clues to the attitudes and perceptions Transparency International. Corruption Perceptions
of the multiple publics vis-à-vis the police as well as Index 2017. Annual Report, 2017 edition. https://www.
one another. transparency.org/news/feature/corruption_perceptions_
index_2017#table (accessed March 1, 2018)
The difference in trust and satisfaction levels
becomes starker, for instance, when one talks to the United Nations. Goal 16: Sustainable Development
society’s disadvantaged sections who are more likely Knowledge Platform. https://sustainabledevelopment.
to be victims of exclusion and procedural injustice. un.org/sdg16 (accessed March 12, 2018)
It is important, therefore, to examine the issues of World Justice Project. Rule of Law Index 2017-
public cooperation and compliance with law as 2018. Annual Report, 2018 edition. http://data.
well as those of police excesses and atrocities from worldjusticeproject.org/#/groups/IND (accessed
the point of view of the poor and the vulnerable February 12, 2018)

CHAPTER 1:
LET THE NUMBERS SPEAK: POLICE
PERFORMANCE REVIEW
Police attempting to control protesters following the murder of a child in a school in Gurgaon
(Credits: Parveen Kumar, Hindustan Times, 10 September 2017, Gurgaon)

Let The Numbers Speak: Police Performance Review
The report that you are reading through the specific guidelines for the implementation of police
following pages contrasts the findings of a reforms. Its implementation in the past 12 years is a
nationwide survey of 22 states with the insights story of a monumental non-compliance.
derived from objective data accessed and collated
Prakash Singh vs Union of India, 2006 was a landmark
from the police establishments. The idea is to work
judgement that sought to make it incumbent upon
towards restoring the citizens’ trust and confidence
the government to make much-needed changes in
in the police force by studying the gaps in the levels
policy which were long overdue. The judgement
of their performances and the peoples’ expectations
was a huge victory for many and the specific nature
from them.
of the directions given by the Court made it seem
It is tough to achieve an appreciable improvement that there was little or no scope for non-compliance.
in the existing practices of policing without A progress report on the action taken by the police,
understanding where we are going right or wrong. or the lack of it, is given in Appendix 7. The seven
This report analyses official law enforcement data directives mandate the union and state governments
in order to offer some evidence about efficiencies to set up the following institutions or take specific
on the ground. Once we are able to rate state-wise actions:
performances, it is easier to measure them against
1. State Security Commissions (SSC) with the
peoples’ perceptions and expectations presented in
Leader of Opposition, judges and independent
the next few chapters.
members to ensure that the state police is
This chapter has been divided into two sections. In the able to function independent of unwarranted
first section, we are discussing the broad framework government control, influence or pressure.
of the study regarding different states’ compliance
2. The Director General of Police (DGP) to be
to the Supreme Court directives of 2006 followed by
selected from amongst the three senior-most
an analysis of objective data collected from official
officers and to have a minimum tenure of 2
sources and arranged and presented thematically
years.
in six groups. The Supreme Court directives in the
Prakash Singh case, in which Common Cause was 3. Minimum tenure of I.G. of Police and other
a co-petitioner, are central to the issue of police officers on operational duties should also have a
reforms in India and a study of the compliance with prescribed minimum tenure of two years.
the directives has been presented.
4. Separate wing for investigation of cases
The following section, however, based on the rating
5. Police Establishment Board (PEB) for all
of the states in the form of indices, goes much
transfers, postings, promotions and service
beyond that. For the analysis of the objective data,
matters of officers up to the Dy Superintendent
we have come up with an easy-to-understand Index
of Police rank, and to hear their appeals.
to measure the performance of Indian states across
selected parameters. Our endeavour is to let the 6. Police Complaints Authority: Both at the
reader measure the results of the state-wise analysis state and district levels to hear complaints
of the official data against the findings of the survey against police officers up to the rank of Dy
which will follow in the next few chapters. Superintendent of Police.
1.1 A momentous non-compliance 7. National Security Commission (NSC) for

selection and placement of Chiefs of the Central
India has a long history of making big promises on Police Organisations (CPOs) and to review the
police reforms without effective delivery. Common effectiveness of the police forces.
Cause Journal (July-September, 2015) lists all These directives are a culmination of the main
important committees and commissions formed recommendations of the different committees on
since the colonial period (Ayaz, 2015). However, police reforms. As pointed out by Joshi (2013),
India’s watershed moment came on September 22, some critics have called these “National Police
2006, in Prakash Singh vs Union of India, when after Commission model of reforms” i.e., with focus
over a decade of hearing the petition, and after the only around the reduction of political influence on
failure of states to implement recommendations policing, instead of addressing structural problems.
of a number of expert panels, the Supreme Court However, there is no denying the fact that these
delivered a momentous judgement that contained

directives could be the first set of building blocks for made wherein the recommendation may be rejected
more reforms in future. in exceptional cases by the state government for
reasons given in writing. These, and many other
1.1.1 The role of the states examples, show that the Supreme Court directives
Under the Constitution, the law and order, are being observed in breach rather than in
including the police, is a state subject. The states are compliance.
empowered to enact their own laws but most of them
Overall, Himachal Pradesh is found to be the most
still retain features of the Indian Police Act, 1861,
compliant with the Supreme Court directives, with
which is not just archaic but also colonial. Seventy
Uttarakhand close behind. Kerala too has shown
years after independence, the police establishments
relatively better compliance with the SC directives
in India continue to function in the same repressive
and has made progress on the issue of police
way -- with a rigid hierarchy of control and an
autonomy. The Union Government too was to form
unilinear command system -- and largely without
a National Security Commission along the lines of
transparency and accountability required under
the SSC. This, however, was formed in 2017 more
norms of democratic governance.
than 10 years after the SC judgement, according
It is hardly an exaggeration to say that almost all to a January 2018 statement given by Minister of
states of India are guilty of the contempt of court State for Home Affairs, Hansraj Gangaram Ahir in
for not implementing the Apex Court’s directives Parliament.
fully and unequivocally. Their failure is at both levels
1.1.2 Beyond SC directives
-- in absolute non-compliance with the directives, as
well as in inserting dubious provisos in legislation One crucial problem with the police structure
which run counter to the spirit of the judgement. Mr. which has not found a sufficient safeguard in the
Prakash Singh, the main petitioner in the 2006 case, court directives is that of discrimination against
has filed a contempt petition in the Supreme Court the subordinate rank officers and undue abuse of
for effective implementation of the directives. This authority by the senior rank officers. This hierarchy
petition is still being heard and the case is nowhere is so deeply ingrained within the police structure
near resolution, 11 years after it was first filed. that the Police Act of 1861 continues to use the
terminology “inferior officers” under Section 7
The state governments are seriously deficient in
of the Act1. Cases of harassment by senior police
compliance to the directives and are, in effect, in
officers meted out to those in the subordinate ranks
flagrant violation of the Supreme Court’s order.
are common news. In 2015 alone, a total of 167
Appendix 7 gives a quick comparison among the 13
police personnel committed suicide in India.
states which have passed state Police Acts after the
judgment. For instance, while all of these 13 states Besides cases of harassment by seniors,
have a provision for a State Security Commission dissatisfaction with job and lack of professional
(SSC), its membership seems seriously compromised growth continue to ail the Indian Police. According
i.e., five states do not have provisions for the Leader to a study conducted by BPRD in 1990, only 22
of Opposition to be a member, and four states percent police constabulary could get promotion,
do not provide for non-political or independent and the remaining 78 percent were stagnating at the
member. Worse still, specific provisions to make same rank in which they joined the force (Joshi et
the recommendations of the SSC binding on the al., 1990). Such, and related issues that have an effect
government exist only in two states, Kerala and on the overall police functioning and efficiency, are
Himachal Pradesh. some issues that have not been adequately addressed.
The other two bodies directed to be formed by the Another crucial omission is Section 197 of the
court – Police Establishment Board (PEB) and the Criminal Procedure Code (CrPC), 1973, which
Police Complaints Authority (PCA) — have similarly makes provisions for a prior sanction by the
been weakened through legislative loopholes. Kerala government for the prosecution of judges and
and Haryana Police Acts have altogether done away public servants.2 Section 19 of the Prevention of
with the core function of deciding transfers and Corruption Act, 1988, also makes it mandatory
postings of officers, while Bihar Police Act does not to have a previous sanction by the government
have a provision for such a board, all in violation for court to take cognizance of an offence under
of the court’s order. The recommendations of the the Act. Thus, police officers in effect have been
PEB have been made binding only in Karnataka, protected through legal measures from prosecution
Kerala and Uttarakhand. Similarly, in the case of without government sanction.
Police Complaints Authority (PCA), specific legal
While the PCA has been set up as an institution to
provisions making its recommendations binding
deal with grievances and complaints against police
exist in only two states- Himachal Pradesh and
officers, but in cases where the recommendations of
Kerala, and in Maharashtra, a provision has been
the PCA are not binding, (which is the case with 10

out of the 13 states studied above), these forums will 2012 to 2016, for each specific year, as well as for
be rendered futile since further appeals in the courts the average of the five years. The formula used for
will be complicated. According to media reports creating the Index was:
(M. Raghava, 2012), sanctions for prosecution
State Index =
of police officers have not been provided by state
governments in a majority of cases, such as in the (x-minimum observed in the last five years)
case of Karnataka where the government refused
(maximum observed in the last five years-minimum
to give sanction for prosecuting seven officials
observed in the last five years)
against whom the Lokayukta had conducted raids where ‘x’ is the actual state figure for the variable
and found assets disproportionate to their known
sources of income. In the Subramaniam Swamy The indices for the different variables were averaged
case, the court gave guidelines to Parliament to out to arrive at a thematic Index. Thus, the final
introduce a time limit of three months in Section 19 thematic Index is an average of the individual
of the Prevention of Corruption Act within which variable indices under it. A detailed list of the state-
the decision regarding a sanction should be taken, wise figures on all variables is given in Appendix 6.
failing which the permission will be deemed to have The need to create these indices arose from the fact
been granted. However, the legal position in this that the data, in some cases, were not in comparable
matter is not settled, and Section 19 continues to formats and could not be fused together without
apply in cases of corruption against police officers. bringing in uniformity in the computation of the
values. Additionally, the indices, calculated using
1.2 Where do we stand? A state-wise the maximum and minimum values observed
analysis of objective data across all states in the previous five years, enables
a relative ranking for each variable. While it would
Data on state-wise police performance using
be presumptive to accord direct correlational
objective parameters was aggregated from publicly
associations between any of these variables at this
available, official statistics from the National Crime
level, a comparative ranking can go a long way in
Records Bureau (NCRB) and Bureau of Police
enhancing institutional performance by encouraging
Research and Development (BPRD). The three main
good practices and state-driven inputs in the police
reports relied on for gathering the datasets are Crime
machinery.
in India (NCRB), Data on Police Organizations
(BPRD) and Prison Statistics India (NCRB). During the initial analysis, the objective was to
develop a cumulative ranking of the different
This section looks at the performance of the police
indicators for an overall police performance
in states because law and order comes under the
evaluation. However, this process was amended
state list and it is at the state level at which crucial
in the later stages seeing as how the nature of the
decisions about policing are taken. The large base
thematic groups were entirely different, as were the
of data sets was reduced to a total of 43 variables
institutions from which this data was emanating:
collected over a span of five years from (2012 to
police, prisons, court and the State.
2016) for the purpose of developing comparable
and measurable state-wise parameters on the Another major consideration behind the decision
performance of the police. The figures have been to abstain from a cumulative ranking was that
averaged for five years so that the effect of state- increasing crime rates in a society is not necessarily
specific turbulences in a single year due to unrelated an outcome of poor policing. It could very well, in
factors may be minimized. All figures given below are fact, be the other way around – increasing crime rates
averages of five years, unless otherwise mentioned. may be indicative of improving registration of cases
These variables were categorised and divided into by the police – in itself a positive indicator of police
six different themes, which are as follows: performance. Yet, as is seen in the survey findings
in the subsequent chapters, decreasing crime rates
1. Crime rates
in a locality have a positive impact on people’s
2. Disposal of cases by police and courts levels of satisfaction with the police. Being a crucial
determinant for policing, it is not feasible to leave
3. Police diversity
out crime rates altogether when studying the police
4. Police Infrastructure structure of a country. In order to balance out these
seeming contradictions, the different thematic heads
5. Prison data
have all been addressed separately in this analysis.
6. Disposal of cases of crimes against SCs, STs,
Coming back to poor registration of cases by the
women and children
police and to the issue stemming from that -- how
For each of the variable under the above themes, an reliable are the figures projected by the police and
Index was created for the five-year period between the State? Even with all the shortcomings and

doubts raised about the data, these are the only 2. Rate of violent crimes7
comprehensive and verifiable data sets that have
3. Rate of total cognizable crimes against
State accountability. Doubts raised regarding the
women8
accuracy of this data, although not completely
unfounded3, need to be set aside for any analysis 4. Rate of total cognizable crimes against
aimed at policy recommendation and advocacy. Scheduled Castes9
In the following sub-sections, each of the above 5. Rate of total cognizable crimes against
indices will be discussed separately. The indices Scheduled Tribes10
have been computed such that it provides a range
6. Rate of total cognizable crimes against
between 0 and 1 for each indicator, with 0 depicting
children11
the poorest performance and 1 being the benchmark
for best performance. Higher figure for each Index Crime, from a criminological perspective, lends itself
is indicative of better performance4. This data only to various causes, ranging from societal turbulence
pertains to those collected from official sources and to failure of the law and order machinery. Modern
is independent of the survey results and findings. theories of crime can be traced back to atavistic
school of thought propounded by Lombroso (1876)
1.1.1 Crime rate Index5 which puts the burden of deviancy on inherent
The variables used for calculating the crime rate individual traits. On the other hand, sociological
Index are6: theories such as the social disorganisation theory or
the strain theory place the onus on the breakdown
1. Rate of total cognizable crimes under Indian of social norms in a society or disproportionate
Penal Code (IPC) and the Special and Local opportunities available to different communities.
Laws (SLL) Therefore, trying to understand crime from a
Table 1.1: Crime rate Index

State rank States ranked from least crime rates to most crime rates Crime Rate Index
1. Punjab 0.91
2. Himachal Pradesh 0.88
3. Jharkhand 0.84
4. Tamil Nadu 0.84
5. Uttarakhand 0.83
6. West Bengal 0.83
7. Gujarat 0.81
8. Maharashtra 0.81
9. Uttar Pradesh 0.79
10. Bihar 0.78
11. Karnataka 0.78
12. Haryana 0.77
13. All India 0.77
14. Andhra Pradesh 0.74
15. Assam 0.74
16. Andhra Pradesh + Telangana 0.74
17. Odisha 0.72
18. Chhattisgarh 0.71
19. Telangana 12
0.71
20. Madhya Pradesh 0.69
21. Rajasthan 0.63
22. Kerala 0.52
23. Delhi UT 0.51
Note: Index interpretation- 0 indicates worst performing and 1 indicates best performing
(Nagaland excluded because data not available/not computable)

sociological or legal perspective brings out various increase, from 8.9 to 24, between 2012 and 2016.
conflicting dilemmas, and to put the responsibility The introduction of new laws such as Protection of
of crime squarely on the police as an institution Children from Sexual Offences Act (POCSO), 2012
would be undue. may have created an enabling structure for increase
in registration of these crimes.
Another difficulty with making crime rate a marker
of police performance is the common practice of 1.1.2 Disposal of cases by police and courts
non-registration of crime (Rao & Tiwari, 2016), Index
stemming from the requirement of police to project
a better law and order situation in the state. In a The variables used for the calculation of disposal of
study done by UP Police Commission in 1970-71, cases by police and courts Index are:
it was unanimously admitted by the officers that 1. Chargesheeting rate of IPC and SLL cases by
concealment and minimization was commonly the police13
done by them. It has been noted, contrastingly,
that increase in crime rates in some cases may be 2. Disposal percentage of IPC and SLL cases by
a result of improving registration of crimes in that the police
state (Chandra, 2016). In order to get a better picture 3. Conviction rate of IPC and SLL cases by the
of the crime rate in a state, rather than looking at courts14
the rate of overall crimes (rate of total cognizable
4. Disposal percentage of IPC and SLL cases by
crimes in official terms), a look at the rate of violent
the courts
crime in the state would be more effectual. This is
based on the premise that violent crimes, which The disposal Index is an indicator of the prompt
include offenses such as murder, rape, robbery and action taken by the police and by the courts on the
kidnapping, are a lot more difficult to suppress in reported cases of crimes. India is notorious for a
documentation. huge backlog of cases in courts and severe delays
in justice. In the landmark judgement of Hussainara
Therefore, crimes such as murders are more reliable
Khatoon vs Home Secretary, State of Bihar the Supreme
markers of the law and order situation prevailing in
Court held speedy trial to be a part of Article 21
a region. For instance, according to the annual crime
of the Constitution which guarantees the citizens’
data released by United Nations Office on Drugs
right to life and liberty. However, speedy trial is
and Crime (UNODC), while India has crime rates
not the only marker for proper disposal of cases,
lower than developed countries such as Sweden and
and there is a need to ensure that cases are being
Netherlands, the rate of murder in India, at 3.28
registered by the police and that justice is being
per lakh of population, is much higher than these
delivered. In order to get a more wholesome idea
countries which have rates ranging between 0.71 in
of this, the disposal percentage of cases with police
Sweden and 1.06 in Australia.
and courts, along with the chargesheeting rate by
Looking at figure 1.1 gives an idea of the fact that the police and conviction rate by court have been taken
rate of violent crimes is disproportionately higher here for the calculation of this Index (see table 1.2).
than the rate of total cognizable crimes across most Uttarakhand, Chhattisgarh, and Uttar Pradesh
states, particularly in states such as Delhi, Assam figure at the top of this Index while Assam, Delhi
and Bihar. Whereas in contrast, states such as and West Bengal figure at the bottom.
Kerala have much higher rates of total cognizable
A disaggregated look at the individual variables used
crimes than violent crimes. This difference may
in this Index reveals that the chargesheeting rate and
be attributable to differences in reporting and
disposal percentage of cases by police is much higher
registration of crimes in different states, which are
than the conviction rates and the disposal percentage
reflective of the accessibility and responsiveness of
of cases by courts. As is evident from Figure 1.2, the
the police in those states.
disposal of cases by police is uniformly better than
The cumulative crime rate Index reveals Punjab, the disposal of cases by the courts, indicating that
Himachal Pradesh and Jharkhand to be the states the police are in general more prompt and efficient
recording least rates of crime, while Delhi, Kerala in dealing with cases as compared to the courts.
and Rajasthan have the highest. Delhi also has
However, one of the variables used for the disposal
the highest rates of crimes against women and
Index of courts needs to be analysed more carefully:
children. At the all India level, while the rate of
the conviction rate. There is nothing particularly
total cognizable crime has more or less remained
unusual about conviction rate, but as will be seen in
constant, with a slight decrease in the year 2016,
the following sections, the same becomes lamentable
but the rate of crimes against women, children, SCs
in cases of crimes against vulnerable communities.
and STs have been increasing. The rate of crimes
However, it would be a mistake to put the onus of
against children has had an almost three times
poor conviction rates on the courts alone. Police in

Figure 1.1: Rate of total cognizable crime and rate of violent crime indices
0.1
Andhra Pradesh 0.1
0.1
Assam 0.7
0.1
Bihar 0.4
0.5
Chhattisgarh 0.3
0.3
Gujarat 0.1
0.2
Haryana 0.5
0.1
Himachal Pradesh 0.2
0
Jharkhand 0.3
0.1
Karnataka 0.4
0.9
Kerala 0.5
0.2
Madhya Pradesh 0.3
0.1
Maharashtra 0.4
0
Nagaland 0.1
0.1
Odisha 0.4
0.1
Punjab 0.2
0.2
Rajasthan 0.2
0.4
Tamil Nadu 0.2
0.1
Telangana 0.2
0.4
Uttar Pradesh 0.2
0.6
Uttarakhand 0.2
0.1
West Bengal 0.4
0.3
Delhi UT 1
0.1
AP+ Telangana 0.1
0.2
All India 0.3
Rate of total cognizable crime index Rate of Violent crime index

Note: Index interpretation: 0 indicates worst performing and 1 indicates best performing

Figure 1.2: Disposal of cases by police and court Indices
Note: Index interpretation: 0 indicates worst performing and 1 indicates best performing. Police disposal Index is cumulative score of
chargesheeting rate and percentage of cases investigagted by the police count disposal index is a cumulative score of conviction rate
ands percentage of cases tried by the court

Table 1.2: Disposal of cases by police and courts Index
State rank States arranged from the best performing to worst performing Disposal Index
1. Uttarakhand 0.90
4. Kerala 0.79
5. Tamil Nadu 0.77
7. All India 0.69
8. Nagaland 0.67
9. Rajasthan 0.67
10. Gujarat 0.63
13. Haryana 0.58
14. Karnataka 0.56
15. Telangana 0.56
16. Punjab 0.55
18. Jharkhand 0.45
19. Maharashtra 0.45
20. Odisha 0.45
21. Bihar 0.43
22. West Bengal 0.43
23. Delhi 0.39
24. Assam 0.29
India has three primary responsibilities- to uphold in 2016, less than half the rate of 99 percent in Japan
and enforce law and order, to investigate offences and in China.
and to assist in the prosecution of offenders15. The
quality of investigation and prosecution have a huge
1.1.3 Police diversity Index
impact on the conviction rates of cases. Considering The variables used for police diversity Index are:
the poor state of the forensics departments in India
1. Percentage of SCs in police in proportion to the
such as vacancies of almost 80 percent in the labs
reserved percentage for SCs16
of Bihar (more details given in Chapter 7 on CAG
audit of police) and the lack of sanctioned staff for 2. Percentage of STs in police in proportion to the
investigation, the poor conviction rates are a matter reserved percentage for STs
of concern comes as no surprise.
3. Percentage of OBCs in police in proportion to
However, a caveat needs to be highlighted here. For the reserved percentage for OBCs
the purpose of this study, the data for IPC and SLL
4. Percentage of Muslims in police in proportion
crimes has been merged for all variables so as to get
to the Muslim population in the state17
an overall picture of the police performance. When
looking at the combined rate for both these kinds 5. Percentage of women in police
of offences, the conviction rate ranges between a Reservations or positive discrimination by the
decent 65 and 80 percent in the last five years. But State is an essential tool for a democracy to ensure
when looking at only the IPC crimes, under which proportionate representation of minorities and
a majority of the crimes fall, the conviction rate at vulnerable groups in different sections of society.
the all India level has not crossed even the 50 percent When it comes to the police, this requirement
mark in the last five years, with 46.8 being reported becomes doubly essential, as it has become evident

Table 1.3: Police diversity Index
State rank States arranged from best performing to worst performing Police diversity Index
1. Odisha 0.39
2. Maharashtra 0.37
4. Tamil Nadu 0.36
7. Karnataka 0.30
8. Uttarakhand 0.29
9. Jharkhand 0.26
10. Punjab 0.26
11. Delhi 0.25
12. Kerala 0.24
14. Rajasthan 0.22
15. Bihar 0.21
17. Gujarat 0.20
18. Assam 0.19
19. Haryana 0.19
(Nagaland, Telangana and All India excluded because data not available/not computable)
in the last few years that there is a disproportionate Maharashtra, Odisha, Punjab, Telangana and
incarceration of minorities and other vulnerable Uttarakhand).
communities (more in the section on prisons below).
While this number might seem high at first glance,
Legal provisions for the reservation of SCs, STs
but it needs to be understood in the context of
and OBCs are in place, yet there has been a failure
the fact that not even half the number of selected
to meet even the basic reservation quota for these
states (22) have been able to meet the reservation
communities in the police.
criteria for OBCs, and much lesser for STs and SCs.
To understand this Index in the context of Popular myths pertaining to reservations “eating
reservations, we have calculated the actual percentage up” on the general seats are largely unfounded,
strength of SCs, STs and OBCs in the police force particularly since reservations criteria are set mostly
in proportion to the sanctioned percentage for these in proportion to the percentage of the community
groups in the state. In other words, the final figure in question in that state. Even as of 2016, UP Police
shows the percentage of reservation that has been has met less than 40 percent of the reserved quota
met by the state for these groups. And although the for OBCs, and the percentage of reserved seats filled
all-India figures are not available because of different has indeed fallen drastically in UP from 61 percent
reservation quotas in different states, but a bare look in 2013 to 39.6 percent in 2016. Similarly, in Tamil
at the state-wise figures leaves a lot to be desired. Nadu, as in many other states, the percentage share
of seats reserved for SCs filled has fallen from 91.1
When looking at the five year average, only two out
percent in 2012 to 63 percent in 2016. There is
of the 22 selected states for this study have been
reason to believe, therefore, that things are in fact
able to meet the reserved quota for SCs (Punjab
deteriorating instead of improving when it comes to
and Uttarakhand); six states have been able to fulfil
representation of SCs, STs and OBCs in the police
the reserved quota for STs (Bihar, HP, Karnataka,
force.
Nagaland, Telangana, Uttarakhand); and a slightly
higher number of nine states have been able to Contrastingly, the representation of women in
achieve the reservation benchmark for OBCs police has been going up over the years almost
(Andhra Pradesh, Assam, Jharkhand, Karnataka, uniformly across all states and at the all-India level.

While this in itself may be good news, but there’s to their population in the states and at the all-India
not much to celebrate in the fact that even despite an level, is constantly less than half of the size of
upward graph, the all-India percentage of women their population in India. An even more distressing
in police rests at a shameful 7.3 percent as of 2016. fact is that the data on Muslim representation in
The state with the highest representation (amongst police_provided under Crime in India, NCRB until
the selected states), Tamil Nadu, goes up to 12.9 2013_has since been discontinued. The absence
percent_ just slightly more than 1/3rd of the already of information on this crucial aspect of diversity
modest benchmark of 33 percent reservation. further clouds the possibilities of improvement in
Of the 22 selected states, 17 have provisions for this aspect of policing.
reservation for women in the police force_ranging
between 20 percent in Uttar Pradesh and Karnataka
1.1.4 Police infrastructure Index
to 38 percent in Bihar. Besides this, several laws, The variables used for calculating the police
particularly with respect to women and children, infrastructure Index are:
have legal provisions that make the presence of a
1. Percentage utilisation of grants for police
female police officer mandatory.
modernisation
One section that does not have legal provisions
2. Strength of actual police per lakh of population
guarding its representation is that of Muslims18
as a percentage of sanctioned police per lakh of
who happen to be present in almost all states of
population
India and continue to have disproportionately low
representation in the police force. This is worrying 3. Strength of actual police stations as a percentage
notably in the backdrop of disproportionately high of sanctioned police stations
Muslim representation in the prisons. Muslim
4. Police expenditure as a percentage of state
representation in police, calculated in proportion
budget
Table 1.4: Police infrastructure Index

State rank States arranged from best performing to worst performing Police infrastructure Index
1. Delhi 0.48
2. Nagaland 0.42
3. Rajasthan 0.41
4. Punjab 0.38
5. Tamil Nadu 0.38
7. Haryana 0.35
8. Jharkhand 0.34
9. Kerala 0.33
11. Uttarakhand 0.33
12. Assam 0.32
14. All-India 0.31
17. Odisha 0.30
18. Telangana 0.30
19. Bihar 0.29
20. Karnataka 0.27
22. Gujarat 0.25

5. Training expenditure as a percentage of total 1. Percentage non-utilisation of prison budget
police expenditure
2. Percentage of SC prisoners in proportion to SC
6. Police personnel given in-service training as a population in the state
percentage of total police strength
3. Percentage of ST prisoners in proportion to ST
Any progress in different dimensions of policing population in the state
is necessarily based upon the presumption of the
4. Percentage of Muslim prisoners in proportion
existence of a basic enabling structure which is in
to Muslim population in the state
place and is functional. However, there are apparent
gaps within this infrastructure itself, as is evident 5. Number of undertrial prisoners as a percentage
from the above Index. One of the most obvious of the total strength of prisoners
failure of the State is in its inability to minimise
Several studies in different states have been conducted
vacancies in the police force. Averaged out for 2012-
on the disproportionate representation of minorities
16, a gap of 24 percent is seen in the strength of
and vulnerable communities in the prisons. This
actual police proportionate to the sanctioned police
has been found to be so particularly in the case of
strength per lakh of population. While the gap
Muslims. When coupled with poor conviction rates
between actual and sanctioned number of police
and incidents of false implication, as recognised
stations is minimal, but when the BPRD standards
by courts, this points to a deeper problem of biases
for police stations are applied, the states fall short of
within the structure leading to hyper-incarceration
as much as 44 percent as reported in UP by the CAG
of a particular section of the society. As we will see
Performance Audit of the police there.
later in the survey findings, there is also a significant
According to Global Peace Index 2017, India public agreement to the statement that often Dalits,
lost close to $742 billion on violence in 2016, or Adivasis and Muslims are falsely implicated.
8.6 percent of the GDP. In comparison, the State
For the analysis of this Index, the percentage of
budgeted only 3.56 percent for the police in the same
SC, ST and Muslim prisoners has been taken in
year, thus reflecting a skewed cost-benefit ratio.
proportion to their respective populations in the state
Another issue is that of under-utilisation of funds, as a five-year average, and it was found that in case
which continues to be a chronic problem in many of SCs, only four states (West Bengal, Uttarakhand,
government institutions. Of the amount allocated Punjab and Karnataka) out of the selected 22 have
for police modernisation, as an average of five years, SC prisoners in proportion to or less than their
13 of the 22 selected states, i.e., more than half the population in the State; in case of STs this number
number, have not been able to spend even 50 percent is three (Himachal Pradesh, Madhya Pradesh,
of the amount. The utilisation percentage is as low Nagaland), and in case of Muslims, all of the 22
as 0 percent in Andhra Pradesh, West Bengal, Tamil states have a higher proportion of Muslim prisoners
Nadu and Bihar in the year 2015. CAG reports than the Muslim population in the State, as evident
across states tell a similar story of lapsing of funds in Figure 1.3. The differences are as glaring as more
under the police modernisation scheme due to than 7 times the Muslim population in Nagaland
under-utilisation by states. in 2014, almost 6 times the population of STs in
Uttar Pradesh in 2015 and more than double the
Training is an indispensable component in the task
percentage of SC population in three states in 2015
of ensuring a modern, upskilled and sensitive police
(Kerala, Gujarat and Assam). At the all-India level
force. But a mere 6.5 percent of the total police force
as well, this ratio continues to be skewed adversely
has received training in the last five years. While
against SCs, STs and Muslims through all five years.
states such as HP have progressed in this respect,
with 23 percent of the police force being imparted A major consequence of the failure of quality
training in the year 2016, but the overall situation and timely disposal of cases by the court is the
has ample room for improvement. Percentage of overcrowding of jails due to excessive number of
police personnel provided training depends largely undertrials not commensurate with the available
on the percentage funds allocated for training, but capacity of the prisons. Overcrowding in prison
that is a poor 1.38 percent at the all India level for is to the extent of more than twice the available
the last five years. Many states have allocated less capacity, as in the case of Delhi and Chhattisgarh
than 1 percentage of the total police expenditure for in 2015. Undertrials in the whole country form the
training. major chunk of the prison inmate population, with
their percentage ranging between 50.6 percent in
1.1.5 Prison data Index19 Himachal Pradesh and 84.4 percent in Bihar (2011-
The variables used for the calculation of prison data 15 average). In an adversarial justice system such as
Index are20: the one India has wherein the accused is presumed to
be innocent until proven guilty, to have the accused

Table 1.5: Prison data Index
State rank States arranged from best to worst performing21 Prison data Index
3. Kerala 0.78
5. Karnataka 0.77
7. Gujarat 0.76
8. Rajasthan 0.76
9. West Bengal 0.76
10. All India 0.76
11. Assam 0.74
12. Jharkhand 0.74
14. Odisha 0.73
15. Uttarakhand 0.73
16. Tamil Nadu 0.71
17. Bihar 0.68
(Nagaland, Telangana, Haryana, Punjab and Delhi excluded because data not available/not computable)
form the major share of prison population appears While the overall disposal indicators discussed above
unreasonable. may in themselves be a cause for worry, the disposal
indices for the cases of crimes against SCs, STs,
Unlike as in the police infrastructure, however, the
women and children are almost everywhere much
utilisation of prison budget is up to the mark and is
below the overall disposal Index. This difference
less than 80 percent in only two states, Assam and
is shown clearly in figure 1.4. When looking at the
Jharkhand, as an average of five years. The all-India
disposal of cases against SC Index, it is only in the
average for five years of prison budgets is 89 percent.
state of Bihar and Maharashtra that the disposal
With the introduction of educational facilities to
indices for crimes against SCs are not below those
prisoners, vocational training imparted to prisoners,
for all crimes. In disposal of cases of crimes against
introduction of the concept of open jails, etc.
STs, only Bihar, Maharashtra and Odisha have
the prison infrastructure is adopting progressive
indices equal to or higher than the disposal Index of
measures.
overall crimes. Assam, Bihar, Jharkhand, Nagaland
1.1.6 Disposal of cases of crimes against SCs, and Odisha are the five states that have higher or
STs, women and children Index same disposal indices of cases of crimes against
children than the overall crime disposal Index.
The variables used for the calculation of disposal
of cases of crimes against SCs, STs, women and The disposal Index of cases of crimes against
children indices are: women is not being compared with other disposal
indices because the data on the variables of disposal
1. Chargesheeting rate of cases of crimes against percentage of cases of crimes against women by
SCs, STs, women and children22 police and by courts is not available for the years
2. Disposal percentage by police of cases of crimes 2012-15, and only the 2016 data has been used for
against SCs, STs, women and children23 this variable. However, the apparent gaps are in the
rates of conviction. While the all-India conviction
3. Conviction rate of cases of crimes against SCs,
rate for total IPC and SLL crimes is 75 percent,
STs, women and children
that for cases of crimes against women is only
4. Disposal percentage by court of cases of crimes 21.1 percent, less than one-third of the former.
against SCs, STs, women and children This difference is further compounded with the
presumption that a significant number of cases of

Figure 1.3 : Percentage of Muslim prisoners vis-a-vis percentage Muslim population in the states
Andhra 12.3
Pradesh 9.6
34.4
Assam 34.2
17.1
Bihar 16.9
5.0
Chhattisgarh 2.0
23.0
Gujarat 9.7
9.2
Haryana 7.0
Himachal 4.4
Pradesh 2.2
20.5
Jharkhand 14.5
Karnataka
16.4
12.9
26.9
Kerala 26.6
Madhya 11.7
Pradesh 6.6
30.3
Maharashtra 11.5
12.4
Nagaland 2.5
Odisha
5.6
2.2
3.9
Punjab 1.9
17.5
Rajasthan 9.1
15.6
Tamil Nadu 5.9
Uttar Pradesh
25.1
19.3
Uttaranchal 29.5
14.0
West Bengal 46.0
27.0
Delhi 21.6
12.9
AP+Telangana 15.3
9.6
Total (All 19.8
India) 14.2
Percentage of Muslim prisoners (2011-15 average) Percentage Muslim population in the state (Census 2011)
Notes: Figures are in percentages

Table 1.6: Disposal of cases of crimes against SCs, STs, women and children Index
S.no. States Disposal of cases of Disposal of cases of Disposal of cases Disposal of cases
crimes against SCs crimes against STs of crimes against of crimes against
women children
1. Andhra Pradesh 0.43 0.45 0.64 0.48
2. Assam NA NA 0.44 0.35
3. Bihar 0.45 0.48 0.66 0.45
4. Chhattisgarh 0.61 0.61 0.81 0.58
5. Gujarat 0.50 0.49 0.74 0.46
6. Haryana 0.57 NA 0.75 0.54
7. Himachal Pradesh 0.49 NA 0.72 0.50
8. Jharkhand 0.38 0.43 0.59 0.53
9. Karnataka 0.46 0.46 0.61 0.45
10. Kerala 0.38 0.42 0.68 0.47
11. Madhya Pradesh 0.61 0.61 0.82 0.57
12. Maharashtra 0.45 0.46 0.63 0.40
13. Nagaland NA NA 0.80 0.71
14. Odisha 0.42 0.45 0.66 0.45
15. Punjab 0.46 NA 0.56 0.50
16. Rajasthan 0.56 0.51 0.86 0.55
17. Tamil Nadu 0.46 0.46 0.57 0.51
18. Telangana 0.43 0.38 0.62 0.45
19. Uttar Pradesh 0.59 0.62 0.77 0.57
20. Uttarakhand 0.57 0.47 0.74 0.57
21. West Bengal 0.39 0.38 0.67 0.36
22. Delhi 0.35 NA 0.50 0.36
23. Andhra Pradesh + 0.42 0.45 NA NA
Telangana
24. All-India 0.49 0.50 0.66 0.48
crimes against women continue to go unreported 1.2 Concluding objective data analysis
in the country, a presumption which is further
strengthened by comparing the NCRB data with The whole point of doing a state-wise comparison is
that of the National Family Health Survey (NFHS) to be able to pick out the best and worst performing
as demonstrated by Gupta (2014). states. Overall, HP has performed consistently well
on multiple parameters, while Chhattisgarh, UP
Similarly, as in the case of women, the all India and MP have good track records in the category
conviction rate for cases of crimes against SCs is of disposal of cases. On the other hand, UP is
24.5 percent, that for STs is 19.9 percent and for performing poorly in all other aspects of policing,
children is 31.9 percent. Differences in all other while Delhi, West Bengal and Assam are performing
variables are also apparent throughout the five poorly on several parameters.
years. However, an improvement that needs to be
noted is that the disposal Index of cases of crimes It also needs to be admitted here that many
against women, although quite poor, has improved progressive policing practices and policies have not
over the years from 0.56 in 2012 to 0.63 in 2016. been given a due mention. However, the criminal
One of the contributing factors towards this upward justice system, when seen as a public good from
thrust may be the introduction of new legislation on the Rawlsian perspective of distributive justice,
prohibition of crimes against women, the Criminal fails to prove itself an egalitarian institution, and
Law Amendment Act, 2013. functions detrimentally against the least advantaged

Figure 1.4: Disposal Indices
0.43
Andhra 0.45
Pradesh 0.48
0.61
0.45
0.48
Bihar 0.45
0.43
0.61
0.61
Chhattisgarh 0.58
0.85
0.5
0.49
Gujarat 0.46
0.63
0.38
0.43
Jharkhand 0.53
0.45
0.46
0.46
Karnataka 0.45
0.56
0.38
0.42
Kerala 0.47
0.79
0.61
Madhya 0.61
Pradesh 0.57
0.73
0.45
0.46
Maharashtra 0.4
0.45
0.42
0.45
Odisha 0.45
0.45
0.56
0.51
Rajasthan 0.55
0.67
0.46
Tamil 0.46
Nadu 0.51
0.77
0.43
0.38
Telangana 0.45
0.56
0.59
Uttar 0.62
Pradesh 0.57
0.83
0.57
0.47
Uttarakhand 0.57
0.9
0.39
0.38
West Bengal 0.36
0.43
Disposal of cases of crimes against SCs Disposal of cases of crimes against STs
Disposal of cases of crimes against children Disposal of Cases by Police and Courts Index
Note: Disposal of cases of crimes against women left out of the graph because data on disposal percentages of police and courts
no available for the years 2012-15. States for which data on any one of the indices is not available have been left out as well.Index
interpretation- 0 indicates worst performing and 1 indicates best performing

Table 1.7: Comparison of state performance
Index Three best performing states Three worst performing states
Crime Index Punjab, Himachal Pradesh, Delhi, Kerala, Rajasthan
Jharkhand
Disposal of cases by police and Uttarakhand, Chhattisgarh, Uttar Assam, Delhi, West Bengal
courts Index Pradesh
Police diversity Index Odisha, Maharashtra, Himachal Uttar Pradesh, West Bengal, Haryana
Pradesh
Police infrastructure Index Delhi, Nagaland, Rajasthan Andhra Pradesh, Uttar Pradesh,
Gujarat
Prison data Index Himachal Pradesh, Madhya Pradesh, Uttar Pradesh, Bihar, Tamil Nadu
Kerala
Disposal of cases of crimes against Chhattisgarh, Madhya Pradesh, Delhi, Kerala, Jharkhand
SCs Index Uttar Pradesh
Disposal of cases of crimes against Uttar Pradesh, Chhattisgarh, Madhya West Bengal, Telangana, Kerala
STs Index Pradesh
Disposal of cases of crimes against Rajasthan, Madhya Pradesh, Assam, Delhi, Punjab
women Index Chhattisgarh
Disposal of cases of crimes against Nagaland, Chhattisgarh, Madhya Assam, Delhi, West Bengal
children Index Pradesh
communities such as SCs, STs and minorities. This data were not corresponding. In some instances, the
is evident from the fact that the parameters on methods of data calculation were changed midway.
which the failures of the criminal justice system Data on indicators such as on custodial violence,
appear the most egregious are those relating to the was not available in a state-wise format for several
diversity within the police force, disproportionate years, because of which it had to be left out. Some
representation of minorities in the prisons and variables were dropped because of inconsistencies,
disposal of cases of crimes against SCs, STs, particularly in prison data. Also, data on some of
women and children. This signals the need for the variables has either been discontinued, such
urgent systemic reform targeted at the society’s most as the data on Muslims in police forces, or is not
vulnerable sections. available for some specific years, such as the data
on strength or percentage of SCs, STs and OBCs in
1.3 Limitations of the study police force for the year 2015.
One of the major limitations of the study is its References
inability to tap into an important component
on policing, that of excesses and human rights Ayaz, Shakeb. “Police Reforms: A Chronology” Common
violations by the police. While an attempt has been Cause Journal, 34, no. 3 (July-September 2015): 14-24.
made in the survey to get citizens’ perceptions Burke, James. “Japan’s Court System Has A 99%
on and their experience of police excesses and Conviction Rate”. Vision Times, January 06, 2016. http://
atrocities, but empirical data from official sources www.visiontimes.com/2016/01/06/japans-court-system-
on this issue are so scanty that they could not be has-a-99-percent-conviction-rate.html (accessed 21
interpreted in a meaningful manner. Perhaps the February 2018)
issue lends itself to a separate study rather than as Chandra, P. “NCRB Data Names Kerala as India’s ‘Crime
part of a large-scale survey. The study also leaves out Capital’, But Here’s Why It’s a Good Thing”. India Times,
many crucial areas like the functioning of Armed September 27, 2016. https://www.indiatimes.com/news/
Forces (Special Powers) Act (AFSPA) in Jammu and india/ncrb-data-names-kerala-as-india-s-crime-capital-
Kashmir and North-Eastern states, and the presence but-here-s-why-it-s-a-good-thing-262492.html (accessed
of para-military forces in states like Chhattisgarh November 24, 2017)
and Jharkhand which are under the influence of left- Connor, Neil. “Chinese Courts Convict More Than 99
wing extremism. Per Cent of Defendants”. The Telegraph, March 14, 2016.
http://www.visiontimes.com/2016/01/06/japans-court-
When specifically looking at the objective data system-has-a-99-percent-conviction-rate.html (accessed
analysis, a major problem was the discrepancies in 21 February 2018)
official data sets. Even the NCRB and the BPRD

Dr. Subramaniam Swamy vs Dr. Manmohan Singh and 4 In case of the variables that had a negative direction, i.e.,
Anr., 2012 Civil Appeal No. 1193 of 2012 (Arising out of the higher figures were reflecting depreciating performance,
SLP (C) No. 27535 of 2010) such as the variables under crime Index and prison Index,
the indices were subtracted by 1 so as to reverse the direction
Gupta, Aashish. “Reporting and Incidence of Violence of these figures for the purpose of uniformity.
Against Women in India”. Research Institute for 5 For calculation of the crime rate Index, the average of the
Compassionate Economics. http://riceinstitute.org/ variables has been subtracted by 1 so that a higher Index
wordpress/wp-content/uploads/downloads/2014/10/ indicates better state performance
Reporting-and-incidence-of-violence-against-women-in-
6 The NCRB methodology for calculation of rate of total
India-working-paper-final.pdf (accessed 14 January 2018)
cognizable crimes against women, children, SCs and STs
Hussainara Khatoon vs Home Secretary, State of Bihar, changed in 2012. Therefore, for the Index calculation of
AIR 1978 SC 597 these four variables, data only from 2012-2016 has been
used, instead of using maximum and minimum values from
Institute for Economics and Peace. 2017. “Global the previous five years.
Peace Index 2017” http://visionofhumanity.org/ 7 Crimes included under the category of “violent crimes”
app/uploads/2017/06/GPI17-Report.pdf (accessed 9 according to NCRB: murder, attempt to commit murder,
February 2018) culpable homicide not amounting to murder, attempt to
commit culpable homicide, dowry deaths, kidnapping &
Joshi, G.P. Policing in India: Some Unpleasant Essays. New abduction, dacoity, making preparation & assembly for
Delhi: Atlantic and Distributors (P) Ltd., 2013 committing dacoity, robbery, riots, arson, rape and attempt
Joshi G.P., Sharma P.P.P., Bedi B.S. and Sharma J.K. to commit rape.
“Promotion Prospects of Constabulary in the Police”. 8 Crimes included under the category of “crimes against
Bureau of Police Research and Development, Ministry of women” according to NCRB (2014): rape, attempt to
Home Affairs, Government of India, 1990. commit rape, kidnapping and abduction of women, dowry
deaths, assault on women with intent to outrage her
Lombroso, Cesare. Criminal man. Translated by M. modesty, insult to the modest of women, cruelty by husband
Gibson & N.H. Rafter. Durham, NC: Duke University or his relatives, importation of girl from foreign country,
Press, 2006. abetment of suicide of women, The Dowry Prohibition Act
1961, the Indecent Representation of Women (Prohibition)
M, Raghava “Will prosecution of five police officers be Act 1986, the Commission of Sati Prevention Act 1987, the
sanctioned?” The Hindu, March 23, 2012. http://www. Protection of Women from Domestic Violence Act 2005 and
thehindu.com/todays-paper/tp-national/tp-karnataka/ the Immoral Traffic (Prevention) Act 1956 (women related
will-prosecution-of-five-police-officers-be-sanctioned/ crimes only)
article3166010.ece (accessed November 27, 2017). 9 Crimes included under the category of “crimes against SCs”
according to NCRB (2015): Protection of Civil Rights Act
Ministry of Home Affairs, Government of India.
1955, atrocities committed against persons belonging to SCs
“Setting up of National Security Commission,” January by non-SCs, i.e., where SC/ST (Prevention of Atrocities)
03, 2018, http://pib.nic.in/newsite/PrintRelease. Act has been applied along with various sections of IPC,
aspx?relid=175277 (accessed February 15, 2018). crimes committed against SCs where SC/ST Prevention of
Atrocities Act has not been applied and only IPC sections
Prakash Singh vs Union of India, Writ Petition (Civil) 310 have been involved, the Employment of Manual Scavengers
of 1996 and Construction of Dry Latrines (Prevention) Act 1923,
Rao, U. N., Dr., & Tiwari, Arvind. “A Study on other SLL crimes, assault on SC woman with intent to
outrage her modesty and insult to modesty of SC women.
Non-Registration of Crimes: Problems & Solutions”
(Rep.). Bureau of Police Research and Development, 10 Crimes included under the category of “crimes against STs”
Ministry of Home Affairs, Government of India. according to NCRB (2015): Protection of Civil Rights Act
http://www.bprd.nic.in/WriteReadData/userfiles/ 1955, atrocities committed against persons belonging to STs
by non-STs, i.e., where SC/ST (Prevention of Atrocities)
file/201612200235022990797Report-Non-Registrationof
Act has been applied along with various sections of IPC,
CrimesProblems&Solutions.pdf (accessed November 24, crimes committed against STs where SC/ST Prevention of
2017) Atrocities Act has not been applied and only IPC sections
The National Police Commission. Fourth report, June 1980. have been involved, the Employment of Manual Scavengers
and Construction of Dry Latrines (Prevention) Act 1923,
New Delhi: Government of India. 1980.
other SLL crimes, assault on ST woman with intent to
outrage her modesty and insult to modesty of ST women.
(Endnotes)
11 Crimes included under the category of “crimes against
1 Section 7, The Police Act, 1861: “Appointment, dismissal, children” according to NCRB (2014): murder, attempt to
etc. of inferior officers”. commit murder, infanticide, rape, unnatural offence, assault
on women (girl child) with intent to outrage her modesty,
2 This was the point of contention in the case of Dr. insult to the modesty of women (girl child), kidnapping
Subramaniam Swamy vs Dr. Manmohan Singh and Anr., and abduction, foeticide, abetment of suicide of child,
2012 exposure and abandonment, procuration of minor girls,
3 In the course of data collection, it was discovered that the importation of girls from foreign countries (under 18 years
figures projected by NCRB and BPRD on the same variable of age), buying of minors for prostitution, selling of minor
were often inconsistent. For instance, while NCRB reports for prostitution, Prohibition of Child Marriage Act 2006,
the total police strength in India to be 1731537 for the Transplantation of Human organs Act 1994 (for persons
year 2013, whereas the BPRD figure for the same year is below 18 years of age), Child Labor (Prohibition and
1722786. Regulation) Act 1986, Immoral Traffic (Prevention) Act
1956, Juvenile Justice (care and Protection of Children) Act

2000 and Protection of Children from Sexual Offences Act 18 Some Muslim communities come under the category of
2012. OBCs as scheduled by the government.
12 Data for Telangana, across all variables, has been taken 19 For calculation of the prison data Index, the average of the
only for the years 2014-2016, and a five-year average is variables has been subtracted by 1 so that a higher Index
not available. Therefore, the data for Andhra Pradesh and indicates better state performance
Telangana has been combined.
20 Percentage of OBC prisoners in proportion to OBC
13 Chargesheeting rate is the number of IPC and SLL cases in population in the state left out as a variable because data on
which chargesheets were filed by the police as a percentage OBC population in states not available in Census 2011. All
of the total number of cases disposed by the police other population data taken from Census 2011.
14 Conviction rate is the number of IPC+SLL cases convicted 21 Data for Haryana, Nagaland, Punjab, Telangana and Delhi
by court as a percentage of the total number of IPC+SLL not available/ not computable
cases tried by the court in the year
22 Data on chargesheeting rate and conviction rate for cases of
15 For a more detailed understanding of the role, functions and crimes against SCs and STs not available for the years 2014
duties of the police in India see: http://www.bprd.nic.in and 2015.
WriteReadData/userfiles/file/6798203243-Volume%202.
pdf (accessed 9 January 2018) 23 Data on disposal percentage of cases by police and courts
not available for the years 2014 and 2015 for cases of crimes
16 Data on percentage of SCs, STs and OBCs in police force against SCs, STs and children. Data on disposal percentage
not available for the year 2015. of cases of crimes against women by police and courts not
available for the years 2010-2015, therefore, only the data for
17 Data on Muslims in police force taken from Crime in India, 2016 considered for Index calculation.
NCRB. It was discontinued after 2013. No data on this
variable available in the BPRD report.

CHAPTER 2:
EXPERIENCE WITH THE POLICE
Women protesting at Ashoka Road, New Delhi, against rising prices of essential commodities and calling for better
security of women (Credits: Sonu Mehta, Hindustan Times, 13 July 2013, New Delhi)

Experience with the Police
2.1 Introduction illustrates the key findings of the Common Cause-

CSDS survey on some of these issues, with its main
Mechanisms of social control such as police are focus on people’s actual reported experience with
a universal feature of all human societies. They the police, while the remaining sections intend to
are the most visible representation of state and gauge people’s perception. The first section will
regime power. In democratic societies, the system focus on the experiences of those respondents who
of social control cannot completely neglect popular contacted the police or vice versa in the last 4-5
demands and theoretically and ideally, all police years. The second section will examine respondents’
activities should reflect commitment to the rule of experiences and perception of incidence of crime in
law. While India is known as the world’s largest their locality, their satisfaction with the police and
practising democracy, what is less understood is with registration of crime. The third section will
how it has dealt with policing such a large, complex elaborate on people’s views on policing services
and turbulent society. The Indian Police Service’s such as helpline numbers for women, children etc.
success in combating and handling terrorism
has been well acknowledged (Raghavan, 2003). 2.2. Contact with the police
However, alongside charges of corruption, arbitrary
and discriminate action towards vulnerable groups The most striking aspect of this survey is that a
has been levelled and its religious neutrality has been very small percentage of respondents reported any
questioned. As an organ of the state, it has been contact with the police. Out of 15,562 respondents
subjected to public scrutiny because of its inability who were interviewed, only one in seven (14%) said
to deliver services and failure to develop a functional they had interacted with the police in the last 4-5
relationship with citizens. years (Figure 2.1). This includes direct contact of
respondents and/or that of their family members.
The most important assessment of police work and Over eight out of ten respondents did not have any
performance is done by the people whom it seeks contact with the police in the last 4-5 years.
to manage and control and who are also the clients
of their services. Assessment of police performance Analysing by demographic variables such as gender
is done by the public at two levels— assessment of reveals that men were more likely to have contacted
individual level experience involving contact between the police than women. While only 11 percent of
police and individual and general assessment of women respondents said they had contacted the
police performance by both their overall experience police, among men the figure of police contact was
in society, perception and opinions on policing and six points higher at 17 percent (Figure 2.2). Studies
through contact if any with the police. This chapter have documented that the lack of gender friendly
Figure 2.1: Over four in five respondents did not have any contact the police in recent past
4%
14%
Respondents who had no police contact in the
last 4-5 years
Respondents who had police contact in the

last 4-5 years
No response
82%
Note: Question asked: From time to time, for different purposes, people have some kind of contact with the police. In the last 4-5
years, have you or your family member had any kind of contact with the police?

Figure 2.2: Men are more likely to have contacted the police
85
79
17
11
4 4
Contacted police Did not contact police No response
Men Women
Note: Figures are percentages.
atmosphere in police stations and the misbehaviour help or come in contact with the police than those
and extortionist nature of police personnel in India who are poor (Figure 2.3). As far as educational
inhibits women from visiting the police station alone levels are concerned, those who are more educated,
(Sharma, 2005). college educated or above, were nearly twice more
likely to have contacted the police than non-literates
Qualitative evidence indicates that police brutality
(Figure 2.4). In terms of communities, it is Muslims
towards rickshaw pullers, beggars and working
who reported the highest contact with the police at
class labourers is commonplace in India (Verma,
17 percent. The same does not hold true for other
2005). This in turn is likely to reduce poor people’s
marginalized communities such Scheduled Castes
interaction with the police and their subsequent
and Scheduled Tribes; in fact they were least likely
trust levels. The same is also corroborated by the
of all communities to have contacted the police
survey. Socio-economic class has an evident bearing
(Figure 2.5).
on police contact- those who are rich and well to
do were twice more likely to have sought police’s
Figure 2.3: The well-to-do reported greatest police contact

83 84
78 79
20 18
13
10
6 6
2 3
Upper class Middle class Lower class Poor
Note: Figures are percentages

Figure 2.4: Those who are most educated report highest police contact
81 83 84
79
18
15 12 10
3 4 5 6
College and Up to matric Up to Primary Literate

above
Figure 2.5: Muslims reported highest police contact

82 84 84
81 79
15 17
13 12 12
5 4 4 4 4
Upper castes Other Scheduled Scheduled Muslims

Backward Castes Tribes
Classes
Furthermore, among all those who reported contact Muslims, Scheduled Tribes and those who are poor
with the police, 67 percent approached the police are less likely to contact the police on their own and
themselves and 17 percent were contacted by secondly that they are more likely to be contacted by
the police. About 11 percent said the contact was the police. This aspect of disproportionate minority
mutual (Figure 2.6). Hindu Upper Castes were contact might have possible links with the over
least likely to have been contacted by the police representation of these minorities in different stages
(13 percent). Scheduled Tribes and Muslims on the of the criminal justice system (see Appendix for
other hand were most likely to have been contacted Government of India data on overrepresentation of
by it (23 percent and 21 percent respectively; Table minorities in prisons).
2.1).In terms of class hierarchy, the likelihood of
The most commonly cited reason for police contact
the police contacting a person is nearly twice as
pertains to complaints of property related crime
high amongst the poor compared to the upper class
(15%) and physical assault (14%; Figure 2.8). Nearly
(21 percent as opposed to 12 percent; see Figure
one in ten persons contacted the police to resolve a
2.7). These findings reflect two possibilities: firstly,

Figure 2.6: Mode of police contact
5%
11%
Complainant contacted the police
Police contacted the complainant

17%
Police and complainant contacted each other
67% (multiple cases of contact)
No response
Note: Question asked: Did you or someone from your family contact the police or the police contacted you?
Figure 2.7: The poorest are nearly twice as likely to have been contacted by the police as the rich
74
68 70
60
21
16 16
12 10 11 10 11 8
4 5 4
Complainant contacted police Police contacted complainant Both No response
Table 2.1: Mode of police contact by caste

Complainant Police contacted Police and complainant contacted No response
contacted police complainant each other (multiple cases of contact)
Upper castes 73 13 9 5
OBCs 69 17 10 4
Scheduled Castes 68 16 10 6
Scheduled Tribes 61 23 9 7
Muslims 60 21 15 4
Note: Figures are percentages. Sample size for Christians and Sikhs was low and hence the figures haven’t been reported.

family dispute, about 9 percent contacted because of the police over phone as opposed to 69 percent who
loss of important documents and 8 percent stated visited the police station (Figure 2.9). Men and
domestic violence as the reason for police contact. women did not vary in their type of police contact.
Another important finding of the survey is that not
Disaggregating by class we see that the upwardly
a single respondent’s initial contact with the police
mobile, well-to-do respondents were twice more
was over the internet.
likely to have contacted the police for loss of
goods, documents etc. than the poor. Complaints 38 percent or nearly four in ten respondents were
of domestic violence and family dispute were most accompanied by a family member to the police
likely to be reported by poor. station, 16 percent sought an influential person’s
help and 20 percent of the respondents visited the
A large proportion of persons continue to visit the
police station alone (Figure 2.10). Women were
police station for their work: among those who
more likely to be accompanied by a family member
reached out to the police, only 14 percent contacted
Figure 2.8: Reasons for police contact
Property related crime
Physical assault
15%
Family dispute
17%
Loss of essential goods and documents
Domestic violence
14%
10% Authorization, verification of documents
Petty disputes (water, car parking etc.)
2%
Accompanied a friend/relative to the police station
2%
3% 11% Caste or religion related dispute
4% Sexual assault
5%
9% Other reasons
8%
No response
Note: Question asked: What was the reason for contacting the police?
Figure 2.9: A large plurality of people continue to visit the police station for their work
1%
0%
7%
9% Visited the police station

Over the phone
Police visited home/ workplace
14%
Via internet/ online
Other
69%
No response
Note:Question asked: How did you first contact the police - over the phone, visited the police station, over internet, police visited
home/ workplace?

Figure 2.10: About two-fifth people sought a family member’s help in contacting the police
9%
3%
Family member
14% 38%
Went alone
Neighbour/friend
Influential person
16% Any other person

No response
20%
Note: Question asked: Who assisted you in contacting the police or visiting the police station - family member, influential person,
neighbour/ friend, any other person or you went alone?
(45%) and less likely to visit the police station at the very beginning. Not only is reporting and
alone (14%). This is likely to be due to the deeply recording of crime arbitrary, often complainants
patriarchal attitudes of police personnel, absence of are asked to resolve the dispute by arriving at a
policewomen in every police station and its overall compromise or given the false impression that
hostile environment. Furthermore, 34 percent of the the crime has been registered. In many cases,
male respondents stated that they were accompanied victims do not report the crime to the police due
by a family member and they were also more likely to fear of secondary victimisation, long drawn and
than women to visit the station alone (23%). embarrassing trial proceedings and uncertainty of
the perpetrator getting punished. These challenges
2.2.1. Registration of complaints thereby result in mistrust of police. Findings from our
In India, while the registration of complaints is survey suggest that among those who had any kind
mandated under law, many complaints are not of contact with the police in the last 4-5 years, three-
registered. Preventing, refusing and delaying the fifth respondents were able to successfully register
process of First Information Report (FIR hereafter) their FIR/ complaint1 and about 24 percent were
and complaint registration impede access to justice unable to do so. Those in rural areas were relatively
Figure 2.11: Registration of complaints/ FIR has a direct impact on people’s satisfaction levels
73
60
33
22
FIR/complaint registered FIR/complaint not registered
Satisfied with police help Dissatisfied with police help
Note: Figures are percentages. Rest of the respondents did not respond to the question.

more likely to report success in filing complaint/ non-registration was because they were asked to
FIR. Furthermore, registration of FIR/complaint pay a bribe by the police. A disproportionately high
corresponds with higher satisfaction levels with the number of 1 in two respondents (51%) did not reveal
police. A little less than three fourths (73%) of those the reason for non-registration of complaint/ FIR.
who managed to get their FIR/ complaint registered
There is a difference of ten percentage points in the
were satisfied with the help they received (Figure
complaints/FIR that were read out (52%) and those
2.11). Among those whose complaint or FIR was
that were written (42%). However, when looked at
not registered, satisfaction with police help was 13
in terms of locality a divergent trend emerges. The
points less at 60 percent. That said, it is significant
FIR/complaint was far more likely to be read out
to note that even a majority of them were satisfied
in rural areas than urban areas (57% as opposed to
with the help they received from the police. In other
40%; see Table 2.2). On the other hand, urban areas
words, respondents with negative police contact (for
accounted for a greater percentage of complaints
example: non-registration of complaint/FIR) did
that were formally written and recorded (52%).
not develop too high a negative attitude towards
Analysing by gender of respondents shows that
police as one would have expected. Only one-third
complaints/ FIR were more likely to have been read
of them did, three-fifths did not.
out to women (55%) than men (50%). Men were
Furthermore, the survey suggests that one-fifth more likely to have received written complaints/
respondents (19%) whose complaint or FIR was FIR (44%) than women (38%).
not registered were asked to resolve the matter or
Nearly six in ten of those whose complaint/FIR
arrive at a compromise (Figure 2.12). This was the
was registered, received a copy of their complaint
most commonly cited reason for non-registration
whereas three in ten did not (Figure 2.13). Men and
of complaint/ FIR. Nearly one in ten (9%) said the
Figure 2.12: Reasons for non-registration of FIR
8%
9%
Crime doesn’t require FIR
Asked for a bribe
51% 19% Asked to resolve the matter/ compromise
Miscellaneous reasons
No response
13%
Note: Only among those who reported non-registration of FIR; n= 518

Question asked: Why did the police not file your complaint/ FIR?
Table 2.2: Form of FIR registration by locality and gender

Oral / Read out Written Email/ via internet Other No response
Overall 52 42 1 1 4
Rural 57 37 2 1 3
Urban 40 52 1 2 5
Men 50 44 1 1 4
Women 55 38 1 1 5
Question asked: (If FIR was filed) How was the FIR registered – was it read out, written or via mail?
1
In the survey, the terms complaint and FIR were used interchangeably because many respondents would not be famil-
iar with the difference. However, the authors of this report recognise the distinction between the two.

Figure 2.13: Copy of FIR/ Complaint
11%
Those who received their
FIR/ Complaint
Those who did not receive

30% a copy of their FIR/Complaint
59%
No response
Note: Question asked: (If FIR was filed) Did you get a copy of the complaint/ FIR?
those in urban areas (Table 2.13) were more likely to institution in numerous countries across the world
receive a copy of their complaint/FIR. and those in India considered it as the second most
corrupt institution (Transparency International,
Corruption is a powerful obstacle to economic
2012). According to findings from our survey,
development and a dangerous phenomenon that
over one-third respondents who had contacted
impedes the growth of public institutions. This
the police admitted that they paid bribe whereas a
danger is hugely amplified when it is the police, an
much larger cohort of 1 in two respondents did not
institution that exists to protect the public and is the
(Figure 2.14). While men and women were nearly
enforcer of rule law, themselves who are corrupt.
equally likely to be affected (or lack thereof) by
Evidence from an international report suggests
corruption in the police, across class categories, the
that citizens rated the police as the most corrupt
poor were most likely to be compelled to pay bribe.
Table 2.3: Receipt of complaint/ FIR across localities and gender

Those who received their FIR/ complaint Those who did not receive their FIR/ complaint
Rural 55 35
Urban 66 19
Men 61 30
Women 55 31
Figure 2.14: How many paid bribe
16%
34% Those who paid bribe

Those who did not pay bribe
No response
50%
Note: Question asked: (If during the last 4-5 years whenever you contacted a police officer or visited the police station, did you have to
pay bribe to get your work done?

Table 2.4: Poor Muslims and OBCs are most likely to have paid bribe on contacting the police
Those who paid bribe Those who did not pay bribe
Upper caste 34 51
OBC 37 47
Scheduled Castes 34 50
Scheduled Tribes 22 55
Muslims 38 48
Note: Figures are percentages. Rest of the respondents did not respond.
Similarly, Muslims, OBCs, socio- economically they were satisfied with the help provided by the
poor respondents were also most likely to have paid police. Less than a quarter (24%) stated that they
money to the police (Table 2.4). were very satisfied, a relatively bigger proportion of
41 percent were somewhat satisfied, nearly 1 in ten
2.2.2. Satisfaction with police help (9%) were somewhat dissatisfied and 14 percent were
The police in India, based on people’s experiences fully dissatisfied (Figure 2.15). Men and women did
and their portrayal in media etc., have a reputation not vary significantly in their satisfaction levels. By
for being non-responsive, unapproachable, corrupt locality, rural respondents were a little more likely to
and biased. Such popular perception severely affects be satisfied (Figure 2.16).
people’s satisfaction with the rule of law and results
Table 2.5 shows a ranking of surveyed states based
in a trust deficit. Victim satisfaction with the police
on the respondents’ level of satisfaction with the
is an important measure of police performance. To
police help that was provided. After assigning
gauge this, respondents who reported contact with
individual weights to each answer category and
the police in the last 4-5 years were asked whether
Figure 2.15: A large proportion of persons were moderately satisfied with police’s help
Note: Question asked: How satisfied or dissatisfied are you with the help provided by the police?
Figure 2.16: Satisfaction with police help across localities

42
38
26
21 21
11 11
8
Very satisfied Somewhat satisfied Somewhat dissatisfied Fully dissatisfied
Rural Urban
Note: Figures are percentages. Rest of the respondents did not respond to the question

Table 2.5: State-wise ranking of responses about satisfaction with police help after having contacted it
Rank State Very satisfied Somewhat Somewhat Fully Score N
satisfied dissatisfied dissatisfied
1 Jharkhand 32.4 59.0 2.9 1.9 11.7 104
2 Kerala 50.8 21.4 7.9 13.5 8.8 126
3 Uttarakhand 20.0 61.8 3.6 7.3 8.4 55
4 Telangana 30.9 43.2 8.6 11.1 7.4 81
5 Gujarat 30.0 45.0 5.0 15.0 7.0 60
6 Rajasthan 11.8 60.3 2.9 5.9 6.9 68
7 Odisha 22.1 49.4 11.7 9.1 6.4 78
8 Maharashtra 23.1 38.1 7.5 12.9 5.1 146
9 Madhya Pradesh 19.8 48.9 13.0 13.7 4.8 131
10 Tamil Nadu 9.9 59.3 7.7 13.2 4.5 91
11 Bihar 15.0 48.4 11.0 17.6 3.2 273
12 West Bengal 13.0 42.0 15.9 13.0 2.6 69
13 Uttar Pradesh 9.4 40.6 15.2 12.3 2.0 139
14 Karnataka 19.8 31.9 8.4 24.9 1.3 273
15 Delhi 14.8 32.1 12.3 34.6 -2.0 81
Note: The state rankings for Satisfaction with Police Help are based on summated scores that were arrived after weighing each Index
category. The category of satisfaction includes very satisfied and somewhat satisfied and dissatisfaction includes very dissatisfied and
somewhat dissatisfied. The ‘very satisfied’ category was weighed as 0.2, the ‘somewhat satisfied’ category was weighed as 0.1, the
‘somewhat dissatisfied’ category was weighed as -0.1, the ‘fully dissatisfied’ category was weighed as -0.2. A higher summated score
here indicates positive assessment, i.e. greater satisfaction. States where contact with the police was low and hence sample size was
low (<50) have been excluded from the analysis.
States where contact was low have not been reported.
arriving at a summated score, maximum satisfaction Among those who interacted with the police in the
with police help was found to be in Jharkhand while recent past, the reasons for dissatisfaction are not
the lowest satisfaction was in Delhi followed by very clear due to the high proportion of respondents
Karnataka. Kerala, Uttarakhand and Telangana also who did not provide reason(s) for their dissatisfaction
ranked high in terms of satisfaction. States where (Figure 2.17). One-fifth respondents stated that
the contact with the police low and hence sample they were dissatisfied because the police refused to
size was low have been excluded from the rankings. help. About 12 percent reported that the police was
Figure 2.17: Reasons for dissatisfaction
Note: Sample size:- 492.Answer choices with * have a sample size of less than 50 cases, therefore read them with caution.
Question asked: (If dissatisfied with help provided by the police) What was the main reason for your dissatisfaction?

abusive in behaviour and 13 percent complained of Men and women, different castes and religious
corruption (bribe). communities across rural and urban areas did not
vary significantly in their response on future police
2.2.3 Future police contact contact. Satisfaction could be seen to have an evident
A large plurality of respondents displayed favourable bearing on future police contact- those who were
inclination regarding future police contact - 72 dissatisfied with police performance in their locality
percent admitted that they would be open to seeking were more likely to avoid contacting the police than
police’s help when required. Among those who were those who are satisfied. Similarly, prior positive
not fully certain and as confident, 14 percent stated police contact was associated with greater optimism
that they would probably visit the police station and openness towards future police contact- those
and 4 percent said they would have to do so due who were satisfied with the help they received at the
to lack of other options. The implication of police police station were much more likely to seek their
perception on this likelihood of contact in the future help in the future than those who were dissatisfied
is an important finding that grants us a window into with police help. However, it is worth noting that
how people perceive the police. despite negative police contact, nearly three in five
Figure 2.18: Almost three-fourth respondents said that they will contact the police if the need arises1
4%
6%
14% Yes
Have no other option
4% Probably
No
Don’t know
72%
Note: Question asked: In the future, if you have a problem that requires police help, would you go to the police?
Figure 2.19: Satisfaction with the police likely to encourage police contact if needed
76 77
62
57
31
27
16 16
11
8 6
5
Those who are Those who are Those who are Those who are
satisfied with police’s dissatisfied with police’s satisfied with dissatisfied with
performance in their performance in their police’s help police’s help
locality locality
Seek police’s help Reluctant to seek police’s help Not seek police’s help
Note: Figures are percentages. The rest of the respondents did not respond.

persons are nonetheless willing to contact the police individuals, their fear and interpretation of events on
again in the future. the ground. Overall, nearly one third (32%) of those
who were surveyed expressed that crime does not
2.3. Incidence of crime occur in their locality whereas about 3 in ten (29%)
perceived the occurrence of crime as ‘sometimes’
Perception of incidence of crime is different from
and 9 percent said that crime occurs ‘very often’
crime statistics because it is the subjective opinion of
Figure 2.20: Incidence of crime by locality
40
36 36
32
29 29
25 23 24 22
19 19
15
9 8 11 7 7
5 5
Very often Rarely Sometimes Never No response
All Village Town City

Question asked: How often do incidents such as burglary, murder, physical assault, chain snatching occur in your locality?
Table 2.6: State-wise ranking of responses on incidence of crime

Rank States Very often Sometimes Rarely Never Score
1 Kerala 3.6 14.5 26.1 52.9 11.0
2 Assam 0.4 19.1 39.7 38.6 9.7
3 Odisha 4.3 23.2 12.7 55.8 9.3
4 West Bengal 3.1 19.1 35.1 38.5 8.7
5 Nagaland 2.2 16.6 36.8 33.7 8.3
6 Uttarakhand 5.9 23.0 16.7 49.2 8.0
7 Andhra Pradesh 10.2 12.3 34.8 36.4 7.5
8 Himachal Pradesh 4.8 28.3 16.6 47.6 7.4
9 Telangana 9.0 16.5 15.4 42.6 6.6
10 Gujarat 6.5 26.5 21.9 39.8 6.2
11 Chhattisgarh 1.3 27.3 16.4 37.2 6.1
12 Tamil Nadu 9.0 26.9 30.7 31.8 4.9
13 Punjab 5.5 31.7 32.1 22.3 3.4
14 Bihar 5.4 31.9 39.6 18.0 3.3
15 Maharashtra 6.8 39.1 22.7 27.9 2.6
16 Haryana 23.6 27.6 7.3 39.3 1.1
17 Madhya Pradesh 9.1 47.7 19.6 14.7 -1.7
18 Karnataka 18.0 42.9 14.9 20.7 -2.3
19 Rajasthan 15.7 35.9 14.8 12.7 -2.7
20 Uttar Pradesh 18.4 48.0 11.7 16.2 -4.1
21 Delhi 27.1 37.1 16.7 14.5 -4.6
22 Jharkhand 13.2 59.0 20.2 5.6 -5.4
Note: The state rankings for Incidence of crime are based on summated scores that were arrived after weighing each Index category.
The category of Crime Occurs includes incidence of crime as very often and sometimes and the category of Crime doesn’t Occur includes
incidence of crime as rarely and never. The ‘very often’ category was weighed as -0.2, the ‘sometimes’ category was weighed as -0.1,
the ‘rarely’ category was weighed as 0.1, the ‘never’ category was weighed as 0.2. A higher summated score here indicates positive
assessment, i.e. less incidence of crime.

(Figure 2.20). Respondents in cities were nearly perceived by the people is lowest there, while
twice more likely to report occurrence of crime in Jharkhand comes last, signalling that incidence
their neighbourhood/ area as ‘very often’ compared of crime as perceived by the people is greatest
to those residing in villages. there. Starting from Madhya Pradesh and up till
Jharkhand, the incidence of crime is greater whereas
Based on the state-wise responses to the question
in all the states above Madhya Pradesh, occurrence
on frequency of crime in one’s locality, a unique
of crime is less.
score for each state was derived (See Table 2.6).
This score indicates the overall degree of incidence An important aspect of understanding citizens’
of crime in a single figure. Arranging the states in sense of security entails understanding their
descending order of their scores—highest score first, perception of crime and security in their locality.
representing least amount of incidence of crime, we Fewer people expressed that crime had increased
arrived at the following distribution, which displays (17%) in their locality compared to 37 percent of
the relative position of each state regarding this those who considered a reduction in the incidence
question. We see that Kerala fares best among all of crime (Figure 2.21). A little over one-third (34%)
the states, meaning that the occurrence of murder, stated that there was no change in the occurrence of
physical assault, burglary and related crimes as crime. As the locality increases in size and urbanity,
Figure 2.21: Over one-third believe that crime in their locality has reduced
12% 17%
Crime had increased in my locality

Crime had decreased in my locality
34% Crime level has remained the same in my locality
No response
37%
Note: Question asked: In the last 2-3 years, has crime in your locality increased, decreased or remained the same?
Figure 2.22: Change in the incidence of crime across localities
36 40
38 37 33
30
27
16 13
Village Town City
Crime has increased Crime has decreased Crime has remained the same
Note: Figures are percentages. Rest of the respondents did not answer

the perception of increase in crime also grows i.e. investigation of crime is lowest here. Other states
respondents in cities were most likely to report where satisfaction was extremely low are Karnataka,
an increase in crime in their locality than those in Madhya Pradesh, Jharkhand, Delhi, Telangana etc.
towns (Figure 2.22).
People’s overall satisfaction with the police is
The opinion on police’s investigation of crime such shaped by their experience and perception. Positive
as murder, assault, robbery was very mixed. On perceptions promote engagement, compliance and
one hand, citizens felt that their investigation is trust in the police whereas negative perceptions
satisfactory and proper (37%). On the other hand, erode legitimacy and result in reduced contact with
about 29 percent expressed that the investigation is the police. The survey sought to assess this and it
not up to the mark and often faulty in nature, nearly was found that citizens’ satisfaction levels with
7 percent reported that police harasses people during the police’s performance and work in their area
investigation and 8 percent believed that police does are not concentrated in the extreme categories of
not carry out the needful inspection (Figure 2.23). fully satisfied and fully dissatisfied. Rather, their
Satisfaction with police’s investigation is likely to satisfaction levels are moderate in nature- a little
be highest in towns, among upper castes and those over a quarter were fully satisfied, while a large
who belong to upper class. Furthermore, experience proportion (52%) were somewhat satisfied. Only 5
of harassment by police during investigation is most percent said that they were fully dissatisfied with
likely to be reported by Hindu Scheduled Tribes, police performance in their locality (Figure 2.24).
Scheduled Castes and Muslims and those residing
Respondents were asked if they witnessed a police
in small cities.
officer violating the law and nearly one in five
Disaggregating the state responses (Table 2.7 persons responded with an assertion, while 67
reveals that states with a higher net score (this was percent did not report so (Figure 2.25). Men, urban
calculated by subtracting total dissatisfaction from dwellers (Figure 2.26) and those who had to pay
total satisfaction) indicate positive assessment that bribe to get their work done were more likely to have
is greater satisfaction with investigation of crime. In witnessed this.
contrast, states with a lower net score demonstrate
Among 21 percent who saw a police officer violating
dissatisfaction with police’s investigation of crime.
the law, only 13 percent filed a complaint and 81
We see that Himachal Pradesh fares best among all
percent did not. Respondents in rural areas, women
the states, meaning that satisfaction with police’s
and OBCs and those who had contacted the police in
investigation of crime is greatest here, while Bihar
the recent past and had to pay bribe were more likely
comes last, signalling that people’s satisfaction with
to have filed a complaint against a police officer.
Figure 2.23: Opinion on police’s investigation by locality

Question asked: In your opinion, is the police’s investigation of such incidents (of crime) proper and satisfactory or is it temporary
investigation?

Table 2.7: State-wise perception on police’s investigation of crime
Those who are satisfied with Those who are dissatisfied with Net satisfaction
police’s investigation of crime police’s investigation of crime
Andhra Pradesh 49 40 9
Assam 42 39 3
Bihar 16 80 -64
Gujarat 56 27 29
Haryana 59 38 21
Himachal Pradesh 73 22 51
Karnataka 17 58 -41
Kerala 42 22 20
Madhya Pradesh 26 56 -30
Maharashtra 41 46 -5
Nagaland 22 35 -13
Odisha 49 32 17
Punjab 46 25 21
Rajasthan 30 39 -9
Tamil Nadu 30 52 -22
Uttar Pradesh 32 61 -29
West Bengal 30 43 -13
Delhi 28 58 -30
Jharkhand 27 70 -43
Chhattisgarh 58 24 34
Uttarakhand 55 32 23
Telangana 20 53 -33
Overall 37 45 -8
Note: Figures are percentages. Rest of the respondents did not respond. Net satisfaction here means total satisfied minus total
dissatisfied.
Figure 2.24: Respondents’ overall satisfaction with police
6%
5%
26% Fully satisfied

11%
Somewhat satisfied
Somewhat dissatisfied
Fully dissatisfied
52% No response
Note: Question asked: How satisfied are you with the performance of police and their work in your locality?

2.4. Perception of safety unsafe in the morning and during the day (Table
2.8). In contrast, a high percentage of respondents
Public safety is a key component of a community’s (44%) indicated that they do not feel safe at night.
life. Feelings of safety do not always echo reality, and To get a more comprehensive sense of perception of
the fear of crime can be influenced by neighbours, safety at different intervals of the day, an Index was
friends, and media portrayal of crimes in the computed. This revealed that 34 per cent expressed
neighbourhood. Alongside, the presence of police feeling highly safe in their village/ neighbourhood
constables in individuals’ locality can significantly and 28 per cent stated feeling highly unsafe. While
enhance people’s perception of safety during an equal proportion of men and women feel unsafe
different points of the day. It can also positively at night, a greater proportion (51%) of respondents
shape people’s view that police handle their problems in urban areas stated feeling unsafe at night than
well when there are fewer crimes reported in their their rural counterparts (40%). Similar differences
neighbourhood. Assuming a positive association across rural (28%) and urban areas (34%) could be
between greater police presence (which is likely to seen in the perceived level of safety during the day.
deter criminal activity) and people’s perceived sense
of safety, respondents were asked whether or not they People’s perception of safety has a direct bearing on
feel safe in their village or neighbourhood at night, their satisfaction with police—those who feel highly
during the day and early morning. Nearly an equal safe are also most likely to be satisfied with police’s
proportion of one-third persons (31%) stated feeling work in their locality (Figure 2.27). The feeling
of safety seems to have a direct relationship with
Figure 2.25: Over one-fifth respondents have witnessed a police officer violating the law
12%
21%
Those who saw a police officer
violating the law
Those who did not see a police officer

violating the law
No response
67%
Note: Question asked: Have you ever seen a police officer violating the law?
Figure 2.26: Urban dwellers are more likely to have witnessed a police officer violating the law
69
63
27
17
Rural Urban
Those who saw a police officer Those who did not see a police officer
violating the law violating the law
Note: Figures are percentages and rest of the respondents did not respond.

citizens’ satisfaction with police services and their officers. To better understand this, the survey sought
attitude toward crime. This means that encouraging to ascertain people’s opinion and satisfaction with
a better policing system that focuses on reducing fear police presence in their locality. A large proportion
and ensuring safety in the neighbourhood overall can of 47 percent expressed the need for greater police
generate positive results- it can make communities presence while a relatively smaller proportion of
safer and invoke positive citizen satisfaction with 15 percent said that they would prefer lesser police
police and their services. presence (Figure 2.28). Nearly three out of ten
respondents were not likely to prefer any change. It
2.5. Opinion on police services must be noted here that the demand for greater police
presence can point towards both the existing levels of
Often, the presence of police officers and patrolling
trust in the rule of law and people’s perceived lack of
vans are thought to create feelings of safety. Visible
safety. Men and upper castes were much more likely
policing positively affects citizens’ perception of
to expect an increase in police presence than women
crime, safety and their confidence in the police.
and persons from lower castes. In contrast, caste
At the same time, not all communities are likely
communities such as Hindu OBCs and Scheduled
to echo this sentiment. Vulnerable groups such as
Tribes displayed preference for decreased presence
Scheduled Castes, Scheduled Tribes may associate
of the police (Table 2.9). As the locality increases in
greater police presence with the fear of being
size and urbanity, the preference for greater police
targeted and wrongfully implicated by uniformed
presence also increases.
Table 2.8: Sense of safety among people during different times of the day
Perception of safety early morning Perception of safety during the day Perception of safety at night
Those who feel very unsafe 13 Those who feel very unsafe 12 Those who feel very unsafe 23
Those who feel somewhat 18 Those who feel somewhat unsafe 18 Those who feel somewhat 21
unsafe unsafe
Those who feel not very unsafe 19 Those who feel not very unsafe 20 Those who feel not very unsafe 16
Those who feel not at all unsafe 47 Those who feel not at all unsafe 47 Those who feel not at all unsafe 36
Overall those who feel safe 66 Overall those who feel safe 66 Overall those who feel safe at 52
early morning during the day night
Overall those who feel unsafe 31 Overall those who feel unsafe 30 Overall those who feel unsafe 44
early morning during the day at night
Question asked: How unsafe do you feel in your village/neighbourhood during different times of the day—very, somewhat, not very
or not at all?
Figure 2.27: Feeling safe has bearing on levels of satisfaction with the police
82
78 76 75
69
19 18 19
11 13
Those who feel Those who feel Those who feel Those who feel Non-committal
highly safe somewhat safe unsafe highly unsafe
Those who are satisfied with police’s Those who are dissatisfied with police’s
work in their locality work in their locality
Note: Figures are percentages. For methodology on how Index of feeling safe was constructed, see Appendix.

On the question of whether police’s help was sought To address the safety issues faced by women, police
by calling them on the number 100, only 8 percent introduced a special helpline number which would
replied in agreement whereas a large proportion of provide round-the-clock access to safety services.
82 percent did not (Figure 2.29). This is an alarming To ascertain its popularity and usage, women
finding because the 100-number call service has respondents were asked whether in the last 2-3
been operational for over two decades, yet its usage years they used the special helpline number (Figure
remains abysmally low. Men and those residing 2.31). Only three percent had previously called on
in urban areas were more likely to have called on the special women’s helpline number. Among those
the 100 number than their respective counterparts. who had previously used the helpline number, about
Respondents were asked if access to 100-number 65 percent reported positive experience and received
service had improved in the last two-three years the required help whereas 13 percent did not.
and almost a quarter (23%) agreed. Over one-third
To coordinate and manage the safety and security of
(34%) expressed that it had improved ‘somewhat’,
elderly, the police introduced senior citizen helpline
and about 12 percent opined that there was no
in various states of India. However, awareness of
improvement (Figure 2.30).
Figure 2.28: A large plurality of citizens want greater police presence in their locality
Those who want greater police

9%
presence in their locality
Those who want less police
presence in their locality
Those who are satisfied with the
29% 47%
existing
police presence in their locality
No response
15%
Note: Question asked: What kind of police presence would you like to see in your locality- greater, less, same as before?
Table 2.9: Upper castes most vocal about greater police presence
Those who want greater police Those who want less police Those who are satisfied with
presence in their locality presence in their locality the existing police presence in
their locality
Upper castes 54 9 20
OBCs 47 18 27
Scheduled Castes 43 15 32
Scheduled Tribes 41 17 28
Muslims 43 14 32
Men 50 14 29
Women 44 15 29
Rural 45 15 30
Urban 53 13 27
Note: Figures are percentages. Rest of the respondents did not answer.

Figure 2.29: Respondents who used the 100-number service
8%
10%
Those who called police on 100 number

Those who did not call police on 100 number
No response
82%
Note: Question asked: In the last 2-3 years, have you tried calling police (100 number) on phone in case of an emergency?
Figure 2.30: Improvement in the access of ‘100’ number
Those who believed that access to ‘100’ number

23% has improved a lot
31%
has improved somewhat
has improved not much
5%
has improved not at all
7% 34% No response
Note: Question asked: In your experience, to what extent has access to 100 number improved in the last 2-3 years—a lot, somewhat,
not much or not at all?
Figure 2.31: Seven out of ten women did not call on the helpline
3%
27%
Women who called on the special
helpline number
Women who did not call on the
special helpline number
No response
70%

Figure 2.32: Awareness on police related services
48
21
15 16
Those who are Those who are Those who are Those who are
aware of the aware of the aware of the aware of the
introduction of introduction of introduction of introduction of
all women PCR patrolling senior citizen child helpline
police stations vans helpline number
Note: Figures are percentages. Figures of respondents who were not aware or did not respond have not been reported.
Question asked: Please tell me whether the following services have been introduced in your area or not.
the same remains low and only 13 percent reported trust deficit needs to be addressed by enhancing
that they were aware of this provision (Figure the relationship through sustained communication
2.32). People were most likely to be aware of the and non emergency-based interaction. Such
introduction of PCR patrolling vans with over two- measures would also positively affect the sense of
fifths stating so. A little less than one-fifth (19%) safety and reduced fear of crime among people.
knew that all women police stations had been Regarding corruption and subsequent satisfaction
introduced and about 14 percent were aware of the with the police, the survey found that over one third
introduction of the child helpline number. respondents were compelled to pay bribe in return
for police services, about 23 percent were dissatisfied
2.6. Conclusion: and satisfaction was quite high (65%). The gendered
aspect of police-public interaction revealed that
This chapter highlighted different aspects of
women were much less likely to contact the police,
police-public interaction, people’s perception
more likely to be accompanied by a family member
and experience of crime, corruption, usage of
and less likely to visit the police station alone. For
services such as special help lines for women etc. It
crimes that are experienced predominantly by
illustrated that people’s experiences, perceptions and
women, the police are the first link in the chain of
opinions range from polarity to uniformity across
access to justice. Therefore, not only do women
geographies, gender, caste and class communities.
need to be actively encouraged to report crime to
An important finding of the survey has been the low
the police, issues relating to responsiveness and
incidence of police contact (14%) i.e. people who
discriminatory attitudes of justice service providers
contacted the police or vice versa in recent past.
need to be addressed on priority basis.
There are two possibilities here. First, police contact
might have been affected by stereotypes of mistrust, References:
fear and enmity between police and citizens thereby
deterring latter from approaching them for reporting Raghavan, R.K. 2003. The Indian Police: Problems and
crime or for seeking any other help. Second, it is also Prospects. In Publius: The Journal of Federalism, 33(4)
likely that contrary to our intuition, situations that Verma, A. 1997. Maintaining Law and Order in India:
require assistance or intervention of police are far An Exercise in Discretion. International Criminal Justice
and few, thereby resulting in minimal police-citizen Review, 7: 65-80
interaction. The first possibility indicates that the

CHAPTER 3:
TRUST IN POLICE
A police constable helping a blind couple at Sion, Mumbai (Credits: Kunal Patil, Hindustan
Times, 20 March 2017, Mumbai)

Trust in Police
3.1. Introduction of trust with regard to how one’s perceptions and

experiences guide the overall trust levels.
This chapter focuses on the levels of trust among
people with regard to the institution of police. 3.2. Trust in police: measured directly
Every public institution, by virtue of their role, is
always subject to public scrutiny. Their functioning, Relative to all public institutions, it was noted in
manner of dealing with issues of public concern and the Common Cause-CSDS Survey that people have
attitudes towards the general public guides people’s high levels of trust (a lot of trust) in the army (54%),
perception, which in turn influences people’s trust. followed by the judiciary/courts (31%; see Figure
3.1). The assessment of the police was however not
The trustworthiness of the police is often linked as positive. Only three in every ten people said that
to their performance as well as a close interplay they had significant levels of trust in senior police
of societal factors, which in turn guides public officers and two in every ten in the local police.
assessments of the ability of the police to be a ‘civic Lowest levels of trust were reported for traffic police
guardian’ (Loader and Mulcahy,2003). A number (16%).The police only fared better when compared to
of studies show that policing practices which focus government officers, high trust in whom was found
on respectful treatment and proactive attitude is to be 18 percent. That being said, the image of the
strongly linked to higher levels of trust. A close police seems to have improved in the last few years
linkage exists between perception of police and even as trust in other institutions has declined, albeit
their legitimacy in terms of their role - many widely marginally. When we compare the current trust data
used policing practices and approach are often seen with past data on trust collected by CSDS, we find
as unfair which in turn have compromised the trust a significant improvement in high trust levels with
that people place in police. The image of the police respect to the police and a slight dip in high trust
in India, coupled with the problems and limitations levels with the military, judiciary and government
that this institution faces today, warrants the needs officers. In a national survey conducted by CSDS
to study the trust people place in this institution. in 2013 as part of the larger State of Democracy
In the context of the everyday discourse in India in South Asia Study, high trust in the police as a
with regard to police, we hypothesized that a number whole had been found to be only 16 percent. This is
of social demographic factors would affect the trust around seven points lower than the simple average
people place in the police. Furthermore, we tried of the high trust figures recorded for senior, local
to explore whether experiential factors affect trust and traffic police officers in current survey. On the
and satisfaction of an individual with the police other hand, in the case of the military, judiciary and
functioning. The survey showed interesting trends government officers, high trust of people in them
in this regard. This chapter explores the dimension has declined by about 2-3 points in the last 3-4 years.
Figure 3.1: Trust in police vis-à-vis other institutions

54
48
40 39
34 31 35
29
23 23 22
20
15 14 16 14 16 18
11
7 8 8
5 6
Trust in Trust in Trust in Trust in Trust in Trust in

senior officer local police traffic police army courts government
officer officers
A lot of trust Somewhat trust Little trust None at all
Question asked: Please tell me how much trust do you have in each of the institutions- a lot, somewhat, not much or not at all?

On analyzing trust in different hierarchies of police traffic police state-wise since its presence is largely
by states we found that trust in local police was restricted to bigger cities.
greatest in Andhra Pradesh, Jharkhand, Telangana
State-wise variation was noted in this regard: If
and Kerala (in that order; Table 3.1). On the other
Haryana, Himachal Pradesh, Odisha and Kerala
hand, Rajasthan reported the lowest levels of trust,
were the states where people had high trust in
followed by Uttar Pradesh, Bihar and Nagaland. For
senior police, the top three states in terms of trust
senior police, states such as Haryana (highest trust),
in local police were found to be Andhra Pradesh,
Himachal Pradesh, Odisha and Kerala were ranked
Jharkhand, Telangana and Kerala. Looking at states
high, which indicates that people have significant
which performed poorly on this metric for both
levels of trust in the senior police officers (Table
senior police and local police, we found that both
3.2). Here again, Rajasthan ranked the lowest. Uttar
Rajasthan and Uttar Pradesh appear at the bottom
Pradesh, Punjab and Gujarat also performed poorly
of the rankings.
on this count. We haven’t analyzed trust levels in the
Table 3.1: Trust in local police by state

Rank States A lot Somewhat Not much Not at all Score
1 Andhra Pradesh 35.9 42.1 5.7 3.6 10.1
2 Jharkhand 26.3 58.3 9.2 0.8 10.0
3 Telangana 26.4 55.9 5.4 1.8 10.0
4 Kerala 43.1 32.6 11.3 5.1 9.7
5 Odisha 33.1 40.3 6.7 4.2 9.1
7 Uttarakhand 23.7 57.6 10.0 6.2 8.3
8 Tamil Nadu 30.6 45.1 11.2 6.6 8.2
9 Madhya Pradesh 25.3 53.4 15.1 4.0 8.1
10 Karnataka 26.1 49.7 12.9 5.2 7.9
11 Haryana 11.3 72.0 14.0 2.1 7.6
12 Chhattisgarh 22.6 46.0 11.8 3.8 7.2
13 Assam 18.5 56.1 14.4 4.7 6.9
14 Maharashtra 20.5 49.5 11.9 8.6 6.1
15 Gujarat 22.8 47.0 16.2 7.8 6.1
16 Punjab 25.1 37.4 19.1 8.4 5.2
17 Delhi 13.9 50.5 16.7 9.7 4.2
18 West Bengal 18.1 39.9 14.1 10.2 4.2
19 Nagaland 16.8 47.0 23.7 9.5 3.8
20 Bihar 16.3 44.0 27.4 10.9 2.7
21 Uttar Pradesh 9.4 48.4 19.4 15.7 1.6
22 Rajasthan 6.2 43.5 27.7 6.8 1.5
Note: The state rankings for the question- Please tell me how much trust do you have in local police like police inspector, Sub inspector, SHO - a
lot, somewhat, not much, or not at all?- are based on summated scores that were arrived at after weighting each response option. An ‘a
lot’ answer was weighted as 0.2, a ‘somewhat’ answer was weighted as 0.1, a ‘not much’ answer was weighted as -0.1, and a ‘not at
all’ answer was weighted as -0.2. The category of no response (those who did not answer the question) was weighted as 0 and hence
excluded from the ranking analysis. A higher summated score here indicates a more positive assessment.

Table 3.2: Trust in senior police by state
Rank States A lot Somewhat Not much Not at all Score
1 Haryana 31.0 60.3 6.9 1.0 11.3
3 Odisha 37.9 33.4 6.7 2.5 9.8
4 Kerala 43.9 31.3 9.0 6.5 9.7
5 Bihar 47.5 29.2 15.3 6.8 9.5
6 Uttarakhand 35.1 44.1 15.5 2.6 9.4
7 Andhra Pradesh 28.8 42.1 11.1 2.4 8.4
8 Jharkhand 46.4 19.6 26.6 1.4 8.3
9 Telangana 28.9 40.6 13.6 2.1 8.1
10 Assam 23.2 50.6 13.1 2.1 8.0
11 Madhya Pradesh 32.7 38.5 19.7 3.5 7.7
12 Delhi 28.2 41.6 10.7 5.6 7.6
13 Nagaland 20.4 54.3 19.1 2.9 7.0
14 Chhattisgarh 30.1 33.7 15.2 4.6 7.0
15 Tamil Nadu 27.4 40.5 17.5 7.5 6.3
16 Karnataka 24.6 43.1 19.6 5.8 6.1
17 West Bengal 24.1 36.9 11.3 6.7 6.0
18 Maharashtra 20.5 44.3 16.4 5.5 5.8
19 Gujarat 24.6 41.7 18.3 7.7 5.7
20 Punjab 26.2 29.5 24.6 7.8 4.2
21 Uttar Pradesh 19.6 40.7 20.1 11.2 3.7
22 Rajasthan 13.0 37.7 20.8 11.9 1.9
Note: The state rankings for the question - Please tell me how much trust do you have in a senior police officer like SP, DCP - a lot, somewhat,
not much, or not at all? - are based on summated scores that were arrived at after weighting each response option. An ‘a lot’ answer was
weighted as 0.2, a ‘somewhat’ answer was weighted as 0.1, a ‘not much’ answer was weighted as -0.1, and a ‘not at all’ answer was
weighted as 0.2. The category of no response (those who did not answer the question) was weighted as 0 and hence excluded from the
ranking analysis. A higher summated score here indicates a more positive assessment.
3.3. Trust in police: measuring understand whether the gender of the child could
play a potential role in influencing such decisions,
indirectly we tried to analyze this response with the sex of
Apart from the direct questions related to trust, the the respondents’ children. It was noticed that there
survey also asked a question that helped gauge trust wasn’t any significant difference in opinion between
levels indirectly. Respondents were asked whether all those who had a daughter (53%) or a son (54%).
they would allow their child to visit the police station This is to say, that irrespective of the gender, people
alone to file a complaint in the event that they were show hesitance, which in turn tells us a lot about their
a victim of any crime. Over half (54%) were of the level of trust in the functioning of that institution.
opinion that they would not let their children visit A state-wise disaggregation of the responses found
the police station alone (Figure 3.2). This helped respondents in West Bengal, Assam and Jharkhand
us to make an inference that largely people were of to be more willing about sending their child to the
the opinion that it is usually not safe to approach police station alone, if the need arose. Respondents
this authority without any support. Furthermore, it in Madhya Pradesh, Telangana, Rajasthan and
was interesting to note that women (55%) were only Uttar Pradesh were found to be the most reluctant
slightly more likely than men (53%) to be hesitant (Table 3.3).
in letting their wards approach the police alone. To

Figure 3.2: Majority unlikely to let their wards visit police station alone
10%
Will let my child visit the

36% police station alone
Won’t let child visit the
police station alone
No response
54%
Note: Question asked: If your daughter or son was to be a victim of any crime, would you allow her/him to visit the police station
alone to file a complaint?
Table 3.3: Respondents in West Bengal, Assam and Jharkhand most willing to send their child to the
police station if need arises
Rank States Will let my child visit the police station Won’t let child visit the police station alone
alone
1 West Bengal 87 3
2 Assam 83 6
3 Jharkhand 73 25
4 Himachal Pradesh 47 24
5 Nagaland 46 39
6 Kerala 40 53
7 Tamil Nadu 40 55
8 Haryana 40 59
9 Odisha 38 59
10 Punjab 36 43
11 Bihar 35 57
12 Karnataka 33 64
13 Uttarakhand 31 59
14 Chhattisgarh 30 50
15 Maharashtra 30 59
16 Delhi 27 65
17 Andhra Pradesh 25 64
18 Gujarat 24 62
19 Madhya Pradesh 19 74
20 Uttar Pradesh 17 67
21 Rajasthan 16 67
22 Telangana 12 70
Question asked was: If your daughter/son was to be the victim of any crime, would you allow her/him to visit the police station alone
to file a complaint? The state rankings are based on the ‘yes’ responses. In states where the ‘yes’ responses are of similar proportion, the
no answer has been taken into consideration.

3.4. Overall trust in police see Appendix 3), we found, one in every four (24%)
to be trusting the police highly, more than two in
In order to understand the possible determinants of every five (45%) to be trusting it somewhat, nearly
trust in the institution of police as a whole we created one in every four (23%) to be somewhat distrustful
a trust in police Index. To construct this Index, the of it and less than one in every ten (7%) to be very
two questions directly related to trust were taken distrustful of it (Figure 3.3). This trust in police
into consideration - one that probed people’s level Index helped us to understand the interplay of a
of trust in the local police like police inspector, sub- number of factors which affect levels of trust across
inspector, SHO and second that probed their trust different groups and communities. We report these
in senior police officers like SP, DCP. Once again findings in the following sub-sections.
the question that probed trust levels in the traffic
police was excluded from analysis since it is largely 3.4.1. Socio-economic cleavages
an urban phenomenon. The Index also took into Demographic factors have a significant bearing and
account a third question, the indirect one that asked guide an individual’s interaction and perception of
respondents whether they would allow their child to the police, which in turn influences their level of
visit the police station alone to file a complaint. On trust.
constructing the Index (for a detailed methodology,
Figure 3.3: Overall trust levels in the police based on the trust Index
2%
7%
24%
Highly trust
Somewhat trust
23%
Somewhat distrust
Highly distrust
Non-committal
45%
Note: For methodology on how the Index was constructed, see Appendix 3.
Figure 3.4: Distrust is inversely proportional to class hierarchy
45 46 44 45
27
24 24 23 22 23 23
21
8 9
7
4
1 1 1 2
Highly trust Somewhat Somewhat Highly distrust Non-committal

trust distrust
Upper Middle Lower Poor
Note: For methodology on how the Index was constructed, see Appendix 3.

In the survey, it was found that the poor and the The caste angle on its own also plays an important
lower classes had the highest levels of distrust in role in shaping public trust in police. The Scheduled
police (32%, and 31% respectively; see Figure 3.4). Tribes (STs hereafter) were found to be most
The upper class had the highest levels of trust, distrustful of the police (37%; highly and somewhat
and this was true across all genders. Often class combined), followed by Other Backward Classes
hierarchies tend to impact one’s vulnerability. As (30%) (OBCs hereafter) and Scheduled Castes (SCs
one’s vulnerability increases, their trust levels tend hereafter) (29%; see Figure 3.5).
to decrease (Offe and Patterson, quoted in Warren,
Even within these social groups, regional variation
1999). Such a finding corresponds even with regard
was evident. Among STs, it was those residing in
to other institutions we took into consideration in
Rajasthan that stood out with three-fourths of them
the survey.
(78%) being highly or somewhat distrustful of the
Interestingly, the inverse relationship between police. STs in Gujarat also reported high distrust at
class hierarchy and trust levels coincided sharply 54 percent. However, in the other tribal concentrated
even within the caste-groups and across different states such as Chhattisgarh, Jharkhand, Odisha,
religions. On the whole, the upper class, even within Assam and Maharashtra the level of distrust was not
the various caste-based distinctions weaved together, as high (30%, 19%, 28%, 20% and 16%, respectively).
exhibited highest levels of trust.
Figure 3.5: STs distrust the police more than any other caste groups
48
45 44 44 45
39
36 34
32
29
24 24 24 25 26
21 22 21 22
18 19
8 8 8 7
6 5 6
1 1 3 1 1 1 2
Upper OBC SC ST Muslim Christian Sikh

caste
Highly trust Somewhat trust Somewhat Distrust Highly distrust Non-committal
Figure 3.6: Lower literacy levels correspond to high levels of distrust

46 47
43 44
26 26 26 26
22 21 22
16
13
7 6
4 3
1 1 1
Highly trust Somewhat trust Somewhat distrust Highly distrust Non-committal
Non Literate Upto Primary Upto Matric College and above

As far as SCs are concerned, the highest levels of of distrust was 29 percent. This underlined the need
distrust were recorded among SCs of Punjab (59%), to explore whether one’s gender plays a role as far
Rajasthan (58%), Delhi (54%), Uttar Pradesh (44%), as their interactions are concerned and how such
Bihar (37%), Maharashtra (34%) and Telangana interactions determine the trust levels.
(32%).
On the whole too, women were seen to be more
With respect to religion, the highest ‘highly trust’ distrustful of the police compared to men but not by
figure was recorded among Christian respondents a very wide margin (32% as opposed to 28%, when
(32%). Sikhs (mainly situated in Punjab in our we combine highly distrust and somewhat distrust;
sample) were least likely to highly trust the police see Figure 3.8).
(19%). Muslims were found to be nearly as trusting
The survey showed a consistent trend that the older
or distrusting as the Hindus (Table 3.5).
population had higher levels of distrust in the police,
It was also found that educational levels have a elderly women particularly. While 30 percent of
significant impact on the trust levels of the public. men aged above 55 years reported high distrust of
A clear trend was evident that suggested that people the police, among women from the same age group,
with higher levels of education tend to have greater 37 percent had high distrust. This gender divide was
trust in the police. While high distrust among not as sharp among the middle aged respondents
non-literates was found to be 13 percent, among (those aged between 35-55 years) and it was non-
those educated up to primary and matriculation, existent among the young respondents, that is, those
it dropped to 7 percent and 6 percent, respectively. aged between 18-35 years reported similar levels of
Graduates were least distrustful of the police with distrust of the police.
only 4 percent of them reporting high distrust
Across nearly all states men and women were
(Figure 3.6). This finding could be explained in the
similar in their levels of trust and distrust. That is
context of the police’s attitude towards people with
to say, if men were found to be most distrustful of
lower educational levels and the manner in which
the police in certain states, so were the women of
they address their issues.
that state. There was a very little gap between the
The survey found that across all occupations, two. However if there is one state that stood out it
semi-skilled workers, agricultural workers and was Uttar Pradesh, where the gap was quite big—
non-workers had the highest level of distrust as far while 41 percent of women in Uttar Pradesh were
as police is concerned (Figure 3.7). This could be distrustful of the police, among the men of Uttar
explained in the light of the treatment meted out to Pradesh it was ten points higher at 51 percent.
them by the police. Among the semi-skilled workers
The trends that were noted in the previous sub-
and agricultural workers, women were found to
section with respect to caste and class based
be expressing significantly higher levels of distrust
distinctions and their corresponding linkages with
than men. For instance, 38 percent of female semi-
the levels of trust, were also evident even when we
skilled workers said they were distrustful of the
look at it with respect to gender. This is to say, that in
police as opposed to 31 percent male semi-skilled
the survey, just as STs were found to have the highest
workers. Similarly, while 34 percent of female
levels of distrust with the police, the same pattern
agricultural laborers reported being distrustful of
could be seen even when in terms of gender: ST
the police, among their male counterparts the figure
Figure 3.7: Semi-skilled and agricultural workers more distrustful of the police
48 46
45 45 45 45
39
31
27 28 26
23 25 23 25
20 21 22 20 22
19
7 8 9 7
4 6 6
0 1 1 2 2 1
Professionals Managerial/ Services Skilled Semi-skilled Agricultural Non-workers

Administrative workers workers workers
Highly trust Somewhat trust Somewhat distrust Highly distrust Non-committal


Figure 3.8: Women more distrustful of police
46
44
25
22 22 23
9
6
1 2

trust distrust
Men Women
Table 3.4: Ranking the states in terms of trust in police

Rank States Highly trust Somewhat trust Somewhat Highly distrust Score
distrust
1 Assam 57.0 25.9 14.8 1.7 12.2
2 Jharkhand 44.1 44.5 8.0 3.4 11.8
3 Haryana 34.6 54.5 9.9 1.0 11.2
5 West Bengal 47.2 25.0 22.4 2.6 9.2
6 Uttarakhand 22.2 57.4 16.7 3.2 7.9
7 Kerala 29.9 47.4 14.1 7.4 7.8
8 Tamil Nadu 29.2 43.1 17.6 9.8 6.4
9 Odisha 26.3 46.5 15.2 10.6 6.3
10 Karnataka 17.1 57.3 20.9 4.6 6.1
11 Bihar 26.6 43.4 23.8 6.0 6.1
12 Nagaland 30.7 31.5 34.7 2.9 5.2
13 Andhra Pradesh 19.7 49.5 17.6 12.0 4.7
14 Madhya Pradesh 12.9 55.8 27.4 3.5 4.7
15 Maharashtra 14.6 51.5 23.6 8.8 4.0
16 Chhattisgarh 17.7 43.8 25.3 7.2 4.0
17 Gujarat 16.5 47.3 24.9 9.8 3.6
18 Delhi 19.1 44.9 23.3 12.1 3.6
19 Telangana 7.2 58.6 24.6 6.5 3.5
20 Punjab 17.4 31.9 41.3 8.2 0.9
21 Uttar Pradesh 8.3 42.5 35.7 10.7 0.2
22 Rajasthan 9.9 30.8 38.4 17.0 -2.2
Note: The state rankings for the Index of trust in police (see Appendix 3 for details on how the Index was constructed) are based on
summated scores that were arrived at after weighting each Index category. The ‘highly distrust’ category was weighted as -0.2, the
‘somewhat distrust’ category was weighted as -0.1, the ‘somewhat trust’ category was weighted as 0.1, and the ‘highly trust’ category
was weighted as 0.2. The category of non-committal (those who did not answer any question that went into making the Index) was
weighted as 0 and hence excluded from the ranking analysis. A higher summated score here indicates a greater trust.

women (40%) were also found to have the highest overall trust in it. On analyzing trust levels with the
distrust in the police, which was followed by SC respondents’ level of satisfaction with police work
women (31%). in their area, we noted highest levels of trust in
police (31%) among all those who felt that they were
Probing the geographical angle by looking at the
‘fully satisfied’ with how the police in their area
type of locality (rural or urban) that the respondent
was functioning. Conversely, high trust levels were
resides in, we found small towns to be reporting
nearly three times less at 11 percent among those
the lowest levels of distrust in police (26%). The
who were fully dissatisfied with the performance of
two extreme ends – villages and cities—showed
the police in their area (Table 3.5). Similarly, there
the highest levels of distrust at 31 percent and 30
is also a correlation the other way around—overall
percent respectively. A significant gender divide
trust in the police seems to affect satisfaction with its
could be witnessed in these two areas. While 29
functioning in one’s area (Table 3.6).
percent of male respondents in villages were found
to be distrustful of the police, among female village Even as trust and satisfaction seem to be correlated
respondents the same figure was four points higher when seen in overall terms, people seem to
at 33 percent. In cities, this gap between men and understand the two concepts quite differently in
women when it came to distrusting the police was some of the states. We say this because states that
of six points—27 percent to 33 percent. reported highest trust levels with the police on our
scoring scale are not the states that also reported
3.5. Trust levels by states the highest satisfaction levels. If Assam, Jharkhand
and Haryana were the top three states in terms of
This variation in the levels of trust with the police
trust in police, in terms of satisfaction with police
followed a remarkable spatial pattern as well. After
functioning in one’s area, the top three states on
assigning specific weights to each category of the
a similar scoring scale were found to be Kerala,
trust Index and then summing up the score, we
Himachal Pradesh and Odisha. Bihar ranked the
find Assam ranked first as far as people’s trust in
lowest, followed closely by Uttar Pradesh and
the police is concerned. Jharkhand, Haryana and
Andhra Pradesh on this metric (Table 3.7). This
Himachal Pradesh were the other states which
hints at the fact that while trust and satisfaction
ranked high on this metric, while Rajasthan ranked
are seen to be closely related, people also tend to
the lowest, with Uttar Pradesh and Punjab following
view them differently or that they represent different
it closely, indicating that people are relatively less
notions for the people at large.
trustful of the police in these states (Table 3.4).
3.6. Satisfaction with the police and 3.7. Perception of police and trust
trust in it Other than the demographic variables, people’s
perception of the police as an institution and its
Satisfaction with the police performance in one’s functioning has strong linkages to the level of trust
area also seems to be positively correlated with
Table 3.5: Satisfaction with police performance and its impact on levels of trust in it
Satisfaction with the police Highly trust Somewhat Somewhat Highly distrust Non committal
performance in their area trust Distrust
Fully satisfied 31 47 15 6 1
Somewhat satisfied 24 46 24 5 1
Somewhat dissatisfied 17 45 25 12 1
Fully dissatisfied 11 36 33 18 2
Table 3.6: Trust in police and its impact on satisfaction with its performance
Fully satisfied Somewhat satisfied Somewhat dissatisfied Fully dissatisfied
Highly trust 34 53 8 2
Somewhat trust 27 53 11 4
Somewhat distrust 17 56 12 7
Fully distrust 19 38 18 12
Non-committal 29 28 7 5

Table 3.7: Ranking the states in terms of satisfaction with police performance
Rank States Fully satisfied Somewhat satis- Somewhat dis- Fully dissatis- Score
(%) fied (%) satisfied (%) fied (%)
1 Kerala 51.5 35.4 2.4 1.8 13.2
Himachal
2 29.5 61.8 1.6 0.9 11.7
Pradesh
3 Odisha 36.3 53.6 2.0 3.6 11.7
4 Chhattisgarh 42.9 39.6 4.2 3.2 11.5
5 Punjab 34.8 49.8 3.9 2.3 11.1
6 Gujarat 44.4 37.9 10.3 4.1 10.8
7 Haryana 16.3 72.4 3.5 0.6 10.0
8 Uttarakhand 31.6 50.5 7.8 6.8 9.2
9 Maharashtra 28.7 51.7 8.4 4.3 9.2
10 Jharkhand 32.6 43.0 21.6 2.2 8.2
11 Assam 22.1 57.5 12.4 4.1 8.1
12 Karnataka 18.7 60.8 12.1 2.5 8.1
13 Madhya Pradesh 16.1 63.8 12.1 2.2 8.0
14 West Bengal 26.3 46.7 16.3 2.0 7.9
15 Telangana 18.6 58.7 13.6 3.9 7.5
16 Delhi 21.9 56.8 9.2 8.6 7.4
17 Rajasthan 21.5 54.5 7.5 10.0 7.0
18 Tamil Nadu 17.1 57.7 16.2 3.1 7.0
19 Nagaland 17.1 53.2 11.7 5.6 6.5
20 Andhra Pradesh 27.4 40.2 20.4 5.8 6.3
21 Uttar Pradesh 14.2 52.5 11.1 12.0 4.6
22 Bihar 11.2 52.8 18.2 14.6 2.8
Note: The state rankings for the satisfaction with police performance in the area (see Appendix 4 for details on how the Index was
constructed) are based on summated scores that were arrived at after weighting each Index category. The ‘fully dissatisfied’ category
was weighted as -0.2, the ‘somewhat dissatisfied’ category was weighted as -0.1, the ‘somewhat satisfied’ category was weighted as
0.1, and the ‘fully satisfied’ category was weighted as 0.2. The category of ‘don’t know’ was weighted as 0 and hence excluded from
the ranking analysis. A higher summated score here indicates a greater trust.
they place. Such a perception can be guided by and trust even though we cannot determine whether
experiential instances (direct contact or vicarious the former causes the latter.
experiences) and anecdotal references as well
Furthermore, it was found in the survey that all
(Rosenbaum et al, 2005). Very often, a negative
those respondents who were of the opinion that the
perception leads to low levels of trust, which in turn
police intentionally implicates exuded highest levels
is guided by a multitude of factors. This was evident
of distrust (33%; Table 3.8).
in the survey as well.
Ordinarily, we tend to expect that perception of
Using the Index for perception of police (see
police being corrupt would lead to lesser levels of
Appendix 3 for methodology), its relationship
trust. Such a conception was corroborated by the
with trust in police was explored. It was found that
survey findings. Across all three sub-institutions of
three in every ten people who had a ‘very negative’
police, the perception of corruption correlated with
perception of the police highly distrusted it (Figure
highest levels of distrust- local police (27%, Table
3.9). Correspondingly, highest trust levels were
3.9), senior police officers (32%; Table 3.10) and
found for all those with a ‘very positive’ perception.
traffic police (50%; Table 3.11).
There is definitely a correlation between perception

Figure 3.9: Negative perception of police indicates high levels of distrust
59
54 52
41 40
37 36
29 27
26
21 20
15 14
6 8 7 4
0 0 0 2 0 0 1
Highly trust Somewhat Somewhat Highly Non-committal

trust distrust distrust
Very positive Somewhat positive Somewhat negative Very negative Non-committal
Table 3.8: Distrust levels highest for those who believe police intentionally implicates
Highly distrust Somewhat distrust Somewhat trust Highly trust Non-committal
No, doesn’t implicate 5 19 49 26 1
intentionally
May be does implicate 6 24 46 23 1
intentionally
Yes, implicates 9 24 45 21 1
intentionally
Table 3.9: Perception of corruption and level of trust for local police
A lot of trust Somewhat trust Not much trust Not at all
Local Not at all corrupt 44 38 7 4
police Not much corrupt 20 58 13 3
Somewhat corrupt 17 53 17 6
Extremely corrupt 25 42 15 12
Table 3.10: Perception of corruption and level of trust for senior police officer
Senior Not at all corrupt 53 31 5 3
police Not much corrupt 29 46 13 4
officer
Table 3.11: Perception of corruption and level of trust for traffic police
Traffic police Not at all corrupt 32 29 17 9
Not much corrupt 16 42 25 7

3.8. Experience with police and trust was found in the survey that such a correlation did
exist as an individual who was satisfied with the
The survey found no relationship between having police’s performance when he/she contacted it was
contacted the police and trust in police. Those more likely to have higher levels of trust in it (Table
respondents who said they or a family member of 3.12)
theirs had contacted the police in the last 4-5 years
for some purpose were as likely to trust or distrust Moreover, the study found that among all those
the police as those who said they had not contacted who, after having contacted the police, hadn’t paid
the police during the same duration. a bribe, high or moderate degree of trust in police
was at 74 percent. On the other hand among those
While it seems that the contact with the police does who had paid a bribe trust in the police was at 64
not seem to determine how much one trusts it, the percent (Figure 3.10). However it is still interesting
kind of experience one has had with it does impact to note that even among those who had paid a bribe
it to some extent. The kind of experience that people a majority still were quite trusting of the police.
have tends to have an influence on their trust levels
as well as on their preparedness to trust (Goldsmith, The survey allowed us to study the effect of
2005). For instance, an individual’s satisfaction with satisfactory experience with the police and showed
their engagement with the police, the help that they that this indeed seems to have a positive effect on
received and their functioning in their neighborhood, their inclination to approach the police in the future
has a strong effect on trust. As the levels of trust as well (86%): positive experience contributes
hinge on a number of factors, studying people’s trust to trust, which in turn influences inclination to
in light of their satisfaction levels was important. It approach (Table 3.13).
Table 3.12: Satisfaction with police helps aid trust level

Satisfaction levels with the Highly trust Somewhat Somewhat Highly distrust Non-committal
help at the police station trust Distrust
Very satisfied with 36 48 14 2 0
experience with police
Somewhat satisfied with 23 50 23 4 0
Somewhat dissatisfied with 18 40 32 10 0
Fully dissatisfied with 10 43 33 12 2
Figure 3.10: Paying a bribe and trust levels

47
44
29
27
20 20
5 6
1 1

trust distrust
No, didn’t pay a bribe Yes, paid a bribe

Table 3.13: Satisfaction with the help provided by police is positively related to willingness to approach
it again
Satisfaction levels with the No, will not Probably, will Have no other option but
Yes, will approach
help at the police station approach approach to approach
Very satisfied 4 86 8 1
Somewhat satisfied 8 72 17 3
Somewhat dissatisfied 6 62 21 10
Fully dissatisfied 14 54 22 10
Question asked: In the future, if you have a problem that requires police help, would you go to the police?
Figure 3.11: Those having paid a bribe to the police, relatively more hesitant to approach it in future
78
61
22
11
8 10 7
3
No, will not Yes, will approach Probably, will Have no other
approach approach option but to approach
No, didn’t pay a bribe Yes, paid a bribe
Similarly, the experience of having paid a bribe also did not significantly influence the trust levels.
had a bearing on people’s inclination to approach Studying the correlation between the two across
the police in case of help. It was noted that almost different communities, no reportable consistent
every four out of five people who hadn’t paid a bribe finding was found. For instance, in states where
were inclined towards approaching the police. On SCs are well represented in the police, SC trust in
the other hand, only three in every five of those who the police was much less than SC trust in it in states
had paid a bribe expressed willingness to approach where SCs are not as well represented. With respect
the police in the future (Figure 3.11). to STs, Muslims and women too we saw no clear
pattern with many states with poor representation
3.9. Police diversity and trust in police of these communities in the police showing high
trust of these communities in the police.
One tends to believe that adequate representation
of their respective communities in the police would The survey found that often people are not aware
influence the levels of trust an individual has on the about the level of representation their respective
police, by virtue of a tendency to feel that they are community has in the police force. Their perception
likely to be less vindicated in that scenario. Police of the degree of representation their community has
statistics were available to us on the themes of in the police force of their state vastly differs from
representation of various communities in the police- the actual representation of their community in the
force based on which the states were categorized as police. This could be the reason why contrary to
‘very good/good’ or ‘bad’ depending on the level of what we hypothesized, in states with varying degree
representation. However, it was found in the survey of representation of a community in the police force
that perception of police as diverse or representative does not influence the trust of people.

Even as the degree of representation of a community For STs and SCs, perception of police as diverse
in the police does not affect its trust levels in the and having adequate representation of their own
police, it does seem to be positively associated with community did not seem to have any notable impact
satisfaction with the police. In other words, while on their satisfaction levels.
no consistent linkages were found between actual
Therefore, we see that on the whole, perceptions of
police diversity and trust, some important findings
the police as being diverse or even having adequate
were noted with regard to satisfaction levels.
representation of a particular community has little
In the case of Muslims, in states which are ranked bearing on the degree of trust of an individual even
as ‘very good’ as far as representation of Muslims though it may have an effect on their satisfaction
in the police-force is concerned, 30 percent of the levels in some cases
Muslims stated that they were satisfied with the
police. Meanwhile, in states ranked as ‘very bad’ in 3.10. Conclusion
terms of representation of Muslims in the police,
This chapter has attempted to examine the degree
highest level of dissatisfaction (14%) was noted
of trust that individuals have on the police in India,
(Table 3.14).
by taking into account several factors. Contrary to
This was evident even in case of women. In states impressions, the police seem to enjoy a fairly highly
having ‘very bad’ levels of representation of women degree of trust, even if explicit expression of high
in the police-force, relatively, highest level of distrust was somewhat limited. While in relation
dissatisfaction was noted (7%), while women were to other institutions such as the army and the
most satisfied in states ranked as having ‘Good’ judiciary, the police continues to be less trusted, in
(29%) and ‘Very good’ (24%) representation in this comparison with previous years, the trust levels in
regard (Table 3.15).
Table 3.14: Dissatisfaction in Muslims most in states with ‘very bad’ Muslim representation
States with very good Muslims’ 30 45 13 9
representation in police
States with good Muslims’ 20 56 13 7
States with bad Muslims’ 28 45 11 3
States with very bad Muslims’ 15 49 16 14
Note: Figures are percentages. sample size: 1826. Only Muslim responses have been reported. Rest of the respondents did not respond
to the question.
Question asked: How satisfied are you with the police performance and their work in your area?
States with ‘Very Good’ representation are: Andhra Pradesh, Himachal Pradesh and Odisha. States with ‘Good’ representation are:
Gujarat, Karnataka, Madhya Pradesh, Nagaland, and Chhattisgarh. States with ‘Bad’ representation are Kerala, Maharashtra, Tamil
Nadu, West Bengal and Jharkhand. States with ‘Very Bad’ representation are: Assam, Bihar, Haryana, Punjab, Rajasthan, Uttar
Pradesh, Delhi and Uttarakhand.
Table 3.15: Poor representation of women in police leads to dissatisfaction with police among women
States with very good women’s 24 53 11 3
States with good women’s 29 55 5 4
States with poor women’s 27 50 12 2
States with very poor women’s 22 49 16 7
Note: Figures are percentages. n- 7519. Only women responses have been reported. Rest of the respondents did not respond to the
question.
Question asked: How satisfied are you with the police performance and their work in your area?
States with ‘Very good’ representation are: Himachal Pradesh, Maharashtra and Tamil Nadu. States with ‘Good’ representation are:
Haryana, Odisha, Punjab, Rajasthan, Delhi and Uttarakhand. States with ‘Poor’ representation are: Karnataka, Kerala, MP, WB,
Jharkhand and Chhattisgarh. States with ‘Very Poor’ representation are: Andhra Pradesh, Assam, Bihar, Gujarat, Nagaland, Uttar
Pradesh and Telangana.

the police seem to have increased significantly even community in the police had a very little bearing on
as for other institutions they seem to have declined. the trust of an individual but has some degree of
influence as far as their satisfaction with the police
The survey revealed that trust is to some extent
is concerned. Ordinarily one tends to correlate trust
contingent on an individual’s background, which
with satisfaction, with a close interplay between
was indicated especially by the fact that among
the two, however our survey showed that these two
the communities at the periphery, relatively lower
notions hold altogether different meaning for the
levels of trust were noted. This helped us to know
people.
that as an institution, police and its operation is
also influenced by the social realities or rather the Thus, we see that an individual’s level of trust hinges
inequalities which are so pervasive in the society. on a close interplay of number of factors and varies
It was seen that women, STs, SCs, Muslims and in accordance to their social identity, experience and
poorer sections of the society were warier of the perceptions.
police and as a result, distrusted them slightly more.
Nevertheless, even among these communities, trust References
in police is greater than distrust in police and by a Rosenbaum, Dennis P., Amie M. Schuck, Sandra
wide margin. K. Costello, Darnell F. Hawkins, and Marianne K. Ring,
More than a person’s background however, the “Attitudes toward the Police: The Effects of Direct and
Vicarious Experience,” Police Quarterly 8, no. 3 (2005):
survey found that trust in police is to a large extent
343-365.
a function of one’s satisfaction with it, one’s
perception of it and one’s experience with it. If one Goldsmith, A. 2005. “Police reform and the
is satisfied with its functioning, has a more positive problem of trust,” Theoretical Criminology 9,
perception of it and has had a good experience while no. 4 (2005), http://journals.sagepub.com/doi/
pdf/10.1177/1362480605057727
interacting with it, then one is more likely to trust it.
Moreover, it was noted that the perception of the Lai, Y. and J.S. Zhao, “The impact of race/ethnic,
police as corrupt or as those who falsely implicate neighborhood context, and police/citizen interaction
and misuse their power, led to significantly lower or residents’ attitudes toward the police,” Journal of
levels of trust. Criminal Justice, 38(4) (2010): 685-692.
Loader, Ian and Aogan Mulcahy, Policing and the
While the common perception is such that we tend
Condition of England: Memory, Politics and Culture.
to believe that if the police is more diverse and Oxford University Press, 2003.
has adequate levels of representation of different
communities, it would not only ameliorate this “State of Democracy in South Asia” (SDSA) Reports I &
institution from its often criticized, disorderly II. Jain University Press, 2017.
conduct, but would also ensure a relatively positive Tyler, Tom R., Why people obey the law. New Haven:
perception of the police in the eyes of those Yale University Press, 1990.
communities. However, our survey does not really Warren, M. “‘Introduction’ In Democracy and Distrust,
find this to be the case, at least not in terms of edited by Mark E. Warren,1-22. Cambridge: Cambridge
trust. Police diversity and representation of one’s University Press, 1999.

CHAPTER 4:
PEOPLE’S PERCEPTION OF
DISCRIMINATION BY THE POLICE
A police officer helps a blind man to cross the road near Mahim Durgah, Mumbai
(Credits: Vijayanand Gupta, Hindustan Times, 18 January 2009, Mumbai)

People’s perception of discrimination by the police
4.1. Introduction 4.2. Opinion on discrimination by

One of the key objectives of the Common Cause- police on the basis of caste
CSDS study was to examine citizens’ perception Respondents were asked to share their views
and experience of discrimination by the police. on discrimination by the police on the basis of
As an important part of the executive, the police caste, that is, whether they thought that the police
in India wields immense power due to its primary discriminates on the basis of caste. Contrary
responsibility of maintaining law and order. to many critical writings on the discriminatory
However, often, in the name of investigating crime, practices of the police, this view was not found to be
extracting confession and punishing perpetrators, as widespread. One in four (25%) respondents stated
torture is inflicted not only upon the accused, but that the police discriminates against people on the
also upon complainants, informants and bystanders basis of caste while a large plurality (57%) did not
in the form of physical abuse, custodial death, rape, think so. Nearly one in every five (18%) refused to
humiliation, intimidation and deprivation of food, answer the question (Figure 4.1).
medical attention etc.
With regard to religion, we found that the belief
In addition to the physical police brutality, that the police discriminates on caste grounds was
institutional discrimination is another component strongest among Muslims (30%) and weakest among
of policing that cannot be denied. Discrimination Sikhs and Christians (8% and 11% respectively;
manifests itself in a variety of ways and may be Figure 4.3). while the degree of difference between
motivated by intolerance towards traditionally responses based on religious disaggregation was
marginalized groups, religious communities, caste not very large, those based on caste distinctions
and class background, and gender. For instance, was greater: Upper Castes in both Muslim and
around the time of communal riots, it has been Hindu communities were much more likely to
widely perceived that the police does not act as a have a favourable response for the police than
neutral law enforcement agency and perceptible their backward caste counterparts. Among Hindu
discrimination is alleged in the use of force, communities, it is the Other Backward Classes
preventive arrests, treatment of detained persons at (OBCs) who were found to be most likely to believe
police stations, reporting of facts and investigation, that the police engages in caste-based discrimination
detection and prosecution of registered cases. This (30%), followed by the Scheduled Castes or Dalits
kind of social marginalisation is likely to have (26%). Scheduled Tribes or Adivasis were least
negative consequences and affect social harmony likely to hold this opinion (20%; Figure 4.2).
in society. In addition to less representation of
minorities in the police force and their over- Like among Hindus, among Muslims too, it is the
representation in prison, discriminatory attitudes of backward sections (Muslim OBCs) who were most
the police are one of the ways in which stereotypes likely to believe that the police discriminates on the
and prejudices against certain communities play out grounds of caste (Figure 4.4). This perception of
in public life. Muslim respondents may be derivative of the fact
that the proportion of Muslims in jail far exceeds
This chapter aims to explore people’s perception their overall proportion in the country’s population.
of discrimination by the police. It is organized as
follows: the first five sections address discrimination While these percentages appear to be more moderate
by the police on the basis of caste, religion, gender, compared to the grim picture reported in the
class and region/state; the sixth section explains the literature on this issue, yet the fact that more than
findings thrown up by an overall ‘Discrimination a quarter of those who were interviewed reported
Index’ that includes all the various possible perceived caste-based discrimination by the police is
grounds of discrimination discussed in sections in itself an important revelation. In absolute terms,
one to five. The seventh and last section delves this proportion would translate into huge numbers.
into the perception about the false implication of
marginalized groups by the police.

Figure 4.1: Over a quarter think police discriminates on the basis of caste
18%
25%
Police discriminates on the basis of caste
Police does not discriminate on

No response
57%
Note: Question asked: It is widely believed that police discriminates between people on the basis of different things. In your opinion,
does the police discriminate on the basis of caste?
Figure 4.2: OBCs among Hindus most likely to believe that police engages in caste discrimination
62 61
53 55
30
26
23
20
Upper Castes Other Backward Scheduled Castes Scheduled Tribes

Classes
Police discriminates on the basis of caste Police does not discriminate on the basis of caste
Figure 4.3: Muslims see the police as discriminating on caste basis more than other religious
communities
61
57 56
50
30
26
11
8
Hindus Muslims Christians Sikhs

Figure 4.4: Muslim OBCs more vocal than Muslim upper castes in their perception of discrimination
on caste basis
Muslim OBC 51
37
the basis of caste
Muslim Upper castes 50 Police discriminates on the basis of caste
24
Figure 4.5: Economically well-off among communities more likely to perceive caste-based
discrimination
64 63 64 62
61
54 55 54 53 53
51 52 50
49
35 37
32
28 27 28
24 25
22
17
10 12 12
6
Interestingly, when we disaggregated the community- While the overall figures for perception of caste
wise findings further by economic class we found discrimination may appear moderate, examining
that across all communities (except Christians) state-wise opinion is an important comparative
the perception that the police discriminates on the exercise that helps us in understanding the diverse
basis of caste is stronger among the economically range of public opinion with respect to police. In
well-off sections than the economically worse-off. comparison to the nationwide average of perceived
That is to say, the upper and middle classes among caste-based discrimination (26%), we found Bihar,
Hindu Upper Castes, OBCs, SCs, STs, Muslims and Karnataka and Uttar Pradesh to be the bottom three
Sikhs were more likely to hold the opinion that the states on this parameter, implying that in these states
police discriminates on caste lines than the lower people’s perception of caste-based discrimination by
and poorer sections among all these communities. the police is highest (Table 4.1).
This class divide within communities was strongest
among Muslims, Hindu OBCs and STs (Figure 4.5).

Table 4.1: State-wise opinion on caste-based discrimination by the police
States Opinion of Opinion of Opinion of only OBC Opinion of only ST

all respondents only SC respondents respondents
respondents
Police Police doesn’t Police Police doesn’t Police Police doesn’t Police Police doesn’t
discriminate discriminate on discriminates discriminate on discriminates discriminate on discriminates discriminate on
on caste lines caste lines on caste lines caste lines on caste lines caste lines on caste lines caste lines
Andhra Pradesh 21 63 29 57 18 70 3 72
Assam 10 66 - - 6 74 29 62
Bihar 71 28 70 28 73 26 - -

Gujarat 19 63 13 65 14 71 46 36
Haryana 11 84 2 92 33 65 - -
Himachal Pradesh 3 86 1 76 - - - -
Karnataka 53 37 44 47 60 27 16 80
Kerala 5 76 - - 4 79 - -
Madhya Pradesh 46 41 49 32 55 29 23 65
Maharashtra 24 64 20 67 24 63 11 73
Nagaland 5 60 - - - - 6 59
Odisha 9 64 12 57 9 65 5 60
Punjab 11 59 11 76 15 37 - -
Rajasthan 19 50 15 60 22 54 23 68
Tamil Nadu 40 45 42 39 40 47 - -
Uttar Pradesh 49 39 49 32 52 39 - -
West Bengal 10 47 11 43 10 40 22 52
Delhi 22 65 13 68 26 64 - -
Jharkhand 24 64 10 89 20 77 36 64
Chhattisgarh 10 52 10 68 13 43 5 61
Uttarakhand 10 68 10 59 14 62 - -
Telangana 30 54 28 56 31 54 17 69
Note: Figures are percentages. Rest of the respondents did not respond. The views of respondents from states with a hyphen (-) have not been reported due to a small sample size.
SCs, STs and OBCs reported above belong to all religious communities.
Additionally, our analysis suggests that Scheduled when it is confronted with situations of inter-caste
Castes in Bihar, Madhya Pradesh and Uttar Pradesh strife. Over three in five (63%) respondents said
were more likely to perceive caste-based discrimina- that police is not partial towards a particular caste
tion than Scheduled Castes in other states. Bihar, group in such situations whereas about one in ten
Madhya Pradesh and Uttar Pradesh, along with (8%) thought it to be so. A large chunk of about one-
Karnataka, are also states where OBCs were found third respondents chose to not answer this question.
to be more likely to be of the opinion that police Muslims (11%) and Hindu OBCs (9%) were more
engages in caste-based discrimination than OBCs in likely to opine that police tends to be partial (Table
other states. In terms of Scheduled Tribe responses 4.2). Among Muslims, OBC Muslims tended to
to the question, Gujarat and Jharkhand stood out believe far more than ‘Upper caste’ Muslim that the
compared to other states. Tribals in these two states police is impartial in such situations (13 percent as
were far more likely to believe that police discrimi- opposed to 8 percent).
nates on the basis of caste than tribals in other states
(Table 4.1). 4.3. Opinion on discrimination by
Probing another aspect of caste-based police on the basis of religion
discrimination, the study also asked the respondents Respondents were also asked about the prevalence
their view on impartiality shown by the police of discrimination by the police on the grounds of
Figure 4.6: People’s opinion on police impartiality in the event of a caste strife
8%
29% Police sides with a particular

caste group in the event of a
caste strife/fight
Police is impartial in the event

of a caste strife/fight
No response
63%
Note: Question asked: In an area, whenever there is an instance of fight between people from two caste groups, do you think the
police sides with any particular caste group or remains impartial?
Table 4.2: Perception of castes and communities on police impartiality in the event of a caste strife
Police sides with a Police is impartial in the No response
particular caste group in the event of a caste strife/fight
event of a caste strife/fight
Upper Castes 7 68 25
Other Backward Classes 9 61 30
Scheduled Castes 8 60 32
Scheduled Tribes 6 63 31
Hindus 8 63 29
Muslims 11 58 31
Christians 4 66 30
Sikhs 4 55 41
Note: Figures are percentages. Figures of Upper Castes, Other Backward Classes, Scheduled Castes and Scheduled Tribes are for
Hindus only.

religion. About one in every five (19%) of them in Bihar, Karnataka and Tamil Nadu (Table 4.3). On
said that it does take place while three in every the other hand, respondents in Haryana, Himachal
five (61%) denied its occurrence (Figure 4.7). One- Pradesh, Kerala and Odisha were least likely to
fifth (20%) did not answer the question. Among all opine that the police discriminates on the grounds
religious communities Muslims were most likely of religion. When we did a state-wise analysis by
to hold the view that the police discriminates on looking at only Muslim responses we found the belief
religious grounds with one in four (26%) of them among community that the police discriminates on
stating so (Figure 4.8). Among Hindus, this figure religious grounds to be greatest in Bihar, Rajasthan,
was much less at 18 percent and among Christians Tamil Nadu and Maharashtra. Among Christians,
it was 16 percent. Sikhs were the least likely to hold the feeling that the police discriminates on religious
the opinion that the police engages in religion-based grounds was found to be strongest in the southern
discrimination. Only 6 percent of them thought so. states except Kerala.
Out of all the States, people’s perception of religious As in the case of caste, respondents were also asked
discrimination by the police was found to be greatest their view on impartiality shown by the police when
Figure 4.7: Three in every five deny occurrence of discrimination by police on religious grounds
20% 19%
Police discriminates on the basis

of religion
the basis of religion
No response
61%
Note: Question asked: It is widely believed that police discriminates between people on the basis of different things. In your opinion,
does the police discriminate on the basis of religion?
Figure 4.8: Muslims most likely to see police as discriminating on grounds of religion
62 64
58
54
26
18 16
6
Police discriminates on the basis of religion Police does not discriminate on the basis of religion
Note: Figures are percentages. Rest of respondents did not respond.

Table 4.3: State-wise opinion on religious discrimination by the police
States Opinion of all the respondents Opinion of only Muslims
Police discriminates Police does not Police discriminates Police does not
on the basis of discriminate on the basis of religion discriminate
religion on the basis of religion on the basis of religion
Andhra Pradesh 19 63 26 64
Assam 12 66 14 54
Bihar 40 57 56 44
Gujarat 15 64 15 59
Haryana 2 93 - -
Himachal Pradesh 0 89 - -
Karnataka 37 51 32 65
Kerala 5 75 4 77
Madhya Pradesh 20 56 35 31
Maharashtra 22 64 50 43
Nagaland 13 68 - -
Odisha 5 64 - -
Punjab 8 61 - -
Rajasthan 22 51 55 36
Tamil Nadu 34 48 50 22
Uttar Pradesh 26 51 39 44
West Bengal 11 47 13 38
Delhi 23 64 41 47
Jharkhand 24 74 25 74
Chhattisgarh 5 52 - -
Uttarakhand 8 68 7 76
Telangana 21 63 26 57
Note: Figures are percentages. Rest of the respondents did not respond. The views of respondents from states with a hyphen (-) have
not been reported due to a small sample size.
Figure 4.9: Two in every three view police as being impartial when faced with an inter-religious
conflict
7%
Police sides with a
27% particular caste group in
the event of a religious
strife
Police is impartial in the

event of a religious strife
No response
66%
Note: Question asked: In an area, whenever there is an instance of fight between people from two religious communities, do you think
the police sides with any particular religious community or remains impartial?

it is confronted with situations of inter-religious 4.4. Opinion on discrimination by
strife. Two in every three (66%) respondents
said that police is not partial towards a particular
police on the basis of gender
religious group in such situations whereas only 7 On the question of gender discrimination, three out
percent thought it to be so. Nearly three in every of ten respondents stated that police discriminates
ten (28%) chose to not answer this question (Figure between men and women. One in every two (50%)
4.9). Muslims were twice as likely to opine that denied its occurrence and 20 percent remained silent
police tends to be partial than Hindus, 12 percent on the issue, that is, they did not answer the question
as opposed to 6 percent. Christians and Sikhs were (Figure 4.11). Interestingly, there was not much of a
the least likely to believe so (Figure 4.10). Muslims difference between how men and women responded
in Delhi, Haryana and Rajasthan stood out in their to this question. If anything, men were slightly more
responses to this question compared to Muslims in likely to view the police as being discriminatory on
other states. In other words, Muslims in these three the grounds of gender than women were, particularly
states were far more likely to believe that the police in towns and cities (Figure 4.12).
is partial to a particular religious group whenever
there is a fight between people from two religious Educational attainment seems to make a significant
communities. difference to views on gender discrimination by the
Figure 4.10: Perception of religious communities on police impartiality in the event of a religious strife
67 67
61
52
44
30
27 27
12
6 4
3
Police sides with a particular caste Police is impartial in the event No response
group in the event of a religious strife of a religious strife
Figure 4.11: Perception on gender-based discrimination by the police
20%
Police discriminates on
30% the basis of gender
Police does not

discriminate on the basis
of gender
No response
50%
Note:Question asked: It is widely believed that police discriminates between people on the basis of different things. In your opinion,
does the police discriminate on the basis of gender?

Figure 4.12: Urban men and women more likely to see gender bias among police than rural men and
women
53 54 55
48
43 44
37
34
29 29 31
28
23 22
18 20
15 17
Men in Women in Men in Women in Men in cities Women in

villages villages towns towns cities
Police discriminates on Police does not discriminate No response
the basis of gender on the basis of gender
police, both among men and women. The more Rajasthan, Punjab and Kerala were least likely to
educated a man or a woman, the more likely he or hold such a perception (Table 4.4).
she is to see the police as being biased on gender
lines (Figure 4.13). 4.5. Opinion on discrimination by
A state-wise analysis of responses revealed that police on the basis of class
respondents in Bihar, Telangana and Maharashtra In comparison to caste, gender and religion,
were most likely to view the police as being gender- we found that in the eyes of the people class
biased. Nearly half the respondents in these three was the more significant and telling cleavage of
states saw the police as discriminating between discriminatory attitudes of the police. Overall, one
men and women. This finding held true for women in every two (51%) respondents was of the opinion
as well, that is, women of these three states were that police discriminates on the basis of class, that
also far more likely to view the police as being is, between rich and poor (Figure 4.14). There is no
discriminatory on the grounds of gender than difference between how the economically well-off
women in other states. Women in Chhattisgarh,
Figure 4.13: Highly educated women and men more likely to express gender discrimination by police
54 53
50 51 49
48 48 48
34 36
32 32
27 29 27
25
Non- Primary High College Non- Primary High College

literate pass school educated literate pass school educated
women women pass women women men men pass men men
Police discriminates on the basis of gender Police does not discriminate on the basis of gender
Note: Figures are percentages. Rest of the respondents did not respond

Table 4.4: State-wise opinion on gender discrimination by the police
State Name Opinion of all respondents Opinion of only women respondents
on the basis of discriminate on the on the basis of discriminate on the basis
gender basis of gender gender of gender
Assam 37 49 39 45
Bihar 52 45 53 44
Gujarat 26 55 25 52
Haryana 28 67 26 69
Himachal 30 59 23 61
Pradesh
Kerala 13 67 14 62
Nagaland 25 60 24 63
Odisha 16 55 17 49
Punjab 17 54 14 55
Delhi 41 46 40 45
Chhattisgarh 12 50 12 50
Figure 4.14: Opinion on class- based discrimination by the police
16%
Police discriminates on
the basis of class
Police does not

51% discriminate on the basis
of class
33%
No response
Note: Question asked was: It is widely believed that police discriminate between people on the basis of different things. In your
opinion, does the police discriminate on the basis of class?

and worse-off answered this question. Upper and on the grounds of economic class (Table 4.5).
middle-class respondents were found to be as likely Maharashtra too reported a high proportion of such
to see the police as discriminating on class lines discrimination perceivers – over two in every three
as lower class and poor respondents (Figure 4.15). respondents (68%) there saw the police as engaging
Among the poorest sections, it is those living in big in class-based discrimination. If we just analyse the
urban centres that were more likely to believe that responses of the poorest respondents by states, then
the police discriminates on class lines than those those in Bihar, Punjab and Delhi were more likely to
living in small towns and villages (Figure 4.16). affirm to the police’s differential treatment between
the rich and poor than those in other states.
Nearly three out of every four respondents (73%)
in Bihar and Delhi saw the police as discriminating
Figure 4.15: Both rich and poor equally likely to believe that police discriminates on class lines
51
Upper and middle class Police does not discriminate
34
on the basis of class
(between rich and poor)
51
Lower class
34
Police discriminates on the
basis of class
51 (between rich and poor)
Poor 31
Note: Figures are percentages. Rest of the respondents did not respond
Figure 4.16: Urban poor more likely to view police as being discriminatory than rural poor
57
52
48
34 32
25
Poor in big cities Poor in small cities Poor in villages

and towns
Police discriminates on the basis of class Police does not discriminate on the basis of class
(between rich and poor) (between rich and poor)

Table 4.5: State-wise opinion on class-based discrimination
States Opinion of all respondents Opinion of poorest respondents only

on the basis of class discriminate on the on the basis of class discriminate on the
(between rich and basis of class (between (between rich and basis of class (between
poor) rich and poor) poor) rich and poor)
Assam 56 35 64 25
Bihar 73 25 81 19
Gujarat 44 43 58 29
Haryana 55 39 31 64
Himachal Pradesh 40 50 59 31
Kerala 40 43 41 38
Nagaland 31 53 20 59
Odisha 42 36 30 43
Punjab 49 23 77 13
Delhi 73 20 73 18
4.6. Opinion on discrimination by another state. Recent residents (those who had been
living in a big city for 1-5 years) were least likely to
police on the basis of state believe that police discriminates against people from
The study also tried to find out people’s opinion another state (only 14% thought so) than those who
about discrimination by the police against people had been living there for a longer time. In fact, big
from other states/migrants. Only 16 percent said city residents who had been living in the city since
that such discrimination by the police takes place birth were most likely to hold such a view (24%;
whereas 48 percent denied its occurrence. More Figure 4.19). In other words, non-migrants/original
than one in every three (36%) did not answer the inhabitants in big cities were ten percentage points
question (Figure 4.17). Respondents residing in more likely to view the police as discriminating
big cities were more likely to perceive the police as against people from another state than migrants
discriminatory on this question than those living who had arrived in the city only recently. We cannot
in small towns or villages (Figure 4.18). We did however identify how many of these recent migrants
not however find any association between being a to a city had come from another state since we did
recent migrant/resident of a big city and holding the not ask a question probing the state of origin of such
opinion that police discriminate against people from a respondent. The pattern with respect to big cities

also holds with respect to small towns but not to the 4.7. Overall perception of police
same degree.
discrimination
This belief that the police discriminates against
people from another state was found to be strongest In order to get a holistic view of people’s perception
in Karnataka, Himachal Pradesh, Tamil Nadu and of discrimination by the police, we constructed an
Jharkhand (Table 4.6). If we take into account only all-encompassing Index (see Appendix for details
urban responses, then Karnataka, Tamil Nadu on how it was constructed) that included all the 7
and Maharashtra stand out. Respondents in urban discrimination-related questions that have been
centres in these states were most likely to believe separately probed in sections 5.2. to 5.6. above.
that the police discriminates against people from We found that only about 7 percent thought that
other states.
Figure 4.17: Opinion on discrimination by the police against people from another state
16% Police discriminates

against people from
36% another state
Police does not

discriminate against
people from another
state
No response
48%
Note: Question asked was: It is widely believed that police discriminate between people on the basis of different things. In your
opinion, does the police discriminate against people from another state?
Figure 4.18: People in cities more likely to believe police discriminates against people from another
state
52
45 47
21
16 15
City Town Village
Police discriminates against Police does not discriminate

people from another state against people from another state

Figure 4.19: Non-migrants in big cities more likely to see police as discriminating against people from
another state than recent migrants
69
60
55 55
51 51 51
47 47
41
24
20 18 19 17 17 15
14 14 14
Big city Big city Big city Big city Big city Town Town Town Town Town
resident resident resident resident resident resident resident resident resident resident
living living living living living living living living living living
there since there for there for there for there for there since there for there for there for there for
birth over 15 over 10-14 over 6-9 over 1-5 birth over 15 over 10-14 over 6-9 over 1-5
years years years years years years years years
Police discriminates against people Police does not discriminate

from another state against people from another state
Table 4.6: State-wise opinion on state-based discrimination

States Opinion of all respondents Opinion of urban respondents only
against people from discriminate against against people from discriminate against
another state people from another state another state people from another state
Assam 10 53 17 75
Bihar 24 60 25 52
Gujarat 8 49 10 35
Haryana 3 84 1 87
Kerala 16 54 15 55
Nagaland 18 56 7 69
Odisha 4 50 7 66
Punjab 16 53 13 32
Rajasthan 9 47 4 14
Delhi 24 51 24 51
Note: Figures are percentages. Rest of the respondents did not respond to the question. The views of respondents from states with a
hyphen (-) have not been reported due to a small sample size.

police either discriminates very rarely or doesn’t 4.8. Opinion on false implication of
discriminate at all, 13 percent fell in the category
of those who felt it discriminates rarely, 67 percent
certain communities by the police
or over two-thirds thought that police discriminates Another aspect of police discrimination is the unfair
somewhat and 9 percent or nearly one- tenth targeting of certain vulnerable communities by the
were found to view the police as being highly police and their false implication in cases. This
discriminatory (Figure 4.20). section deals with people’s opinion on this aspect.
We also tried to find out how each of the 22 states Respondents were asked to share their views on the
where the survey was conducted fared with respect false implication of Dalits in petty crimes, Adivasis
to this comprehensive discrimination perception on Maoist charges and Muslims in terrorism-related
Index. The states were ranked on the basis of cases. Nearly two in every five (38%) respondents
summated scores that were arrived at after assigning agreed with the proposition that the often the police
different weights to each of the Index categories falsely implicates members of backward castes such
(see note below Table 4.7 or the Appendix 4 for as Dalits in petty crimes such as theft, robbery, dacoity
methodology). West Bengal, Chhattisgarh and etc. However, agreement with the proposition that
Odisha fared the best – respondents in these three Adivasis are falsely implicated on Maoist charges
states were least likely to perceive the police as being and that Muslims are falsely implicated in terrorism-
discriminatory compared to the other states. On the related cases was not as high. Only a little over one
other hand, Bihar, Jharkhand and Haryana fared in every four (28%) were of the opinion that such
the worst with respondents here being most likely false implication of Adivasis and Muslims occurs
to perceive the police as being discriminatory in its (Table 4.8).
functioning.
Figure 4.20: Index on perception of discrimination
4% 9%
7%
13% Highly discriminates

Somewhat discriminates
Rarely discriminates
Very rarely discriminates
Non-committal
67%
Note: See Appendix to know how Index was constructed.

Table 4.7: State-wise ranking based on overall perception of discrimination
Rank State Very rarely Rarely Somewhat Highly Score
discriminates discriminates Discriminates discriminates
(%) (%) (%) (%)
1 West Bengal 19.9 17.3 47.5 4.7 0.0
2 Chhattisgarh 17.1 15.6 57.1 0.6 -0.9
3 Odisha 15.0 15.3 60.8 1.2 -1.8
4 Rajasthan 16.4 16.0 55.2 8.4 -2.3
5 Nagaland 4.9 26.7 65.3 1.1 -3.1
6 Uttarakhand 8.3 18.2 66.9 4.1 -4.0
7 Madhya Pradesh 4.3 24.4 66.4 3.8 -4.1
8 Andhra Pradesh 8.9 16.3 69.8 2.8 -4.1
9 Himachal Pradesh 4.1 18.1 70.5 0.2 -4.5
10 Uttar Pradesh 9.4 15.1 62.7 9.6 -4.8
11 Kerala 8.5 13.4 71.9 3.3 -4.8
12 Gujarat 7.7 13.1 72.3 4.2 -5.2
13 Punjab 4.7 3.7 66.6 3.9 -6.1
14 Telangana 2.6 13.9 77.6 3.6 -6.6
15 Assam 5.4 10.9 77.0 5.8 -6.7
16 Maharashtra 4.0 8.7 73.4 12.0 -8.1
17 Tamil Nadu 4.7 9.6 56.9 21.8 -8.2
18 Karnataka 3.6 7.7 65.1 17.8 -8.6
19 Delhi 5.6 7.2 68.2 18.5 -8.7
20 Haryana 4.6 0.0 86.4 8.4 -9.4
21 Jharkhand 0.4 3.6 88.6 7.0 -9.8
22 Bihar 0.2 0.7 66.8 31.8 -12.9
Note: The state rankings for The Index of Perception of Discrimination by Police (see Appendix 3 for more details on how the Index was
constructed) are based on summated scores that were arrived at after weighting each Index category. The ‘highly discriminates’ category
was weighted as -0.2, the ‘somewhat discriminates’ category was weighted as -0.1, the ‘rarely discriminates’ category was weighted as
0.1, and the ‘very rarely discriminates’ category was weighted as 0.2. The category of non-committal (those who did not answer any
question that went into making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher summated score
here indicates a positive assessment, i.e., weaker perception of discrimination.
Table 4.8: Opinion on false implication of marginalized communities by the police

Those who agree that there Those who disagree that
is false implication there is false implication
View of all respondents on false implication of SCs in 38 39
petty crimes by the police
Views of only SCs on false implication of SCs by the 35 43
police
View of all respondents on false implication of STs on 28 42
Maoist charges by the police
View of only STs on false implication of STs on Maoist 27 42
charges by the police
View of all respondents on the false implication of 27 43
Muslims in terrorism related cases
View of only Muslims on the false implication of Muslims 47 31
in terrorism related cases
Question asked: Now I will read out three statements. Please tell me whether you agree or disagree with each? (Probe further whether
‘fully’ or ‘somewhat’ agrees or disagrees) a. Often members of backward castes such as Dalits are falsely implicated in petty crimes
such as theft, robbery, dacoity by the police b. Often tribals are falsely implicated on Maoist charges by the police c. Often Muslims are
falsely implicated in terrorism related cases by the police.

Table 4.9: State-wise opinion on false implication of SCs in petty crimes
States Among all the respondents Among only Scheduled Castes
Those who agree Those who disagree Those who agree Those who disagree
that there is false that there is false that there is false that there is false
implication of Dalits implication of Dalits implication of Dalits implication of Dalits
in petty crimes by the in petty crimes by the in petty crimes by the in petty crimes by the
police police police police
Assam 25 58 37 63
Bihar 36 61 26 71
Gujarat 28 47 28 58
Haryana 27 67 19 79
Kerala 24 45 28 34
Nagaland 5 33 - -
Odisha 40 35 23 48
Punjab 52 26 33 42
Delhi 33 50 35 47
Jharkhand 53 42 66 34-
Table 4.10: State-wise opinion on false implication of tribals on Maoist charges

States Among all the respondents Among only Scheduled Tribes
implication of tribals implication of tribals implication of tribals implication of tribals
on Maoist charges by on Maoist charges by on Maoist charges by on Maoist charges by
the police the police the police the police
Assam 13 67 35 62
Bihar 28 66 - -
Gujarat 17 50 47 31
Haryana 14 61 - -
Himachal 30 55 - -
Pradesh
Kerala 22 37 - -
Nagaland 5 33 6 37

States Among all the respondents Among only Scheduled Tribes
implication of tribals implication of tribals implication of tribals implication of tribals
on Maoist charges by on Maoist charges by on Maoist charges by on Maoist charges by
the police the police the police the police
Odisha 42 29 45 26
Punjab 18 27 - -
Tamil Nadu 37 34 - -
Uttar Pradesh 33 40 - -
Delhi 21 55 - -
Uttarakhand 16 47 - -
Table 4.11: State-wise opinion on false implication of Muslims in terrorism related cases
States Among all the respondents Among only Muslims
implication of implication of implication of implication of
Muslims in terrorism Muslims in terrorism Muslims in terrorism Muslims in terrorism
related cases by the related cases by the related cases by the related cases by the
police police police police
Assam 22 65 31 46
Bihar 26 70 49 48
Gujarat 15 53 17 64
Haryana 16 63 - -
Kerala 20 41 32 37
Nagaland 2 34 - -
Odisha 31 30 - -
Punjab 28 25 - -
Delhi 30 54 65 20
Chhattisgarh 6 39 - -

While Muslims were much more likely to see a large plurality of respondents did not adhere to
their community as being falsely implicated by the view on discriminatory practices of the police.
the police than non-Muslims, quite interestingly Nevertheless, the opinion of those who affirmed
the same pattern did not hold with respect to the discriminatory attitudes of rule of law revealed
Dalits and Adivasis. Respondents belonging to core cleavages of caste and community. Overall,
the two communities were less likely to see their people were much more likely to report class-based
communities as victims of false implication by the discriminatory attitudes of the police, followed by
police than non-Dalits and non-tribals (Table 4.8). gender and caste-based discrimination, and least
likely to perceive religious discrimination by the
An examination of state-wise opinion suggests
police. While on one hand, a large plurality of
that Dalits in Jharkhand, Telangana and Madhya
respondents stated that police remain impartial in
Pradesh (Table 4.9), Adivasis in Maharashtra,
case of inter-community conflict, at the same time
Gujarat and West Bengal (Table 4.10) and Muslims
among those who expressed otherwise Muslims
in Telangana, Delhi, Tamil Nadu, Maharashtra and
were most likely to endorse the view of police’s
Karnataka (Table 4.11) were most likely to hold the
partiality. The scope of these findings builds on
opinion that their respective communities are falsely
several previous studies on the fragile relationship
implicated by the police than their counterparts in
between public institutions such as the police and
other States.
citizens in general. This chapter has shown that
4.9. Conclusion the sense of discrimination and marginalisation
is not restricted to Muslims and the very fact that
This chapter sought to understand people’s discrimination continues to be felt among minority
perception of discrimination by the police. It is a communities needs to be recognized and duly
positive sign that perceptions of discrimination addressed by the state.
were articulated only on a moderate scale and

CHAPTER 5:
FEAR OF POLICE AND ATTITUDES
TOWARDS ITS EXCESSES
Police clash with passengers blocking railway lines at Diva Station in Mumbai after rail traffic was severely disrupted following violent
public protests at the Central line in Mumbai (Credits: Rishikesh Choudhary, Hindustan Times, 2 January 2015, Kalyan)

Fear of Police and Attitudes towards its Excesses
5.1. Introduction being beaten by the police, being approached by it or

of being sexually harassed by policemen.
Custodial torture, extra-judicial killings and
unwarranted incarceration are some of the Fear or anxiety regarding the police is as much a
most violent forms of police excesses that one function of people’s perception as it is about the
closely associates with the functioning of police actual actions of the police. When citizens view
departments across the country (Human Rights the police as representatives of the state’s coercive
Watch, 2016). In this chapter, we seek to understand power and not the protectors of their rights, it is
people’s perceptions about these issues and how it understandable that they fear the possibility of the
might shape their attitudes towards police behaviour. police’s unjustified actions. Such trepidation not only
Broadly, questions pertaining to this topic probe the affects their perception of the police as prejudiced or
following three themes - fear of the police, attitudes submissive to their political masters, but also affects
towards custodial excesses and awareness of police the likelihood of one approaching the police in times
brutality. of need, as we show in later sections. It therefore
becomes important to understand the various
In this chapter we present citizens’ responses to dimensions of this fear, as well as possible reasons
questions about police excesses; we check these for it. We do this first by trying to ascertain what is
responses against demographic variables of region, it about the police that makes people most anxious,
caste, class, gender and religion, to report the and then to see the interaction of the overall fear
possible patterns that emerge from this exercise. with other responses. The Common Cause-CSDS
Once this step firmly established the groundwork, survey found about two in every five respondents
we turned towards finding correlations and possible being fearful (either ‘very’ or ‘somewhat’) of the
explanations using other questions of police police when it came to being beaten up by them
perception and empirical information. The chapter (44%), arrested by them (38%), falsely implicated by
is thematically divided into three sections, each them (38%) and being approached by them at home
disaggregating and analysing responses on questions (38%). On the issue of sexual harassment by the
pertaining to the leading theme for that section— police personnel, a little over one in every four (27%
fear of police, police brutality or awareness about overall and 29% among women) said they either
police excesses. feared it a lot or somewhat (Table 5.1).
5.2. Fear of police To get a more comprehensive sense of this fear
of police, we constructed an Index which would
To understand people’s overall attitude towards the aggregate responses from various sub-parts of the
police, it was important to understand their level of question to give us overall levels of fear among the
trepidation and how it shaped their interaction with people (for details on how the Index was constructed,
the police. In this section, we have used an Index see Appendix 3). The Index found 14 percent of the
especially created for this purpose. This Index on respondents to be highly fearful of the police and 30
fear of police combines a set of questions which percent to be somewhat fearful of it. Twenty-four
each probed a different aspect of the fear - whether percent were found to be not much fearful and 27
it was the anxiety about being wrongfully arrested, percent turned out to be not at all fearful.
Table 5.1: People most fearful of being beaten up by the police

A lot Somewhat Not much Not at all
Fear of being beaten up by the police 16 28 18 32
Fear of being arrested by the police 13 25 20 36
Fear of police coming to your house 14 24 19 36
Fear of being falsely implicated by the police 14 24 19 35
Fear of sexual harassment by the police 11 16 17 41
Question asked: Often people are scared of the police due to various reasons. What about you, how scared are you of following - very
fearful, somewhat fearful, not much fearful or not at all fearful?

Figure 5.1: Index of fear of police
5%
14%
Highly fearful of police
27% Somewhat fearful of police
Not much fearful of police

30%
Not at all fearful of police
Non-committal
24%
Note: See Appendix 3 on how the Index of fear of police was constructed.
In Table 5.2, a state-wise distribution of the various country. From the state responses we derived a unique
levels of fear in different states of the country is score for each state, which would reflect its overall
reported, which gives us a relative and comparative levels of fear in a single figure (for methodology
picture of the trepidation felt by citizens across the see note below the table). Arranging the states in a
Table 5.2: State-wise ranking of responses about fear of police: Punjab is the most afraid
Rank State Not at all Not much Somewhat Highly Non- Score
fearful fearful fearful fearful committal
1 Himachal Pradesh 83.2 9.1 2.9 0.2 4.5 17.2
2 Uttarakhand 71.0 17.6 9.6 1.4 0.5 14.7
3 Haryana 53.7 39.9 4.0 2.3 0.2 13.9
4 Kerala 51.9 27.1 12.9 3.8 4.4 11.0
5 Delhi 58.0 19.3 14.9 6.2 1.6 10.8
6 Rajasthan 41.6 25.0 18.7 2.2 12.4 8.5
7 Maharashtra 39.1 30.5 23.7 4.6 2.1 7.6
8 Nagaland 27.7 39.5 24.0 4.4 4.4 6.2
9 Chhattisgarh 29.5 30.9 23.2 4.2 12.2 5.8
10 Assam 25.5 32.5 31.9 6.9 3.2 3.8
11 Gujarat 27.0 31.1 25.8 14.0 2.1 3.1
12 West Bengal 27.0 29.5 26.3 13.8 3.5 3.0
13 Bihar 26.8 24.5 43.4 4.8 0.5 2.5
14 Madhya Pradesh 11.0 35.3 43.7 8.0 2.0 -0.2
15 Odisha 24.5 17.3 26.2 23.2 8.8 -0.6
16 Jharkhand 19.0 17.6 56.6 6.4 0.4 -1.4
17 Uttar Pradesh 15.4 20.4 41.5 17.8 4.9 -2.6
18 Telangana 9.5 22.4 54.4 8.9 4.8 -3.1
19 Andhra Pradesh 9.0 17.1 43.4 25.1 5.3 -5.9
20 Tamil Nadu 4.8 14.8 33.5 39.2 7.6 -8.8
21 Karnataka 3.8 14.7 44.1 33.9 3.6 -9.0
22 Punjab 6.1 9.8 20.9 46.7 16.4 -9.2
Note: The state rankings for the Index of fear of police (see appendix 3 for details on how the Index was constructed) are based on summated
scores that were arrived at after weighting each Index category. The ‘highly fearful’ category was weighted as -0.2, the ‘somewhat fearful’
category was weighted as -0.1, the ‘not much fearful’ category was weighted as 0.1, and the ‘not at all fearful’ category was weighted as
0.2. The category of non-committal (those who did not answer any question that went into making the Index) was weighted as 0 and
hence excluded from the ranking analysis. A higher summated score here indicates a positive assessment, i.e., lesser fear.

descending order of their scores—highest score first, most fearful of the police here. The control and
representing least amount of fear - we arrived at the responsibility of the police falls under the state list
following distribution, which displays the relative and such a relative analysis tells us about the vastly
position of each state regarding this question. The differing perceptions the same institution inspires in
negative sign denotes that the presence of fear different states, opening the space for deliberating
outnumbers those respondents who are unafraid of and comparing different state policies of policing.
the police. We see that Himachal Pradesh fares best It is interesting to note that nearly all southern states
among all the states, meaning that here people are reported high levels of fear. Kerala was the only
the least fearful of being a victim of police excesses, exception where fear of police was found to be quite
while Punjab comes last, signalling that people are low.
Figure 5.2: Religion-wise distribution of police fear: Sikh fear higher than other communities
37
34
30 29 30 28 29
26
23 23
17 16
14
10 11 9

Highly fearful Somewhat Not much Not at all
of police fearful of fearful of fearful of
police police police
Note: Figures are percentages. Rest of the respondents did not respond to the question. Figures for other religions have not been
reported due to their small sample size.
Figure 5.3: Muslims in the South are more likely to fear the police
44
34 34 35
31
29
27 26 26
24
20
17
15
10 10
7
North India South India East India West-Central India
Highly fearful Somewhat fearful Not much Not at all

of police of police fearful of police fearful of police
Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:1826 (Only Muslim responses).
States in the North include: Haryana, Himachal Pradesh, Punjab, Rajasthan, Uttar Pradesh, Delhi, Uttarakhand; States in the South
include: Andhra Pradesh, Karnataka, Telangana, Kerala and Tamil Nadu; States in the East include: Assam, Bihar, Nagaland,
Odisha, West Bengal and Jharkhand; States in the West-Central region include: Gujarat, Madhya Pradesh, Maharashtra and
Chhattisgarh.

5.2.1. Community demographics and fear of was found to be 7 percent, 3 points less than the
police overall average, but in states with a low population
concentration, the situation reverses itself with
From a region-based explanation, we now go 20 percent Muslims reporting high fears, which
towards delineating the possible reasons as to is double the national average (Table 5.3). For a
why certain states are ranked higher than others, minority community such as Muslims in India,
while certain others lag behind. A religion-wise it seems to appear that higher concentration also
distribution shows us that Sikhs have the highest equals lesser fear of police violence, whereas lower
levels of fear, which turns out to be much higher population makes for more insecurity.
than the national average (Figure 5.2). The state-
wise distribution shows the high incidence of this Among Hindus, the highest fear is displayed in the
fear in Punjab (Table 5.2). Given that a majority of Scheduled Caste (SC) category, followed by the
the Sikh responses are coming from this state, it is Other Backward Classes (OBC) (Figure 5.4). Among
the high fear levels in Punjab which are contributing SCs, fear is highest among the poorest class (23%).
to the high figures. When further disaggregated, Upon further disaggregating this and considering
the likelihood of poor Sikhs being scared is higher, only the category of highly fearful, we see that SCs
a trend which is repeated across all other groups. in the south are more likely to be scared (33%) than
But in comparison to the response of upper classes SCs in the North (13%) or East (14%). Historically
from other religions, the Sikh upper class is much and socially, movements against caste oppression
more likely to be scared (42%) as against upper class have been more successful in the south than in the
Hindus (14%) or upper class Muslims (9%) (a trend north. In such a scenario, the high fear numbers
that can have a possible connection to the particular from this region become all the more significant. For
history of Punjab in the last four decades). ST respondents, while the overall levels are very low
compared to other groups, some states do become
A region-wise examination of the Muslim significant outliers to this picture. ST responses from
community shows that Muslims in southern India Gujarat, Karnataka and Odisha are significantly
are more likely to be fearful of the police than higher than the average (highly fearful- 36%, 17%
Muslims in other parts of the country (Figure 5.3). and 18% respectively).
Here 17 percent Muslims were found to be highly
fearful of the police, which is 7 points higher than Another significant detail that emerges from
the national average. A state-wise breakup of the further disaggregation is that fear and caste don’t
southern region reveals Muslims in Karnataka, always have a simplistic narrative. For instance, in
Tamil Nadu and Andhra Pradesh to be the most disaggregating Hindu castes, we see that 39 percent
fearful. It should also be noted here that the fear of Upper Castes in rural areas reported to have no
among Hindus was also highest in the southern fear of the police at all. This figure is not only much
states, compared to other regions in the country. higher than the overall national average (27%) but
Class-wise, poor Muslims are more likely to be is also the highest among responses in the similar
highly afraid of the police (11%) than any other category by other caste groups in rural areas. OBC
class of Muslims. respondents from rural areas are almost as likely to
be highly fearful as rural SC respondents (18% and
While overall figures for Muslims are not much 16% respectively). However, this picture changes
different from other communities, we however significantly in the case of urban areas. Whereas
notice that the fear within the community is Upper Castes continue to be the least likely to be
informed by the concentration of their population afraid of the police (37% in urban areas said they
in a particular state. In states which have very high were not at all fearful), SC responses to fear are very
Muslim concentration, the highly fearful percentage
Table 5.3: Muslims most fearful where their numbers are low
States categorised by Highly fearful of Somewhat fearful of Not much fearful of Not at all fearful of
Muslim population police police police police
Very high population 7 25 37 26
High population 7 44 26 21
Moderate population 11 36 20 29
Low population 20 33 36 9
Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:1826 (Only Muslim responses).
The states have been recategorized according to the presence of Muslim population. Very high concentration category has Assam,
Kerala and West Bengal; high concentration states are Bihar, Uttar Pradesh, Jharkhand Uttarakhand; Moderate concentration states
are Gujarat, Karnataka, Maharashtra, Rajasthan, Delhi and Telangana; Low concentration states are Andhra Pradesh, Haryana,
Himachal Pradesh, Madhya Pradesh, Nagaland, Odisha, Punjab, Tamil Nadu and Chhattisgarh.

high, with 22 percent having reported high levels of most likely to feel highly afraid of the police and
fear. This figure is only 12 percent for urban Other formal higher paying occupations the least. While
Backward Classes. Thus, in understanding police on the whole, 16 percent of the people in agriculture
fear, it is important to situate the individual in a have said they are highly fearful of the police, 12
particular social and geographical milieu, which percent and 10 percent of the professional and
would give us a better understanding of how social managerial class respectively feel so. Within these
status coincides with fear. occupational categories however, a clear pattern is
difficult to establish. This is a crucial exercise in that
To probe the relationship between one’s social and
it tells us that the fear of police seems to function
economic situation and its relation to the fear of
independent of general class considerations; class
police excesses, we tried to ascertain which particular
looked at through the prism of occupation, caste
class would be most vulnerable to it. Upon probing
or community does seem to have some determining
by class, education and gender, we did not see any
power in this context.
significant distinction in the responses. Probing with
occupational categories, however, we do see some In Figure 5.5 we see a location-wise distribution of
variation in the responses. For instance, people in fear of police. People in villages are 5 points higher
the informal occupations such as agriculture are on the highly fearful scale than those living in cities.
Figure 5.4: Fear within the Hindu community: Upper Castes least fearful of the police
38
35
33
27 27 28
26
23 22 23 24
22
17 18
12
10
Upper Castes OBCs Scs STs

Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:12380. Caste-wise classification is
based on the responses received.
Figure 5.5: Region-wise distribution: Villages most fearful of the police

40
33
29
27 27
24 25
22 21
15
13
10

Village Town City
Note: Figures are percentages. Rest of the respondents did not respond to the question..

As the locality increases in size and urbanity, the fear population in the state, we again see that a clear
also seems to recede. This could suggest that people picture doesn’t emerge vis-a-vis fear. In states where
living in villages have a more negative perception of the percentage of SCs in prison is high (‘Bad’ states),
police than their urban counter-parts. the fear is the lowest, whereas the same figure for
Muslims is very high for Muslims and STs. Since no
5.2.2. Police statistics and fear of the police discernible pattern can be identified across Muslims,
Police statistics are available to us on the themes SCs and STs, it seems that here too police statistics
of representation of various communities in the do not provide an explanation for the variation in
force and the percentage incarceration of different levels of fear.
communities. On categorising states based on
Since these classifications provide us with little
these, we came up with a relative classification of
explanation to the variations and patterns that we
states—whether they were very good/good or bad
see in the responses, in the following sub-section
depending on their performance in the respective
we try to find associations between the attitudes of
categories. However, relating these classified states
fear and people’s perception of the police in other
with responses about fear of police didn’t bring
matters such as diversity or discrimination. In
out any significant patterns or general trends. For
doing so, we seek to get a fuller understanding of
instance, the actual numbers of SCs in the police
what invokes and increases fear and effects people’s
force had no effect on fear outcomes. Thus, there
perception towards policing.
was no association between fear levels of people
from states with high representation of SCs in the 5.2.3. Association between sense of
police force and fear levels of respondents of those discrimination by police and fear of it
states where number of SCs in the police force was
There seems to be a high degree of association
low. The same lack of relationship between actual
between fear of the police and how discriminatory
numbers of one’s community in the force and ones
one perceives it to be. We see in Table 5.4 that as
level of fear held true for even STs and Muslims.
the discrimination perception increases, so does
On classifying states according to the level of the likelihood of being afraid of possible police
incarceration of backward classes in relation to their brutality - one is more than twice as likely (31%)
Table 5.4: Fear highest when one believes police is highly discriminatory
Discrimination Perception Highly fearful of Somewhat fearful Not much fearful of Not at all fearful of
Index police of police police police
Highly discriminates 31 35 19 15
Discriminates 13 32 24 29
Doesn’t discriminate much 9 27 31 29
Doesn’t discriminate at all 13 25 20 28
Note: Figures are percentages. Rest of the respondents did not respond to the question. The Discrimination Perception Index has been
computed by combining responses to questions about police discrimination (for methodology, see Appendix).
Figure 5.6: Experience has a significant effect on fear- bad experience leads to more fear.
51
35
29
22 24
19
11
8

Had to pay a bribe when I Did not have to pay a bribe when
contacted the police I contacted the police
Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:2126. ‘Yes’ n=727. ‘No’ n=1058.
Question related to bribery was: (If contacted the police recently) During the last 4-5 years, whenever you contacted a police officer or
visited the police station, did you have to pay a bribe to get your work done?

to be highly afraid if one believes that the police is fearful is 15 percent. This figure jumps to 36 percent
highly discriminatory in nature than if one believes among respondents who know of many cases of
it isn’t (13%), which perhaps goes to show that one’s police brutality. Thus, it seems that the level of
level of confidence has a significant effect on one’s awareness does have an effect of how fearful one
perception and attitude towards the police. feels of the police.
Since there are not many ways to probe a possible Among SCs who believe that the police is often
harrowing experience that a person might have gone complicit in falsifying charges against them, the
through, we have here taken bribery as a proxy for level of fear (40% highly fearful) is more than double
bad experience with the police. In Figure 5.6 we those who believe that police doesn’t frame SCs. In
see that the experience of having paid a bribe has a fact, among those SCs who believe police never
significant effect on one’s levels of fear. If one has implicates their community under false charges, 41
paid a bribe in the past one year, 19 percent report percent say they are not at all fearful of the police
to being highly fearful, whereas if one hasn’t paid a (Table 5.6).
bribe the figure is 8 points lower (11%). Similarly,
Strong agreement or disagreement with police
people who have not had to pay a bribe are much
implicating Muslims in false terrorism charges
more confident and nearly four times more likely to
seems to significantly influence one’s level of fear.
be completely unafraid (29%) than those who have
If one fully denies that it happens, the high fear
paid a bribe (8%).
percentage among Muslims is 9 percent, but when
In Table 5.5, we have tried to ascertain where one’s one fully agrees with the vulnerability of the Muslim
awareness about police excesses informs fear or not. community towards falsified terrorism charges, the
The table seems to suggest a clear relation between high fear percentage increases more than two times
the two—in cases where a person knows some cases to 20 percent (Table 5.7).
of police violence, the likelihood of being highly
Table 5.5: Relation between awareness and fear of police: fear increases with awareness
Awareness about police Highly fearful of Somewhat fearful Not much fearful of Not at all fearful
brutality police of police police of police
Know of quite a few cases 15 46 29 9
Know of many cases 36 43 10 10
Don’t know of any case 13 25 22 33
Note: Figures are percentages. Rest of the respondents did not respond to the question. The Awareness Index is made by combining
responses for questions which probe whether a person knows of a victim of police brutality For methodology on how the Index was
constructed, see Appendix 3
Table 5.6: SCs very fearful of police if they believe it falsely implicates their community
False implication of Scheduled Highly fearful of Somewhat fearful Not much fearful Not at all fearful of
Castes by the police police of police of police police
Strongly believe it happens 40 33 12 15
Somewhat believe it happens 16 42 22 16
Don’t much believe it happens 11 31 34 21
Don’t believe at all it happens 17 18 23 41
Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:2214 (SC responses only).
Question asked was police often implicates SCs under false charges - do you agree or disagree? (Probe further whether fully or somewhat)
Table 5.7: Muslims who believe police falsely implicates their community in terror related cases are
more fearful of it
False implication of Muslims Highly fearful of Somewhat fearful Not much fearful Not at all fearful of
by the police police of police of police police
Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:1824 (Muslim responses only).
Question asked was- Police often implicates Muslims in false terrorism charges. Do you agree or disagree? (Probe further whether
fully or somewhat)

As in the case of Muslims and SCs, in the case of fear invoked by the police. However, these do
of STs too we see that one’s perception of unfair not tell us about the implications this fear could
accusations about Maoist activities leads one to be have on the interaction of citizens with the police.
highly fearful of the police. Among STs who fully Considering how important a part the police plays
disagreed about false accusations being levelled in the functioning of modern states and in the
against them by the police, the proportion of highly daily interaction of the state with its citizens, it is
fearful respondents is 6 percent, but among STs who imperative to delineate the result of fear with its
fully agree with it, the figure is 20 percent, which is causes. In this section we gauge the nature and
three times more (Table 5.8). extent of the effects of fear on the overall perception
of police among the people and the likelihood of
It therefore seems to become clear that in
approaching the police, if the need arises.
understanding and examining the fear of police
brutality and unfair convictions, it is important to Fear and negative perception feed on each other and
see how the police interact not with individuals but so are closely associated. Among people who are
with the whole community. Responses from SCs, highly fearful of the police, the level of extremely
STs and Muslims, who are often the most vulnerable negative perception of the police (for methodology
sections of the population, tell us that the fear is on how the Index of perception was constructed,
largely informed by how they perceive the police to see Appendix 3) is the highest, while positive
behave with them1. perception of the police is the lowest (Table 5.9). In
terms of their perception about police biases as well,
5.2.4. Effect of fear on perception about we find a significant association between fear and
independence, autonomy and partiality perception of discriminatory attitudes by the police
In previous sub-sections, we have tried to identify (Table 5.10). For methodology on how the Index
and evaluate the possible constituents of the sense was constructed, see Appendix 3).
Table 5.8: STs who believe police falsely implicates them are three times more likely to be highly fearful
of it than those who don’t
False implication of Highly fearful of Somewhat fearful Not much fearful Not at all fearful of
Scheduled Tribes by the police police of police of police police
Note: Figures are percentages. Rest of the respondents did not respond to the question. sample size:1172 (ST responses only).
Question asked: STs are falsely implicated in false Naxalism charges. Do you agree or disagree? (Probe further whether fully or
somewhat)
Table 5.9: Highly fearful respondents most likely to have a negative perception of the police
Fear of police Index Extremely negative Somewhat negative Somewhat positive Extremely positive
perception of police perception of police perception of police perception of police
Highly fearful of police 23 19 37 20
Somewhat fearful of police 14 15 51 20
Not much fearful of police 16 15 40 28
Not at all fearful of police 15 14 32 38
Note: Figures are percentages. Rest of the respondents did not respond to the question. The Index of Perception of Police has been
made by combining responses about questions dealing with public’s perception of police For methodology on how the Index was
constructed, see Appendix 3.
Table 5.10: Fear increases the level of partiality people perceive in the police
Fear of police Index Doesn’t Doesn’t Discriminates Discriminates a lot
discriminate at all discriminate much
Not at all fearful of police 8 14 72 5
Note: Figures are percentages. Rest of the respondents did not respond to the question. The Index of Perception of Discrimination by
the police has been made by combining responses about questions dealing with public’s perception of discrimination by the police For
methodology on how the Index was constructed, see Appendix 3.

People who are most fearful of the police are also section we try to gauge people’s attitudes towards
least likely to want to approach them in times of police brutality - whether they are critical of it or
need (Table 5.11). They also feel the need for lesser do they condone and accept violence towards the
police presence in their localities compared to those accused and the criminals. The answers to the
who are less fearful (Table 5.12). Thus, it becomes question asked in this respect reveal ambivalence
apparent that fear has a significant effect on how the tilted more in favour of police using violence (Figure
police is perceived by the people. 5.7).
5.3. Attitudes towards police brutality In Table 5.13, a state-wise summated score that was
arrived at after compressing responses of people to a
The previous section tried to analyse fear of police question that measured their acceptability of violent
among citizens and factors associated with it. In this police behaviour towards criminals/custodial
Table 5.11: People afraid of the police are less likely to approach them
Fear of police Index Won’t approach Will definitely Will probably Have no other
police if a problem approach police if a approach police if a option but to
requires help problem requires help problem requires help approach
Not at all fearful 3 81 7 3
Note: Figures are percentages. Rest of the respondents did not respond to the question. Question asked: In the future, if you have a
problem that requires police help, would you go to the police?
Table 5.12: Fear leads to a demand for lesser police presence.

Fear of police Index Prefer greater police Prefer lesser presence in Prefer no change in police
presence in my area my area presence in my area
Highly fearful of police 45 22 23
Somewhat fearful of police 48 16 28
Not much fearful of police 46 13 31
Not at all fearful of police 51 9 32
Note: Figures are percentages. Rest of the respondents did not respond to the question. Question asked: What kind of police presence
would you like to see in your village/area- greater, less or no change?
Figure 5.7: Half the respondents condone the use of violence on criminals in police custody
‘There is nothing wrong in the police being violent towards criminals'
16% Fully agree

21%
Somewhat agree
14% Somewhat disagree
Fully disagree
29%
20% No Response
Note: Rest of the respondents did not respond to the question. Question asked: There is nothing wrong in the police being violent
towards criminals. Do you agree or disagree? (Probe further whether fully or somewhat)

violence. The table is arranged in descending order, scores, here too it is difficult to establish similarities
meaning that state which has the least acceptability between states which are at the top or the bottom
towards police violence is ranked highest, and state of the table, suggesting the need to disaggregate and
with highest acceptability is ranked lowest. analyse this data based on some other parameters.
From this table we see that states at the bottom of The responses elicited seem to be uniform across
the table such as Tamil Nadu and Gujarat have a class. The poor are as likely to condone or reprove
high acceptance of police brutality, meaning that police violence as the upper class (21% of the poor
they are most likely to agree with the statement ‘fully agree’ and among the upper class, this figure
that “there is nothing wrong in the police being is 22%). Similarly, we find little to no variation in
violent towards criminals”. In states at the top responses across educational levels. There is not
of the table, these attitudes are reversed and we much difference between those who are unlettered
see that they are more likely to disagree with the and those who have a college degree in this context;
above statement. In fact, the negative sign depicts whatever little difference that exists can be attributed
that number of respondents who condoned police to the propensity of unlettered people to not respond,
violence outnumbered those who were against it in which is not observed in the more educated groups.
that particular state; there are only four states where
Some variation does exist in particular caste/
a majority seem to reject police violence—Himachal
religious groups with respect to their attitudes
Pradesh, Odisha, West Bengal and Nagaland. Like
towards police violence. For instance, Christians are
in the case of the first table demarking the ranked
Table 5.13: State-wise arrangement of scores showing attitudes towards police violence
“There is nothing wrong in the police being violent towards criminals”
Rank State Fully agree Somewhat agree Somewhat disagree Fully disagree No response Score
1 Himachal Pradesh 11.8 20.5 40.9 13.2 13.6 2.3
2 Odisha 9.4 20.6 23.2 19.0 27.7 2.2
3 West Bengal 12.0 15.0 15.0 21.0 37.1 1.8
4 Nagaland 12.2 36.4 19.3 28.4 3.8 1.5
5 Uttarakhand 24.8 19.6 11.6 25.2 18.8 -0.7
6 Telangana 13.6 29.8 19.9 13.1 23.7 -1.1
7 Karnataka 18.5 29.6 23.5 14.9 13.5 -1.3
8 Chhattisgarh 16.0 22.9 13.7 13.9 33.5 -1.3
9 Uttar Pradesh 17.6 26.6 24.7 11.7 19.4 -1.4
10 Jharkhand 12.8 43.6 18.2 18.2 7.2 -1.5
11 Madhya Pradesh 19.7 30.0 20.8 12.5 17.0 -2.4
12 Maharashtra 19.8 36.0 21.4 14.3 8.6 -2.6
13 Assam 24.7 32.4 20.0 14.6 8.4 -3.3
14 Haryana 30.7 25.3 31.3 10.6 2.1 -3.4
15 Punjab 19.9 27.7 14.3 7.8 30.3 -3.8
16 Bihar 33.3 23.3 28.0 12.0 3.3 -3.8
17 Andhra Pradesh 21.6 34.8 17.6 9.1 17.0 -4.2
18 Delhi 32.0 28.6 12.5 18.7 8.2 -4.3
19 Rajasthan 21.6 35.7 13.0 7.5 22.3 -5.1
20 Kerala 32.0 29.2 11.6 15.3 11.9 -5.1
21 Gujarat 26.7 31.4 19.2 5.8 16.8 -5.4
22 Tamil Nadu 31.7 29.8 14.2 11.6 12.6 -5.6
Note: The state rankings for the question (Please tell me if you agree or disagree with this statement – There is nothing wrong in the police being
violent towards criminals (If agree or disagree probe further whether fully or somewhat) are based on summated scores that were arrived at
after weighting each response option. An ‘fully agree’ answer was weighted as -0.2, a ‘somewhat agree’ answer was weighted as -0.1,
a ‘somewhat disagree’ answer was weighted as 0.1, and a ‘fully disagree’ answer was weighted as 0.2. The category of no response
(those who did not answer the question) was weighted as 0 and hence excluded from the ranking analysis. A higher summated score
here indicates a greater disagreement with police violence.

the least likely to condone police violence towards people in urban areas, which perhaps shows that
criminals (17% fully agreed with the statement), they are less likely to condone such acts of violence.
which is 8 points less than the Muslims who are most
likely to condone it (25%). Hindus seem to have a
5.3.1. Condoning police violence: citizen’s
high percentage of agreement, but also a high level responses
of disagreement with police violence. In the case of Demographic variables in the previous section help
caste and community-wise distribution, we find that us identify some particular qualities which inform
STs are the least likely to accept police violence at attitudes towards police violence. In the following
11%, and are significantly lower than the national section we try to further identify social attitudes
average, which is 21%. But further disaggregating which have an effect on one’s acceptance/rejection
any of these categories into class categories ceases of such violence.
to show a pattern.
A clear pattern emerges when we associate sympathy
In terms of locality those living in rural areas are for police with acceptability for its violent behaviour.
much less likely to condone police violence towards As sympathy for police working conditions increases,
criminals (20% fully agreed with the statement) so does the likelihood of agreeing with custodial
than those living in urban areas (25%; Figure 5.9). violence. If one believes the working conditions
Even though their responses in the other categories for police are difficult, the agreement percentage
are similar to their urban counterparts, in the ‘Fully is 58 percent (‘fully’ and ‘somewhat’ combined),
agree (condone)’ category they are 5 points behind which is 8 points higher than the average and 21
Figure 5.8: STs least likely to agree with police violence

38
35
31
28 29 28
24 25
22 23 22 22
21 20
19 19 19 18
16 17 16
14 15
13 12 12
11
7
Upper OBC SC ST Muslims Christians Sikhs

Caste
Fully agree Somewhat agree Somewhat disagree Fully disagree
Figure 5.9: People in rural areas less likely to accept police violence towards criminals
25
Fully Agree 20
30
Somewhat Agree
28 Urban
18 Rural
Somewhat Disagree
20
13
Fully Disagree
14
Note: Figures are percentages. Rest of the respondents did not respond to the question. Question asked: There is nothing wrong in the
police being violent towards criminals. Do you agree or disagree? (Probe further whether fully or somewhat)

points higher than if one doesn’t have a sympathetic doesn’t seem to be dictating the level of agreement/
attitude towards the police. disagreement with police violence and even though
condoning attitudes seem to be falling as the level
There is no clear relation between level of crime and
of violent crimes decreases, a parallel movement in
acceptance of police violence. The rate of violent
the ‘agree’ category is not observed. Thus, it would
crime comes to us from police statistics and is
appear that a significant relationship between the
based on actual crime rates from NCRB. Crime rate
two does not exist.
Table 5.14: Sympathy for police likely to influence attitudes towards police brutality
Sympathy towards police working conditions Index “There is nothing wrong in the police being violent
towards criminals”
Fully Agree Somewhat Fully Somewhat
Agree Disagree Disagree
Does not believe police works under stressful conditions 15 22 17 17
Somewhat believes police works under stressful conditions 19 30 22 14
Strongly believes police works under stressful conditions 27 31 20 14
Note: Figures are percentages. Rest of the respondents did not respond. Sympathy for police Index was made by combining responses
which gauged the level of sympathy for police with respect to their working conditions (for methodology on how it was constructed,
see Appendix).
Figure 5.10: Few people have reported awareness of cases of police violence: Maximum know victims
of police torture
Know of a victim of police torture 13
Know a victim of police firing/lathi charge 12
Know of a victim of fake encounter 7
Know a woman who was a victim of sexual harassment 7

by the police
Know anyone who’s been taken into wrongful custody 9
by the police
Know anyone who died in police custody 6
Note: Figures are percentages. Rest of the respondents did not respond to the question. Responses under ‘Know a few cases’ and
Know quite a few cases’ have been merged under ‘Yes’.
Question asked: Do you know of anyone who was affected by the following (as mentioned above).
Figure 5.11: Awareness of police excesses: Majority report being unaware of police violence
2%
Awareness of Police Excesses Index
7%
Don’t know of any cases
16% of police excesses
Know of a few cases of

police excesses
Know of quite a few

cases of police excesses
75%
Know of many cases
Note: For details on how the Index was constructed, see Appendix 3.

5.4. Awareness about police excesses of ‘Know of quite a few cases’. For instance, states
such as Bihar, Telangana, Jharkhand and Karnataka
As is visible from Figure 5.11, in most cases three- have high figures in the last category (know quite a
fourths of the respondents were not aware of any few cases), which are a significant deviation from the
such cases. Of the remaining who were aware, even norm observed in other states. The low sample size
fewer were aware of more than a few cases. Thus, does not allow us to make further disaggregation,
the sample size reduces at each level of awareness, but this finding is important in itself in that it shows
which does not allow for further meaningful analysis the level of awareness must have a parallel story of
to take place. When checked against demographic excessive violence here. NCRB data reports a total
variables of region/caste/class and education, little of 66 cases of custodial deaths in 2015, of which
variation was found in this regard and they have as many as 14 came from one state—Maharashtra.
therefore not been reported. However, in a sample Yet the reported level of awareness about police
survey based on randomised sampling, it is quite violence is not very high here. Similarly, Madhya
likely that one isn’t able to capture people who Pradesh reported the highest number of complaints
would have personally heard about such crimes. against police personnel (10089 out of a total of
In the Table 5.15, which shows us a state-wise 41424 cases in 2015), but only a relatively lesser
distribution of the Index of awareness about police number of people have reported knowing more than
excesses (made by combining questions shown in a few cases of police violence (even though more
Figure 5.10; to know how the Index was constructed, have reported knowing a few cases). This tells us
see Appendix 3), some important differences that the relation between actual cases that end up
appear. For ease of analysis, the category of ‘Know getting reported and people’s awareness of police
of many cases’ has been merged within the category excesses is somewhat ambiguous.
Table 5.15: People in Jharkhand are most aware about quite a few cases involving police violence
States Don’t know of any police Know of a few cases of police Know of quite a few cases of
excesses excesses police excesses
Andhra Pradesh 77 15 8
Assam 82 12 5
Bihar 56 26 18
Gujarat 94 5 1
Haryana 87 12 1
Himachal Pradesh 98 2 0
Karnataka 55 22 24
Kerala 83 16 2
Madhya Pradesh 39 48 13
Maharashtra 70 23 7
Nagaland 75 13 13
Odisha 85 9 6
Punjab 94 5 1
Rajasthan 88 10 2
Tamil Nadu 81 12 7
Uttar Pradesh 63 29 8
West Bengal 83 14 3
Delhi 89 9 2
Jharkhand 61 12 27
Chhattisgarh 94 5 1
Uttarakhand 97 3 0
Telangana 48 29 23
Note: Figures are percentages. The awareness of police excesses Index was computed from the individual questions about awareness
of police excesses. For details on how the Index was constructed, see Appendix 3. “Know of many cases” has been added to “know
quite a few cases”.

5.5. Conclusion to levels of sympathy for the police, which could
suggest why people with sympathetic attitudes
Fear of the police is greatly predicated on some towards police personnel are also more likely to have
particular social demographics. We found that a more positive perception of the police, as different
incidences of fear were lowest among people living parts of the report mention.
in metropolitan cities, among people belonging to
upper castes and among upper caste people in rural Throughout this report we have made an effort
areas. Communities which were vulnerable and to understand the relation of social variables
fearful in one context might not feel so in another to attitudes towards police and policing in the
situation: thus, the levels of fear among Muslims country. The thrust of this report has been to report
was dependent on their levels of concentration in citizens’ perception of the police, and how certain
the area. In states where population concentration perceptions can contribute to the overall image of
of Muslims is higher, their levels of fear are lower the police. In doing so, we find that some interesting
whereas in states with lower concentration, the findings have emerged in this regard, which have
levels of fear are seen to be higher. challenged conventional notions of policing, even
as other findings have reiterated oft-perceived mores
Experiences with police and perceptions about and tropes of Indian society. Most importantly,
their attitude have a very significant effect on how these findings are grounded in survey-based social
fearful one feels. Among people who believe the science research, which finds few precedents in the
police is discriminatory in its practices, the levels of available scholarship on this subject in the country.
fear are fairly high. When one adds the particular This adds not only to the saliency of this report, but
respondent’s caste/communication to this question, also to the credibility of research. The gap between
such as asking SC respondents if they believe the intention and outcome in the matter of police reform
police discriminates against them, the responses to is one that needs to be fulfilled with availability of
fear of such respondents who agree with this are meaningful research which talks of the complexities
likely to be much higher than those SC respondents that are involved in an institution such as the police.
who believe the police doesn’t discriminate against This report is an effort to address this lacuna and
them. add to the vibrancy of debate on this issue.
In reversing the relation between fear and perception, References
we see that a high level of fear of the police has
significant effects on how one perceives the police, Chevigny, Paul. Edge of the Knife: Police Violence in
on people’s perception of police autonomy and Americas. New York: NY:New Press, 1995.
impartiality. Thus, understanding fear is useful Cohn, Steven, Steven E Barkan, and William A
not only to understand the relation of particular Halteman. 1991. “Punitive Attitudes Toward Criminals:
communities to the police, but also to realise that Racial Consensus or Racial Conflict?” Social Problems
it might have significant implications on overall 38(2):287-296.
perception and likelihood of contact. Jackson, Pamela. Minority Group Threat, Crime and
After understanding the fear of police in this detail, Policing. New York: Praeger, 1989.
it is interesting to note the high levels of popular Snyder, Benjamin. Policing the Police: Conflict Theory
agreement that police violence against criminals and Police Violence in a Radicalized Society. MA Thesis,
enjoys. While there is significant variation across University of Washington, 2013.
states, generally it can be stated that a majority Watch, Human Rights. “Bound by Brotherhood: India’s
of respondents seem to condone it instead of Failure to Killings in Police Custody.” 2016.
disapproving of it. Such approval has close ties
Wide-ranging research on this subject in the US has established how race has significant effects on levels of trust, satisfaction and
1
perception of the police (Chevigny 1990; Cohn et al 1991; Jackson 1989). Similar studies, especially in the context of fear, are not
present for India. Responses of SCs, STs and Muslims reported here, however, do suggest that marginality and vulnerability can play
an important role in relations with the police.

CHAPTER 6:
PEOPLE’S PERCEPTION ON
DIFFERENT ASPECTS OF POLICING
Police officers presented with vehicles to control the law and order situation in Noida
(Credits: Virendra Singh Gosain, Hindustan Times, 9 September 2017, Noida)

People’s Perception on Different Aspects of Policing
6.1. Introduction The chapter also seeks to understand which social

cleavages produce differences in the way police
This chapter considers some of the factors that go is perceived. In a deeply stratified society such
into shaping people’s perception of police. A wide as India, it is of immense importance to arrive at
range of studies and literature on public perception an understanding of how individual and group
of police have established the primacy of personal identities interact with state institutions. The
experience in shaping one’s perception of the chapter is organised as follows: the first section
police. These experiences are related to not only analyses people’s perception of police using a set
the personal interaction or relationship that one has of questions; the second section probes people’s
had with the police, but also the consequences of opinion on the extent of police independence;
police action on an individual’s life. For instance, it the third section unpacks people’s empathy levels
is believed that the effectiveness of police presence towards the police and its working conditions, and
in one’s area has an effect on how people perceive the fourth section delves deeper into the complex
the police. However, another set of variables that relationship of gender and policing.
have been found to be consequential in shaping the
perception on police deal with the particular socio- 6.2. People’s perception of the police
cultural position of individuals. Thus, class and
caste have been found to be important signifiers of We attempted to analyse people’s perception of the
the variation in people’s perception of the police. police by constructing a comprehensive Index that
took into account their responses to 9 questions that
A 1977 report by the National Police Commission, probed various aspects of perception - their trust
India on the perception of people about the police levels in the police, their assessment of corruption
strongly condemned the general view of the police, in the police, their opinion on police bias and their
stating that “the egregious features of police are sympathetic attitude towards the police. These
politically oriented partisan performance of duties, questions along with the answers they elicited have
brutality, corruption and inefficiency, degrees been listed below –
of which vary from place to place and person to
person”. Since then, empirical research on Indians’ Q1. Please tell me how much trust do you have in
attitudes toward the police has been relatively scant, local police like police inspector, Sub inspector,
sporadic and descriptive in nature. The present SHO – a lot, somewhat, not much or not at all?
survey seeks to fill the gap in understanding citizen’s Q2. Please tell me how much trust do you have in
perception by analysing the findings on public senior police officer like SP, DCP - – a lot, somewhat,
perception of police in 22 states of India. not much or not at all?
In this chapter, we have used questions which Q3. Please tell me how much trust do you have in
probed citizens’ perception of the police along traffic police - – a lot, somewhat, not much or not
parameters of trust, honesty and fairness to arrive at at all?
an overall understanding of how policing and police
Q4. On a scale of 10 points where the 1st point on
personnel are viewed by common citizens. Such an
the left stands for extremely corrupt and the 10th
exercise has far-reaching implications not just for
point on the right stands for not at all corrupt,
the self-image and confidence of police personnel,
where would you place the local police, i.e., police
but also for how the public chooses to interact with
inspector, Sub inspector, SHO, beat constable, in
the police.
terms of corruption?
Table 6.1: Trust levels lowest for traffic police

Trust a lot Trust somewhat Don’t trust much Don’t trust at all
Local police 23 48 14 7
Senior police 29 40 15 5
Traffic police 16 34 23 14

Q5. On a scale of 10 points where the 1st point on where would you place the traffic police in terms of
the left stands for extremely corrupt and the 10th corruption?
point on the right stands for not at all corrupt, where
Q7. Do you think the police intentionally implicate
would you place the senior police officer, i.e., SP,
people under false charges?
DCP in terms of corruption?
Q8. Please tell me whether you would agree or
Q6. On a scale of 10 points where the 1st point on
disagree (and to what extent) with –police is blamed
the left stands for extremely corrupt and the 10th
unnecessarily even when it does its job well?
point on the right stands for not at all corrupt,
Table 6.2: Senior police thought to be least corrupt

Very corrupt Corrupt Not much corrupt Not at all corrupt
(Rated 1-2 on scale) (Rated 3-5 on scale) (Rated 6-8 on scale) (Rated 9-10 on scale)
Local police 27 34 22 11
Senior police 17 33 29 16
Traffic police 27 30 25 11
Figure 6.1: Nearly 1 in 6 feel police implicates under false charges
16%
20% Police implicates people under false
charges
Police perhaps implicates people under
false charges
Police does not implicate people under
27%
false charges
37% No response
Figure 6.2: 54% feel police is blamed unnecessarily despite doing its job well
15% 13%
Fully agree with the statement
Somewhat agree with the statement
12%
Somewhat disagree with the statement
Fully disagree with the statement
41% No response
19%

Q9. Please tell me whether you agree or disagree in total only 33 percent of the respondents displayed
with this statement –police is corrupt, it does not do an overall negative perception of the police. An
its job without a bribe? (Probe further whether ‘fully’ overwhelming majority (66%) were found to hold
or ‘somewhat’ agree or disagree) a positive perception of the police (26% had a very
positive opinion and 40% had a somewhat positive
The response categories for each question listed
opinion; Figure 6.4).
above were assigned scores that ranged from 0 to
4, following which the scores of all questions were Disaggregating the perception of police across
summed up to arrive at a comprehensive perception demographic variables reveals interesting variations.
Index (for detailed methodology, see Appendix 3). Despite accusations by many of a gender unfriendly
The Index results revealed that almost one-in-five approach of the police, men and women do not
respondents (18%) hold a very negative opinion of differ much in their perception of police. While
the police, while about one in seven (15%) have a 68 percent of men were found to have a positive
somewhat negative opinion of it. This means that perception (strong and moderate combined), among
Figure 6.3: 55% feel that the police is corrupt and does not do its job well
14%
25% Fully agree with the statement
Somewhat agree with the statement
11%
No response
20%
30%
Figure 6.4: Index of perception about the police-majority display a positive perception
1%
18% Very positive perception

26%
Somewhat positive perception
Somewhat negative perception
15%
Very negative perception
Non-committal
40%
Note: See Appendix 3 for methodology of Index construction.

women the same assessment was just four points People’s perception of the police also appears to be
less at 64 percent (Figure 6.5). affected by their economic status. Various studies
have found economic situation to have a significant
An analysis of perception by the respondent’s level
effect on people’s opinion of the police. Our survey
of education throws up a clearer pattern. Non-
also attests to this view, as is visible from Figure
literates were found to be twice as likely to have
6.7. Between the two extreme points of the class
a very negative perception of the police as the
hierarchy - the upper class and the poorest - there is
college educated (25% as opposed to 13%). In fact,
a seven-percentage point difference as far as a‘very
a highly negative perception of the police decreases
negative’ perception of the police is concerned (14%
as people’s level of education increases (see Figure
among the upper class and 21% among the poorest).
6.6).
There is an even higher difference of 11 points in
Figure 6.5: Female perception of police: not very different from male perception
27
Very positive
25
Somewhat positive 41
39 Male
15
Somewhat negative
16
Female
16
Very negative
19
Figure 6.6: Non-literates most likely to have a negative perception of the police
44 42
38
35
27 29 27 25
22
16 14 17 15 16 18
13
Very positive Somewhat positive Somewhat negative Very negative
College educated Up to matriculation Up to primary Non-literate
Figure 6.7: Poorer individuals more likely to have a negative perception of the police
41 40
39 39
33
27 28 27
22 21
17 17
13 15 15 14 15

the ‘very positive’ perception category - while 33 negative perception of the police, followed by
percent of upper class respondents were found to Muslims. Christians held the most positive opinion
carry a very positive perception, among the poorest (see Figure 6.8).
respondents this outright positivity was way less at
Examining state wise opinion on how police is
22 percent.
perceived by people is an important comparative
In India, caste and class often interact in definitive exercise that helps us in understanding the diverse
ways. The responses for police perception on the range of public opinion with respect to policing. We
basis of caste also follow the logic of class-based weighted each response category of the Perception
perception: Hindu upper castes generally reflected about the Police Index and arrived at a summated
a more favourable opinion of the police than other score for each state (see note below Table 6.4 or the
Hindu communities. Hindu Scheduled Tribes were Appendix 4 to know more about methodology). On
found to carry the most negative opinion followed doing this we found Punjab, Uttar Pradesh and West
by Hindu Scheduled Castes (see Table 6.3). Within Bengal to be the bottom three states when it comes
castes however, class comes back into play: with to police perception, implying that in these states the
the exception of Scheduled Tribes, in all other level of negative perception of police is the highest
communities the poor were more likely to hold a (see Table 6.4). The negative sign for states towards
negative opinion of the police and the upper class a the bottom of the table signifies that a majority of
more positive one. people have a negative opinion of the police, instead
of a positive one. The states with the most positive
In terms of religion, Sikhs (mostly from Punjab in
perception of police are Haryana, Himachal Pradesh
our achieved sample)were found to have the most
and Jharkhand.
Table 6.3: Scheduled Tribes most likely to hold a negative perception of the police
Upper Castes 31 39 15 14
Other Backward Classes 23 41 17 18
Schedule Castes 26 37 16 19
Schedule Tribes 27 38 14 21
Figure 6.8: Sikhs most likely to hold a negative perception of the police
42 43
40
36 34
32
27
21 20
16 17 16 18 15
12 10

Table 6.4: Haryana and Himachal Pradesh have the most positive perception of police
Rank State Very positive Somewhat positive Somewhat negative Very negative Score
1 Haryana 70.9 22.4 3.8 2.9 15.5
3 Jharkhand 46.7 42.5 4.6 5.6 12.0
4 Nagaland 37.2 47.0 6.4 9.5 9.6
5 Kerala 41.4 37.7 10.3 9.5 9.1
6 Bihar 30.3 48.3 13.0 8.4 7.9
7 Uttarakhand 29.6 43.6 14.1 12.0 6.5
8 Andhra Pradesh 24.9 44.5 15.3 14.7 5.0
9 Karnataka 25.7 44.7 12.3 17.2 4.9
10 Assam 24.0 45.7 13.7 16.5 4.7
11 Maharashtra 25.8 40.0 15.9 16.6 4.3
12 Telangana 19.9 44.8 16.9 15.8 3.6
13 Odisha 26.1 37.7 11.2 23.9 3.1
14 Madhya Pradesh 18.7 41.8 25.4 14.2 2.5
15 Gujarat 25.6 34.2 18.5 21.7 2.4
16 Delhi 22.2 34.9 20.8 21.8 1.5
17 Tamil Nadu 15.9 40.2 26.0 17.8 1.0
18 Rajasthan 14.5 39.7 12.4 29.5 -0.3
19 Chhattisgarh 18.3 32.8 16.8 31.5 -1.0
20 West Bengal 15.7 35.6 13.9 31.9 -1.1
21 Uttar Pradesh 8.2 37.2 27.6 26.2 -2.6
22 Punjab 9.4 37.8 17.8 34.6 -3.0
Note: Figures are percentages. The state rankings for The Index of perception about the police (see Appendix 3 for more details on
how the Index was constructed) are based on summated scores that were arrived at after weighting each Index category. The ‘very
negative’ category was weighted as -0.2, the ‘somewhat negative’ category was weighted as -0.1, the ‘somewhat positive’ category was
weighted as 0.1, and the ‘very positive’ category was weighted as 0.2. The category of non-committal (those who did not answer any
question that went into making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher summated score
here indicates a positive assessment, i.e., a better perception of the police.
Perception of the police seems to determine whether 6.3. People’s opinion onpolice
a person is likely to seek the police’s help in the future
if the need arose. Four in every five (80%) of those
independence
who were found to hold a very positive perception People’s view on police independence is another
of the police said they would go to the police in the aspect that we tried to probe in the survey through
future if they have a problem that requires police a series of questions. Respondents were asked about
help. This figure drops to 72 percent among those the extent of interference in the police’s work in
who hold a somewhat positive perception of the their locality by politicians, local goons, senior
police, to 68 percent among those with a somewhat police officers, economically powerful groups
negative perception and to 66 percent among those and influential castes in the area. Their responses
with a very negative perception of it (see Table 6.5). reveal that the greatest interference in the police’s
Table 6.5: People with a negative perception of the police least likely to seek police help
Perception about Open to seeking police’s help Reluctant to seek police’s help Won’t at all seek police’s help
police in the future in the future in the future
Very positive 80 13 4
Somewhat positive 72 20 6
Somewhat negative 68 22 7
Very negative 66 21 7

functioning is seen to be coming from political thought it was highly independent and 41 percent
parties and politicians (39%), followed by senior thought it was somewhat so). At the other end, only
police officers (24%) and economically powerful 34 percent thought it was not much independent
groups (23%). Local goons and dominant castes and 4 percent were of the opinion that it was not
were not seen to be influencing police’s work as independent at all (see Figure 6.9).
much (see Table 6.6).
We weighted each response category of the Index to
Based on the responses to all these questions on arrive at a summated score for each state (see note
interference in police’s work by various actors, below Table 6.7 or the Appendix 4 for methodology
we constructed an Index of perception of police on how the weightage was done). We found
independence (see Appendix for methodology respondents in Kerala to have the most positive
on how it was constructed). We found over half perception on the issue of police independence. In
the respondents to be under the impression that other words, the respondents here saw the police
the police in their area works independently (12% as being far more independent in its functioning as
Table 6.6: Extent of interference in the police’s functioning in one’s area

A lot of Somewhat Not much Not interference
interference interference interference at all
Political party 39 36 9 6
Local goons 14 33 24 15
Senior police officers 24 38 15 8
Economically powerful 23 33 18 12
groups
Influential and dominant 14 28 22 16
caste of the area
Question asked: Often the police says that it is not allowed to work independently and free from political interference. Please tell me
how much interference do these groups have in the functioning of the police in your locality – a lot, somewhat, not much or not at all?
Figure 6.9: Nearly half the respondents likely to perceive police as independent
Perception of police independence Index
9% 12%
4%
Highly independent
Somewhat independent
Not much independent
Not independent at all
34%
Non-committal
41%

compared to respondents in other states. Odisha people in positions of power and authority can
fared the worst (see Table 6.7). wield great influence on the police, 45 percent of
the respondents hold a very negative opinion of the
Perception of police independence and the
police (see Table 6.8). This is an important finding
perception of the police in general are correlated
since it shows us that perhaps the biggest roadblock
and the former seems to be informing the latter in
to improving perceptions lies in improving police
a significant way. Among those who considered
independence.
police to be very less independent, implying that
Table 6.7: State rankings on the question of police independence

Rank State Not at all Not much Somewhat Highly Score
independent independent independent independent
1 Kerala 2.1 12.4 14.4 64.8 12.7
2 Assam 2.4 17.3 52.2 20.6 7.1
3 Nagaland 7.5 19.5 42.1 24.0 5.6
5 Gujarat 2.3 30.3 42.6 17.1 4.2
6 Haryana 0.2 36.5 43.2 17.0 4.0
7 Maharashtra 3.3 28.3 52.2 10.8 3.9
8 Rajasthan 1.4 29.6 45.0 9.8 3.2
9 Bihar 1.2 41.7 42.7 11.8 2.2
10 Jharkhand 6.2 32.0 52.0 7.0 2.2
11 Andhra Pradesh 7.3 27.6 51.2 6.3 2.2
12 Karnataka 1.7 43.8 37.9 12.8 1.6
13 Uttarakhand 2.9 40.7 44.9 7.9 1.4
14 Madhya Pradesh 1.8 42.9 51.0 3.7 1.2
15 Tamil Nadu 0.3 44.9 32.4 10.7 0.8
16 Telangana 3.6 39.6 42.1 6.4 0.8
17 West Bengal 11.4 27.3 32.8 12.0 0.7
18 Uttar Pradesh 2.7 43.6 39.3 8.0 0.6
19 Chhattisgarh 13.9 23.2 36.3 7.2 0.0
20 Punjab 4.5 43.2 42.8 3.7 -0.2
21 Delhi 4.8 48.6 33.9 5.4 -1.4
22 Odisha 9.8 40.4 29.4 3.1 -2.4
Note: The state rankings for The Index of Perception of Police Independence (see Appendix 3 for details on how the Index was
constructed) are based on summated scores that were arrived at after weighting each Index category. The ‘not at all independent’
category was weighted as -0.2, the ‘not much independent’ category was weighted as -0.1, the ‘somewhat independent’ category
was weighted as 0.1, and the ‘highly independent’ category was weighted as 0.2. The category of non-committal (those who did not
answer any question that went into making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher
summated score here indicates a positive assessment, i.e., a better perception of the police’s independence.
Table 6.8: Negative perception of police greatest among those who don’t see police as independent
Highly independent 42 36 12 10
Somewhat independent 30 45 15 11
Not much independent 22 42 18 18
Not independent at all 14 27 12 45

6.4. Sympathy for the police 6.10). Respondents in urban areas are more likely to
acknowledge that the working hours of the police
Policing is a relentless activity, requiring efficient are greater than respondents in rural areas (55%
service delivery round-the-clock. Police are required as opposed to 46%). Three out of five respondents
to be available for work 24 hours a day, seven (59%) also acknowledged that it is difficult to work
days a week - a gruelling reality for constables in the police force due to high stress levels and long
and other low-ranking officers in. In addition, the working hours (see Figure 6.11). These questions
problem of personnel shortage and heavy workload are important because they probe a key aspect of
disproportionately affects the investigation of crimes, public-police relations and reflect that while citizens
registration of complaints/FIR thereby affecting may not be content with the functioning of police
complainants’ satisfaction. To ascertain people’s in India, this does not necessarily translate into
understanding of this challenge and complexity of diminished levels of sympathy towards police
policing, their opinion on the demanding nature personnel and recognition of systemic challenges.
of policing was sought. Approximately 1 in two
persons (49%) acknowledged that the working In the survey we asked another question that tried
hours of police officers are greater compared to to indirectly probe people’s sympathy levels towards
other occupations, over one-fifth considered them the police. Respondents were asked about what they
to be the same as other jobs and 1 in ten stated that thought was hindering proper police functioning –
the working hours of the police are less (see Figure whether it waslack of training or other resources or
Figure 6.10: Urban dwellers more likely to acknowledge that working hours of the police are greater
22
Overall 10
49
Same as other jobs

23
Rural 11 Less
46
Greater
21
Urban 8
55
Question asked: In comparison to other jobs, are the working hours of the police greater or less?
Figure 6.11: 60% believe police personnel work under high stress level
‘Compared to other professions, it is difficult to work in the police force
because of high stress levels and long working hours'
14%
24%
Fully agree with the statement
8% Somewhat agree with the statement
19%
No response
35%
Note: Question asked: Please tell me whether you agree or disagree with this statement - Compared to other professions, it is difficult
to work in the police force because of high stress levels and long working hours?

whether it was because the police were lazy and not under tough conditions. On the other hand, eight
motivated to serve people. Their responses to this out of ten respondents either strongly or moderately
question were far less sympathetic. Nearly half the believed that police working conditions are tough
respondents said that what hindered proper police (Figure 6.13).
functioning was the police personnel’s laziness and
Overall sympathy levels for the police’s working
unwillingness to discharge their duty in the correct
conditions appear to be informed by the particular
manner. Only 29 percent (3 in ten) believed that
socio-economic position of the individual: locality,
the police doesn’t function properly due to lack of
education and caste have significant implications
training and other resources.
for this. Urban respondents were more likely (seven
We clubbed all the three questions (responses to percentage points more to be precise) to be fully
which are shown in figures 6.10, 6.11 and 6.12) sympathetic than rural respondents (Figure 6.14).
together to construct an Index which gave us an Similarly,in terms of education, the difference
overall picture of the sympathetic attitudes towards between ‘full sympathy’ for the police working
the police (see Appendix for details on how it conditions among college educated respondents
was constructed). Overall only one in every seven and non-literate respondents was of 14 points (see
respondents (13%) was found to be unsympathetic Table 6.9). In terms of caste Scheduled Castes and
towards the police working conditions. Such people Scheduled Tribes from the Hindu religion were
were of the opinion that the police does not work found to be least likely to be fully sympathetic
Figure 6.12: Half the respondents believe that police is lazy and unmotivated to serve the people
22%
29%
Police doesn’t function properly
due to lack of training and other
resources
Police are in fact lazy and not
motivated to serve the people
No opinion
49%
Note: Question asked: Which of these two statements do you agree with? (1) Police is not able to function properly due to lack of
training and other resources. (2) It is not that the police lack resources, they are in fact lazy and not motivated to serve people. Do you
agree with 1 or 2?
Figure 6.13: 4 out of 5 believe that police work under tough conditions
5%
13%
Strongly believes police works
under tough conditions
42%
Somewhat believes police
works under tough conditions
Does not believe police works
under tough conditions
40%
Non-committal
Note: See Appendix 3 for details on how the Index was constructed.

Figure 6.14: Urban respondents more likely to be fully sympathetic than rural
47
40 41 40
14
10
Fully sympathetic Somewhat sympathetic Not sympathetic towards

the police
Rural Urban
Table 6.9: Sympathy for the police increases with improvement in the literacy levels of respondents
Fully sympathetic Somewhat sympathetic Not sympathetic towards the police
Non-literate 35 38 18
Up to Primary 38 44 13
Up to Matric 45 41 11
College and above 49 40 10
Figure 6.15: OBC respondents most likely to have a fully sympathetic attitude towards police
45 43
42 41
39 38 37 37
18
15 15
11
Upper Castes OBCs Scs Sts
Note: Figures are percentages.Rest of the respondents did not respond.
towards the working conditions of the police(see fifth highest sympathy whereas,Kerala recorded
Figure 6.15). Hindu Upper castes and OBCs were sixth highest sympathy. Sympathy for the police was
found to be highly sympathetic. In fact, urban, lowest in Uttar Pradesh and Chhattisgarh (see Table
college educated Hindu OBC men were more likely 6.10).
than any other similarly situated caste group to
While sympathy for police personnel is an
have a sympathetic attitude towards the police, with
important area of analysis, what is more significant
64 percent of them showing full sympathy for the
is its possible correlation with people’s perception
police. In terms of religious communities, we find
of police, which then gives us a more nuanced
Christians to be most sympathetic and Muslims to
understanding into what might be informing
be least (see Figure 6.16)
people’s attitudes towards the police. In a similar
An interesting pattern emerges when we analyse vein, people who were more sympathetic towards
the sympathy towards police across different states. the working conditions of police personnel were
After assigning weights to each Index category and significantly more likely to hold a positive opinion
arriving at a summated score, we found that the top about them. Of the respondents who were fully
four states with the greatest amount of sympathy sympathetic to police condition, 31 percent held a
towards the police’s working conditions are all very positive perception of the police; on the other
southern states – Telangana, Andhra Pradesh, end of the spectrum, a near reversal of this takes
Karnataka and Tamil Nadu. Assam recorded the place: among those who were least sympathetic, 31

Figure 6.16: Christians most likely to hold fully sympathetic attitude towards police
52 50
42 42 41
40
38 37
14 13
9 6

Table 6.10: Ranking of states on sympathy for police’s working conditions

Rank State No sympathy Moderate sympathy Strong sympathy Score
1 Telangana 3.0 24.8 70.6 43.3
2 Andhra Pradesh 6.6 30.2 60.2 40.5
3 Karnataka 3.0 42.1 54.1 40.3
4 Tamil Nadu 5.3 36.0 56.7 40.2
5 Assam 6.4 36.6 55.7 40.1
6 Kerala 7.0 35.1 54.3 39.1
7 Maharashtra 5.8 34.1 55.2 39.0
8 Bihar 9.0 40.7 49.6 38.8
9 Nagaland 9.1 42.7 45.5 37.4
10 Delhi 14.3 41.0 44.2 37.3
11 Punjab 7.0 52.7 38.1 36.3
12 Uttarakhand 16.1 43.9 37.4 35.1
13 Jharkhand 4.4 63.4 29.2 34.5
14 Odisha 20.6 43.5 30.8 32.6
15 Gujarat 23.6 44.2 27.8 31.9
16 Rajasthan 23.7 34.2 33.5 31.8
17 Madhya Pradesh 20.2 38.1 32.3 31.6
18 Haryana 15.3 68.2 15.5 31.3
19 West Bengal 13.2 37.4 34.7 31.2
20 Himachal Pradesh 32.5 32.3 30.0 31.2
21 Chhattisgarh 16.4 42.3 29.7 30.8
22 Uttar Pradesh 27.4 53.0 14.1 28.4
Note: The state rankings for The Index of Sympathetic attitude towards police working conditions (see Appendix 3 for details on how
the Index was constructed) are based on summated scores that were arrived at after weighting each Index category. The ‘no sympathy’
category was weighted as 2, the ‘somewhat sympathy’ category was weighted as 3 and the ‘strong sympathy category was weighted
as 5. The category of non-committal (those who did not answer any question that went into making the Index) was weighted as 0 and
hence excluded from the ranking analysis. A higher summated score here indicates more sympathy.

Table 6.11: Perception depends on sympathy levels for police
Those who hold Those who hold a Those who hold a Those who hold
a very positive somewhat positive somewhat negative a very negative
perception perception perception perception
Those who are fully 31 44 14 11
sympathetic
Those who are 25 43 16 16
somewhat sympathetic
Those who are not 21 28 18 31
sympathetic
percent held a very negative perception of the police even consider a career in uniform or law enforcement
(see Table 6.11). to begin with due to their misunderstanding of the
nature of the job, resistance from family, or the
6.5. Female police officers: The aggressive and authoritarian images portrayed
gendered aspect of policing in the media. However, once hired, women face
discrimination, sexual harassment, or even peer
In India, norms around the appropriate role of intimidation, and they often lack the necessary role
men and women and gender-based stereotyping models or mentors to help them move up the ranks.
in labour market inhibit women’s entry for certain
occupations such as the police force, army. Like In recent discussions on safety of women and gender
the Indian army, the police force also continues equity within the police force, it is a broadly accepted
to define itself along gendered lines and suffers assumption that the presence of women personnel
from the twin deficits of diversity and design. is likely to make police forces more gender-just,
There is a strong belief that combat, by nature, is help impact patriarchal social norms and gendered
a male occupation; that the police force is a male hierarchies operative in police stations. Following
domain and therefore unsuitable to the female the 1970s, a combination of societal changes and
physique and temperament. In addition to these progressive legislations paved the way for women
culturally privileged beliefs, the abuse and systemic to enter law enforcement. However, the diversity
marginalisation of serving policewomen, an deficit continues to be alarming: just about five per
overall dis-incentivising atmosphere, peer pressures cent (84,479) of the 16.7 lakh-strong Indian police
and gender issues mean that women are actively force is made up of women.
discouraged from joining the police force. While To test people’s perception and preferences, their
research shows that women can be just as effective responses on different parameters such as honesty
as men, uneven hiring practices, societal cultural and hard work of police officers and their overall
values, selection processes and recruitment policies preference towards policemen vs. policewomen
keep the number of women low. were gauged. A large plurality of respondents
According to a survey done by the Commonwealth stated that they consider police officers of both
Human Rights Initiative (2007), most women never the genders honest (43%), hardworking (46%) and
would be open to approaching both for help (51%).
Figure 6.17: Policewomen thought to be more honest, but respondents more likely to approach
Policemen
51
43 46
21 23
20
13 15 14
12 9
7
Police woman Police man Both Neither
Who is more honest Who is more hardworking Whom would you approach for help

However, if we compare the responses towards approach a policeman than a policewoman (20% to
policemen and policewomen specifically, it emerges 15%) for help (see Figure 6.17).
that respondents were more likely to consider a
Analysing responses on these parameters by the
policewoman more honest than a policeman (21%
gender of respondents showed that both men and
to 14%), a policeman more hardworking than a
women were more likely to consider policemen
policewoman (23% to 13%) and were more likely to
more hardworking and policewomen more honest.
Figure 6.18 Across demographic variables, preference for police officers of both the genders is visibly
high
23
Police woman
7
Women
16
Police man
26
48 Men
Both
53
6
Neither
7
Figure 6.19: Women more likely to believe that policewomen are more honest
10
Police woman
14
Women
13
Police man
15
41 Men
Both
45
24
Neither
18
Figure 6.20: Women more likely to think of policewomen as more hard-working
16
Police woman
10
Women
20
Police man
25
45
Both Men
48
9
Neither
10

With regard to approaching either of them for help, inflexible working hours (51%), it is not viable for
respondents of both the genders displayed preference women to join the police force (see Table 6.12).
for police of their gender, that is, women preferred Nearly an equal proportion of respondents affirmed
policewomen (23%) and men preferred policemen that women are incapable of handling high intensity
(26%) for help (see Figure 6.18). In contrast to women crimes and cases (41%) and they should prioritise
who were interviewed, men appear to be more open managing the home (39%) instead of joining the
to the idea of seeking help from police officers of police force. Respondents were most likely to justify
both the genders (53%). With respect to honesty of lack of physical strength and inflexible working
police officers, women were nearly twice more likely hours as reasons for women’s unsuitability in the
to hold a favourable opinion towards policewomen police force. However, they were most likely to not
than policemen, 24 percent as opposed to 13 percent agree that women should give priority to their home
(see Figure 6.19). Even among men a slightly over joining the police force.
greater proportion believed that policewomen were
Paradoxically, the gender of respondents does not
honest than policemen. Finally, as far as opinion on
appear to have any bearing on their responses to
who is more hardworking, both men and women
these questions. In other words, women were nearly
thought that policemen are more hardworking than
equally likely to hold the same opinion on why it is
policewomen;however, women were once again less
difficult for women to serve in the police force and
likely to believe so than men (see Figure 6.20).
prioritising home instead of working (Table 6.13).
The survey also sought to ascertain people’s views This suggests that significant primacy continues
on women and policing and test the extent of to be given to the idea of tying the traditional role
powerful stereotypes that view policing as ‘men’s of women to family, housework and nurturing
work’. Among all those who were interviewed, responsibility. The bias against women suggests that
over 1 in two persons agreed that due to lack of their competence to join the police force is evaluated
physical strength, aggressive behaviour (52%), and not in professional terms but weighed against their
Table 6.12: Responses to questions on women working in the police force

Women lack physical Women should Because of inflexible Women police
strength & aggressive prioritise managing working hours, officers are incapable
behaviour required in home instead of difficult for women of handling high
the police force joining the police to work in the police intensity cases &
force force crimes
Very justified 15 12 19 12
Somewhat justified 37 27 32 29
Somewhat unjustified 20 26 18 24
Very unjustified 18 25 16 21
No response 10 10 15 14
Table 6.13: Gender based responses on women working in the police force
Women should prioritise Because of inflexible working Women police officers are
managing home instead of hours, difficult for women to incapable of handling high
joining the police force work in the police force intensity cases & crimes
Justified Unjustified Justified Unjustified Justified Unjustified
Men 40 50 53 33 43 44
Women 37 52 49 35 39 46

traditionally suited role as mothers, wives and more likely to be favourable to having women in the
homemakers. police force than men (67% to 63%). However, if we
take into consideration only the ‘very favourable’
Taking into account the responses to all these
category, then women fared much better and were
questions (except for the one that deals with inflexible
ahead by six points (27% to 21%).
working hours) we constructed an Index that tried to
measure the overall favourability towards women in There was a significant age-divide that was seen
the police force (see Appendix 3 for methodology). among women. Younger women were far more
We found one-fourth of the respondents to be very likely to be open to having women in the police
favourable towards having women in the police and force than older women. Among men on the other
about two-fifths to be somewhat favourable (see hand, no such age-based pattern was seen (see Table
Table 6.14). However, a significant proportion (one- 6.14). Locality was also found to make a difference.
third) was against it. There wasn’t much of a gap Women in villages were least approving (only 24%
between how men and women fared on this Index. were very favourable to the idea) of having women
If we combine the ‘very favourable’ and ‘somewhat in the police, those in towns and small cities were
favourable’ categories then women were only slightly slightly more favourable (30% were very favourable)
Table 6.14: Overall favourability towards women in the police force

Very favourable Somewhat favourable Not much favourable Not favourable at all
towards women in towards women in towards women in towards women in
police force police force police force police force
Overall 24 41 31 3
Men 21 42 32 3
Women 27 40 29 2
Note: Figures are percentages. Rest of the respondents did not respond. For methodology, see Appendix 3.
Table 6.15: Young women more favourable towards having women in police than older women
Very favourable Somewhat favourable Not much favourable Not favourable at all
towards women in towards women in towards women in towards women in
police force police force police force police force
Women 18-25 years 32 43 23 2
Women 26-35 years 29 40 29 2
Women 36-45 years 27 43 26 2
Women 46-55 years 25 39 32 2
Women 56+ years 22 38 31 4
Men 18-25 years 23 45 29 2

Men 26-35 years 21 44 32 3
Men 36-45 years 20 43 33 4
Men 46-55 years 23 40 32 4
Men 56+ years 21 39 33 4

than them, and women in big cities were most a higher ‘very favourable’ figure than Jharkhand and
favourable to the idea (34% being very favourable). yet scores slightly less than it in overall terms. This is
This pattern holds with respect to men as well but because of two reasons –the ‘somewhat favourable’
not as strongly as it was among women. Men in figure in Jharkhand is double of what it is in Kerala.
cities were found to be more open to the idea than Moreover, unfavorability in Kerala is nine points
men in towns and villages (24% as opposed to 20%). higher than it is in Jharkhand. Similarly, Himachal
Pradesh despite recording a much higher ‘very
We also tried to see how different states did with
favourable’ figure (39.1%) than some other states
respect to this Index. This was done by assigning
ranks below those states because it also recorded
each response category of the Index different
a very high ‘not much favourable’ figure of 43.6
weights and then arriving at a summated score
percent as well.
(see Appendix 4 for ranking methodology). The
greatest favourability for the idea was found to be Respondents were also asked to share their opinion
Uttarakhand, Haryana and Maharashtra (in that on the possible reason(s) for women’s hesitation in
order). Telangana, Uttar Pradesh and Karnataka seeking help from the police and visiting the police
showed the lowest favourability across all states (see station. The responses of men and women were
Table 6.16). A caveat is in order here - Kerala reports found to be nearly identical on the question. Nearly
Table 6.16 State-wise favourability towards women in the police force

Rank Not at all Not much Somewhat Very favourable Score
favourable favourable favourable
1 Uttarakhand 1.7 11.1 39.8 47.4 12.0
2 Haryana 1.5 13.0 37.7 47.8 11.7
3 Maharashtra 1.5 14.9 37.0 45.2 11.0
4 Nagaland 0.0 15.8 43.3 40.7 10.9
5 Delhi 2.4 20.1 40.6 36.4 8.9
6 Jharkhand 1.6 15.0 64.0 19.4 8.5
7 Kerala 2.0 24.1 32.7 39.3 8.3
8 Punjab 1.6 21.5 53.5 23.2 7.5
9 Odisha 4.0 21.9 40.4 30.4 7.1
10 Tamil Nadu 2.0 30.5 44.8 21.7 5.4
11 Madhya Pradesh 0.3 34.2 43.7 21.5 5.2
12 Gujarat 1.9 29.0 49.8 17.4 5.2
14 Chhattisgarh 2.7 34.9 31.6 26.7 4.5
15 Assam 3.6 35.1 38.3 21.8 4.0
16 Bihar 1.2 33.7 57.4 7.4 3.6
17 Rajasthan 3.9 37.1 35.6 19.9 3.1
18 Andhra Pradesh 4.5 37.7 46.5 9.4 1.9
19 West Bengal 6.8 38.2 33.1 16.9 1.5
20 Karnataka 3.8 44.0 42.2 9.5 1.0
21 Uttar Pradesh 6.2 42.6 38.3 10.2 0.4
22 Telangana 5.1 48.1 33.7 9.9 -0.5
Note: The state rankings for The Index of favourability towards women in the police (see Appendix 3 for details on how the Index
was constructed) are based on summated scores that were arrived at after weighting each Index category. The ‘not at all favourable’
category was weighted as -0.2, the ‘not much favourable’ category was weighted as -0.1 the ‘somewhat favourable’ category was
weighted as 0.1 and the ‘very favourable’ category as 0.2. The category of non-committal (those who did not answer any question
that went into making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher summated score here
indicates more favourability.

Table 6.17: Fear of social stigma and harassment biggest reason why women don’t approach the police
Fear of social stigma, harassment & use of abusive language by the police 19
Unsafe for women- absence of policewomen, family doesn’t allow, not possible for women to visit the police 18
station repeatedly
Systemic issues- lack of speedy trials, poor conviction rates; absence of FIR/complaint registration, lack of 11
confidentiality
Behaviour of police officers- demands for bribe, partisan attitude, lack of complete knowledge of law and legal 8
procedure
Other reasons 7
No response 37
Question asked: Often women and young girls are scared to seek help from the police or visit the police station. In your opinion, what
is the main reason for this?
one in five persons stated that the fear of harassment,

use of abusive language by police officers and perceptions of the police. It can also be inferred that
general social stigma deters women from visiting the challenges of women’s safety, improved public
the police station in case of emergency (see Table accountability of police officers, greater reportage
6.17). These issues have also been documented in of crimes rests on an improvement in the fragile
the literature on the twin challenges of women’s police-populace relationship. While it is difficult to
safety and policing. encourage women to join the police force, cultural
attitudes towards women working as police officers
6.6. Conclusion will transform and acceptance will increase only
when more women are seen in the police force.
The aim of this chapter was to report the various
factors that go into informing one’s perception Undoubtedly, the relation between the two is
of the police. Given the salience of police as complex and informed by a multiplicity of factors—
an institution in the everyday life of citizens, it primarily the individual’s position in society and
is important to not only engage with reforms their past experiences with the police. Such exercises
regarding police performance, but also give due are often a common mode of enquiry in other
importance to citizens’ perception of the police. complex societies, with a heterogeneous population.
Our analysis suggests that people’s initial attitudes In a country like India, which is teeming with cross-
about the police play a critical role in shaping their cutting social cleavages and intersectionalities of
overall judgments, subsequent direct and indirect caste, gender, class and religion, people’s perception
experiences as well as their future attitudes. Also, can tell us a lot about the experience of policing at
negative experiences were associated with more the ground level and can go a long way in furthering
negative perceptions of the police, whereas positive the cause of reform.
experiences were associated with less negative

CHAPTER 7:
Analysing CAG Audit Reports:
(Not) Learning from Mistakes
Police escorting the polling officials taking EVMs to Hasnabad for voting. (Credits: Subhendu Ghosh,
Hindustan Times, 9 May 2001, West Bengal)

Analysing CAG Audit Reports: (Not) Learning from Mistakes
Both, the first chapter based on official data and the governments in power without matching action on
following chapters presenting people’s perceptions, the ground.
indicate that improving the police infrastructure
Conventional wisdom suggests that the money well-
may go a long way in improving performance and
spent on law and order should improve the quality of
perception. One way of addressing this aspect is
life for the common citizens by making the country
diligent, transparent and faithful deployment of
a safer place and by improving the performance of
available resources and budgetary allocations. The
the police force on the ground. And that is the main
crucial question is, does that really happen? The
logic as to why India needs to increase allocations
short answer is that our chronic maladies persist
on all aspects of police modernisation.
despite timely diagnoses by existing institutions.
However, our experience of expenditure on MPF
In this chapter we are collating and analysing critical
over almost half a century, belies the conventional
deficiencies which have been pointed out over a
wisdom simply because of lax monitoring by
long period of time by the Comptroller and Auditor
the states. This means that improved grants and
General of India (CAG) in dozens of audit reports
allocations do not always translate into improved
in different states of India. The CAG auditors have
action on the ground. And that is precisely what the
not only looked at the account books but have also
CAG of India has been saying in report after report
assessed performances and extensively interviewed
for decades.
lower level police officials to get to the bottom of
issues. CAG audit reports on the expenditure by the states
makes a depressing reading. It shows that in a
It is common, for instance, for the states to deploy
country where resources are scarce, improper use of
lower-level officers to the field without the requisite
money, pilferage and inefficiency leads to huge costs
training. Many states don’t even bother to train them
in terms of human lives and missed opportunities. In
in handling weapons and some don’t even have a
its audit report on the MPF programme in MP, the
firing range to train them in. The CAG reports are
CAG observed that questionable practices have put
full of insights but the attempt here is to collate
the policemen’s “own security in jeopardy”. It was
critical patterns or egregious violations by the states.
also scathing on the attitude of the UP government,
More details about specific area are available in the
which was indifferent in dealing with issues such
Appendix.
as police modernisation, maintenance of law &
This chapter covers the audit of police of 11 states order and security of citizens. The audit reports
– UP, Uttarakhand, Tamil Nadu, MP, West Bengal, from across the states show how the successive
Assam, Gujarat, Maharashtra, Bihar, Himachal governments have failed the ordinary policeman.
Pradesh and Rajasthan on the following, largely
In Uttarakhand, the auditors remarked that the
comparable, parameters - weapons, training,
police training college was not fully equipped, which
police stations and housing, forensic science labs,
would have a bearing on their ability to work in the
communications and vehicles along with the
field. The MP audit rued the exclusion of issues like
expenditure and financial management. The audit
gender sensitisation and custodial sensitivities in the
spans over the periods 2009-14, 2010-15 and 2011-
MPF training guidelines. The story appears to be the
16. It also includes audit findings on the management
same across states.
of prisons of Assam, Karnataka, Rajasthan and
Himachal Pradesh, the 4 states where prison audits The patterns seem to suggest that the manpower,
were conducted. funds and resource crunch challenges have almost
grounded the traffic police in almost all states. The
In India, police and law-and-order are state subjects.
Home departments seem to be blind to changing
But the central government also makes special
requirements of the Indian cities which are becoming
provisions for federal grants meant to be spent by
more and more chaotic by the day. In Uttar Pradesh,
the states for specific purposes. A good example
the vehicles increased by 2,256 percent in 30 years
is the centre’s allocation of Rs 25,061 crores for a
between 1985 and 2015, but the sanctioned strength
new umbrella scheme for the modernisation of
of traffic personnel remained the same; the shortages
police force (MPF) for three years, beginning in
range from 71 percent to 93 percent in the cadres
2017-18. Since its inception in 1969, the MPF has
of traffic inspectors, sub-inspectors and constables
witnessed many lofty announcements by successive
against the sanctioned strengths. The situation may

be extrapolated in almost all North and central depriving the district police of their sanctioned fleet
Indian states. of vehicles. Thus, the money which was earmarked
for increasing the efficiency of police by increasing
7.1 Poor and patchy Modernisation of the fleet was diverted to buy luxurious vehicles for
Police Force (MPF) the Chief Minister. In Rajasthan, despite a shortage
of operational vehicles the department bought 66
The MPF programme was initiated in 1969 to cars (Indigo, Ambassador, Swift Dzire, Maruti SX4)
augment critical police infrastructure to tackle during 2009-14 in violation of MPF norms.
the threat to country’s internal security. Despite
thousands of crores being pumped into the In West Bengal, senior police officers and bureaucrats
programme, not much headway has been made in attempted to divert funds (Rs.5.72 crores) meant for
the performance of the police, the CAG audits have construction of quarters for Constables and Head
noted. The ratio of the funding between the central Constables, Inspectors, Sub-Inspectors and Assistant
and the state governments under the MPF were Sub-Inspectors to build houses for 12 IPS officers,
75:25 until the year 2011-12, and after that, 60:40 which was a gross violation of the MPF guidelines.
for each year of 2012-16. The North-Eastern states After the CAG red-flagged the decision, the project
and Jammu and Kashmir were eligible for funding was dropped under unclear circumstances leaving
in the ratio of 90:10. no way of knowing whether the houses were
constructed or not. Thus, in spite of the initial
In order to access the central funds, the states are availability of funds, the construction of quarters for
required to prepare Annual Action Plans (AAP) junior officials was delayed by altering the proposal
which should flow from a central government to one for IPS officers, the CAG noted. This episode
approved five-year strategic plan. There is a paucity reflects the attitude of senior police officers towards
of funds, and whatever money is made available, not only the funds meant for specific purposes but
is frittered away due to the states’ administrative also towards the needs of their junior colleagues.
incapability or lack of capacity. This means that the
states refuse to learn and the policemen continue to In Maharashtra, CAG audit points out a shortage
suffer for no fault of theirs. In the implementation of 65,026 modern weapons which constitutes 45
of the MPF, the CAG audit has exposed these percent of the total state need. During an interview
deficiencies, particularly in terms of resources, lack with CAG, 122 out of 150 police personnel (of
of training and crumbling infrastructure. These, Madhya Pradesh), who were stationed in the
coupled with lack of ethical values and alleged bias crumbling buildings, cited the issues of water
against the poor and the vulnerable communities, seepage, space constraints, inconvenience, and
cry for a massive overhaul. In UP, a state particularly even risk to life. Due to a shortage of housing in
hard pressed for funds due to its large population, 80 selected 13 districts where audits were conducted,
percent (Rs.25.65 crore) of funds for procurement of police personnel occupied “683 condemned and 582
training equipment during 2011-16 were surrendered dilapidated houses” in Madhya Pradesh.
due to delay in finalisation of acquisitions. UP, Superintendents of Police, in 13 districts in MP
obviously, is not an isolated case. where audits were conducted stated that such
7.1.1 Apathy towards ordinary cops/citizens buildings caused problems like unsafe records/office
equipment and dangers to the lives of the personnel
The CAG audits have brought out instances of operating from them, besides cramped spaces, all
government and bureaucratic apathy towards round inconvenience, and seepage of water. MP
concerns of policing, particularly the challenges police department has not been able to provide
faced by an ordinary policeman. In UP, the shortage even separate toilets for its women constables. The
of Medium Police Vehicles (used for patrolling) was CAG was apprised by 48 out of 50 women police
as high as 68 percent in civil police and 75 percent in personnel about non-availability of separate retiring
Provincial Armed Constabulary (PAC) compared to rooms and washrooms, which apart from being huge
the BPRD norms. inconvenience and adding to insecurity have had an
Despite these critical shortages, the audit report adverse impact on their work and performance too.
revealed that an exorbitant amount of almost Rs 4 7.1.2 Financial inefficiencies, lack of capacities
crores was incurred on purchasing luxury vehicles
(Mercedes Model M-Guard) for the Chief Minister’s It took 17 years for successive DGPs in UP to
security, instead of Land Cruisers sanctioned earlier. prepare a response to the MHA’s 1995 proposal
The department also spent funds on 10 bulletproof for revision of the Arming Policy of state police,
Tata Safari and eight General Safari vehicles, again despite increasing sophistication of weaponry used
for the Chief Minister’s security. This was at the by terrorists, Maoists & criminal elements. The
cost of 18 condemned vehicles of 11 districts, thus CAG audit report adds that Rs 875.95 crore (28%)
from the UP state’s own modernisation funds were

surrendered despite facing critical shortages of in Uttarakhand, according to the report of 2011-
resources like police stations, arms and ammunition, 16. The situation was found to be similar in West
vehicles and forensic labs. The state police, which Bengal, where audit done in nine police stations
works on a depleted strength of less than 50 percent, have revealed the availability of only 26 (10%)
also failed to utilise 41 percent of Rs 1.165 crore quarters against the requirement of 254. Despite
allotted under the MPF in 2011-16. BPRD advocating residential accommodation for
all staff, the availability of the same was far behind
In MP, the progress of MPF fund expenditure against
the prescribed standards across states.
total available funds was slow for long but improved
largely during 2015-16 when 52 percent of funds 7.1.4 Poor mobility and a severe shortage of
(i.e. Rs 316.47 crore) were spent in what appeared police vehicles
to the auditors as a massive hurry, while the state
reeled under heavy shortage in all the segments. In A recurrent perception about the police, which needs
Bihar, as high as 71 percent of the funds remained to be corrected, is that they always reach the crime
unspent due to the delay in finalisation of tender scene late, almost as a practice. The CAG audits give
for procurement. Bihar was also deprived of central a clue to this laggard behaviour. Audits have brought
fund share of Rs 119.82 crore due to a slow pace of out extreme shortage of vehicles in all categories
expenditure. There were also grave issues of delay in across the states. In MP, there was a shortage of 71
the release of matching share essential for claiming percent or 14,107 motorcycles, 720 heavy vehicles,
the central funds in Bihar. 1382 four wheelers (21%) and still went on to buy
102 Sedan cars (Rs.5.88 crores) in violation of MPF
Rajasthan police could not utilise MPF funds guidelines.
ranging between 36 percent and 79 percent during
the period 2009-14. The audit report observed The BPRD norms call for availability of two light
that the department machinery was ill-prepared to motor vehicles and three motor cycles for each of
implement the scheme. In Maharashtra, the non- the police stations (PS).
utilisation was highest at an astounding 88 percent Bihar was able to procure only 531 vehicles out of
under equipment component. Massive fund under- 2,160 proposed. In Maharashtra, only 662 vehicles
utilisation too remained a concern in Assam, (2009- were procured during 2011-16, leaving a gap of 1,564
14) Himachal Pradesh (2011-16) and Gujarat (2009- vehicles (70%). In nine districts of Maharashtra,
15). there were 5955 vehicles and 5174 drivers against a
requirement of 8933 drivers. As a result, Hawldars
7.1.3 Acute shortage of police buildings
and Naiks (57%) who were supposed to perform
The financial mess is bound to be reflected in other policing duties were put into driving duties.
shortages of critical infrastructure requirements The shortage of vehicles was similar in Tamil Nadu,
across the states. In UP, only 9 percent of the total Uttarakhand, Rajasthan, Assam and other states.
requirement of residential quarters was sanctioned The availability of vehicles was better in only one
during the years 1998-2016. This amounted to state, Gujarat, with surplus two and four wheelers
5.156 quarters against the total requirement of compared to the minimum prescribed under BPRD
59.453 accommodations of different types. Only norms.
499 (47%) out of 1,064 of police stations in Bihar
have their own buildings. The 2011-16 audit also 7.1.5 Erratic communications network
observed a severe shortage of housing - 76 percent Technology-driven crimes have shot up to pose
for ‘upper subordinates’ and 83 percent for ‘lower a huge challenge to policing. To tackle new and
subordinates’. emerging threats like cyber-crimes, online hate
In Maharashtra, during 2011-16, only 8 percent of mongering and identity threats etc. policing needs
planned residential and non-residential buildings to equip itself with technology-driven solutions.
(including police stations) could be constructed/ It requires putting in place a countrywide
upgraded, even as there was an unspent balance of communication infrastructure across police stations,
Rs 205.76 crore at the end of September 2016 in this installation of CCTV cameras at key locations
component. The audit found that in eight selected and providing modern communication gadgets to
districts of Assam, only 36 residential buildings increase connectivity, reach and efficiency.
were constructed against the requirement of 2,748 The CAG audits reflect very poorly on technology-
up to March 2009. This meant a shortage of almost enabled solutions to policing. Uttar Pradesh could
99 percent (2,712) buildings during 2009-14. The not spend nearly 60 percent of Rs 136.51 crore it
situation was found to be similar in Tamil Nadu, received under the MPF for buying communication
Uttarakhand, West Bengal, Himachal Pradesh and equipment, even as about half of the force was
other states. Only 2,553 staff quarters (10.27%) deprived of hand-held sets as per the required
were found available by the audit for 24,863 staff norms. The audit noted that nearly two-third sets

in use have already passed their prescribed lives. was an overall shortage of 75 percent modern arms
As of December 2014, the Assam Cyber Crime during 2009-14 in Rajasthan. In five test districts
Investigation Lab (ACCIL) was non-functional of West Bengal the deficiencies were: 73 percent in
as it struggled with shortages like power-backup, SLRs, 90 percent in AK-47, 100 percent in grenade
broadband, equipment and manpower. During launchers, 100 percent in assault and sniper rifles and
2009-14 Assam CID was able to dispose only 1.82 54 percent in INSAS rifles and 100 percent in night
percent to 16.15 percent of the outstanding cases, vision sight for rifles. Similar situation prevailed in
while its sleuths were regularly pushed into law and Himachal Pradesh and many other states.
order duties. The Maharashtra Home Department
could spend under 50 percent of the grants (Rs 19.51
7.1.7 Severe shortage of forensic labs,
crore out of Rs 44.66 crore) for improving the state’s manpower
communication system. Policemen in Tamil Nadu Audit reports flag vacancies of 46 percent of Senior
too faced shortage of more than half of the required Scientific Officers, 34 percent of Scientific Officers,
quantity of hand-held sets and allied equipment. 48 percent of Lab Technicians, 67 percent of Lab
In Bihar, CCTV cameras bought in 2011-12, were Assistant and 61 percent of Lab Attendants in the
not found installed in eight out of 11 test-checked FSL in Madhya Pradesh. In one particular Regional
districts, and were kept in stores, audits of 2011-16 Forensic Science Lab (RFSL) in Jabalpur, the
have revealed. auditors found that all sanctioned posts were lying
The ambitious CCTNS (Crime and Criminal vacant as of March 2016.
Tracking Network and Systems), launched a year Bihar has one of the worst forensic infrastructures
after the Mumbai attacks of November 2008, in the country with only two regional labs. Even a
simplifies complaints mechanisms and generates smaller state like Andhra Pradesh has five RFSLs
real time crime data, besides connecting 15,000 and 18 MFSU (Mobile Forensic Science Unit) while
police stations and 5,000 offices of top police Odisha has 3 RFSL and 36 MFSU as of March
officials across the country (Chowdhury, 2015). The 20171. The department in Bihar also suffers from a
compliance rate in this crucial area is between 80 massive shortage of manpower with only 66 people
percent and 94 percent in Telangana, Maharashtra, posted in the state-level Forensic Science Lab (FSL)
Gujarat, Himachal Pradesh, Madhya Pradesh, in the capital Patna and the two regional forensic
Haryana, Karnataka, Tamil Nadu, Andhra Pradesh labs (RFSLs), against the sanctioned strength of 314
and Odisha. On the other hand, Bihar at 11.1 percent officials, the audit says. This has led to a colossal
lies at the bottom of the chart, as of January 2018, pendency in the labs. The establishment of four
according to CCTNS dashboard hosted on NCRB mobile units were delayed in Bihar for the reason
website. that the state failed to form an effective forensic
science governing body mandated by the Home
7.1.6 Shortage of weapons and obsolete
Ministry. On the basis of utilisation certificate
equipment
submitted by the body every six months, the central
Amid rising crime graph, Bihar continues to face government releases funds. The central funds to
a heavy shortage of modern weapons - 33 percent the tune of Rs.1.64 crore for the four forensic units
in INSAS rifle, 42 percent in Pistol, 36 percent were diverted to the account of Bihar electronic
in Carbine, 52 percent in AK 47. Due to a heavy corporation where it was lying till July 2016, the
shortage of modern weapons, the state police date of audit. This reflects administrative incapacity
continue to use outdated .303 rifles despite Home on the part of who’s who of Bihar police and state
Ministry orders to phase them out. administration.
The shortage of AK-47s was to the tune of 76 Twenty-six per cent vacant positions were found
percent in Gujarat (as per state’s own assessment). by the auditors in the FSL in Guwahati in Assam.
It does not have an adequate number of modern As against the sanctioned strength of 111, the men-
weaponry (36% shortage) which could compromise in-position were 82. Huge pendency was found in
the striking capability of the force, the audit report the FSL at Dehradun (Uttarakhand) due to non-
of 2009-15 noted. In Uttarakhand, against a availability of adequate workforce. Pendency in
sanctioned requirement of 2,221 modern weapons, Rajasthan forensic labs, due to the shortage of
only 867 (39.04%) were available. Despite the huge technical staff, varied from 37 to over 50 percent in
gap, only Rs 1.12 crore was spent for procurement various laboratories. In West Bengal, the shortage
of weapons, which constituted just over 2 percent of of technical workforce in the state forensic science
the total budget allotted under MPF 2011-16. There lab (SFSL) at Kolkata ranged between 38 and
1
According to the Directorate of Forensic Science Services website, as of March 2017, Bihar has 4 MFUs. The list of state forensic
labs, accessed on December 15th, 2017 was not accessible as the website become dysfunctional on a later date.

88 percent. At least 88 percent positions of lab that prisoners shall not be treated as nonpersons.
assistants to 50 percent positions of senior scientific The manual explains the rights of the prisoners
assistants were vacant as of July 2012. There was which comprises right to human dignity. It says that
only one functional biological department out of a prisoner is guaranteed the right to basic minimum
required seven at the RFSL at Jalpaiguri. needs such as adequate diet, health, medical care
and treatment, access to clean and proper drinking
7.1.8 Crumbling police training infrastructure water, access to clean and hygienic conditions of
Bihar has among the worst training infrastructure accommodation, sanitation and personal hygiene,
for its policeman in the country as required training appropriate clothing and bedding.
equipment was not found in any of the five selected
7.2.1 Assam
districts. Even the Constable Training School (CTS)
at Nathnagar (Bhagalpur) was found ill-equipped, The fund mismanagement story continues in Assam
audits of 2011-16 have revealed. Due to lack of prisons also. Despite the prison security facing
training for operation, bomb disposal suits (body challenges due to acute shortage of guarding staff,
armour designed to withstand pressure of a blast) and weapons, an insufficient height of boundary walls
non-linear junction detector (counter surveillance and absence of watch towers and security equipment,
device) remained idle despite availability. In while Rs.49.15 crore of funds remained unspent
Uttarakhand, PTC (police training college) at during 2011-16. At least 68 escapes were registered
Narendnagar tehsil (Tehri Garhwal district) did not during 2012-15. The audit found that in 2012-
have a firing range which is a fundamental part of 16, while 18 jails faced the issue of overcrowding
training, and there was a huge shortage of drinking of inmates, the registered capacities of 12 prisons
water. Only 5,000 litres of drinking water was given remained unutilised.
per day against the need for 36,000 litres, while the
Despite the issue of overcrowding, over 60 percent
rest were partially arranged from tanker supplies.
of the inmates were undertrial prisoners (UTPs).
In Rajasthan, according to state specified norms The Under-Trial Review Committees meant for
(August 2010), 2,772 pieces of equipment (fibre reviewing the duration of imprisonment of UTPs,
reinforced plastic helmets, poly carbonate lathis, 771 did not function in 15 jails during the audit period.
shields, 615 body protectors) were required in 35 test- Vocational training was imparted to only 4 percent
checked police stations. Of these only 422 (15.23%) and educational training to even less than 20 percent
sets of equipment were available, a shortage of of inmates during each of the last five calendar years
nearly 85 percent, audits of 2009-14 reveal. In West ending December 2015.
Bengal, live training was not imparted for handling
some of the weapons viz. LMG, 0.303 rifle, anti-
7.2.2 Rajasthan
riot gun, grenade and mortar, scrutiny in Swami CAG audits of 2011-16 reveal that in 12 Rajasthan
Vivekananda State Police Academy (SVSPA), jails, security equipment was found out of order and
Barrackpore, according to audits conducted between not provided in three sub-jails, even as incidents of
2009 and 2014. Similar situation prevailed across prisoner escapes increased during 2012-15 in the
the states. state. At least 323 inmates escaped in 275 incidents
during the period.
7.2 Prison audits (of Assam,
Unhygienic conditions prevail in prisons across the
Rajasthan, Himachal Pradesh and states. Prisoners’ stay in the jails gets prolonged
Karnataka) as authorities repeatedly fail to produce them in
courts due to non-availability of vehicles. The audit
The “Modernisation of Prisons” was launched by
disclosed that in nine out of the 16 test-checked
the central government in 2002-03 in 27 states on
units, there was non- availability of any vehicle for
a cost-sharing ratio of 75:25 between the central
transportation of prisoners to courts and hospitals.
and the state governments. The management and
Test checks of selected prisons for 2012-16, also
administration of prisons fall exclusively under the
revealed that in seven of these the mandatory
control of the concerned states which are governed
medical examination to be conducted at the time
by the Prisons Act, 1894, and Prisoners Act, 1900.
of admission was not carried out. In eight jails, 129
The Modern Prison Manual (MPM) prepared by the inmates died during 2012-16 due to diseases such
BPRD 2003 clearly says that a prisoner has the right as TB, AIDs and cancer. Out of these, 49 inmates
to be treated as a human being and as a person. The were infected during their term of imprisonment.
Supreme Court of India has categorically declared The audit clearly revealed that these 146 (49+97)

prisoners were perfectly healthy and free from any 7.3 Conclusion: CAG shows a
ailment or infection at the time of entering the jail.
Similarly, in seven jails, 97 inmates contracted TB in
method in our madness
the jail premises. This shows the pathetic condition The CAG audits of the MPF programme of the
of health services in the prisons. states show that the problem typically begins with
7.2.3 Himachal Pradesh the mishandling of funds, their allocation and
fulfilling of statutory requirements like making of
The state has not yet revised its prison manual as the mandatory Annual Action Plans (AAP), the five-
per MPM 2003 of the BPRD. The CAG audits of year strategic plans, and sharing them with the central
2013-16 revealed that 809 new prisoners admitted government for a smooth flow of resources. These
were not segregated on medical grounds, age include the procurement of weapons, construction
and behaviour. Of this, 456 prisoners found to be of buildings, purchase of vehicles, setting up of
suffering from various diseases like tuberculosis, forensic labs, providing training and provisioning
scabies and heart disease and renal problems were a range of equipment. Who ends up paying the
lodged with other prisoners in the barracks. Audit price of delays, inefficiencies or procedural lapses?
noticed that in two test-checked jails (Kanda and It is not just the common people or the lower-level
Mandi), 65 percent prisoners who were referred to officials who suffer but also the rule of law and the
government hospitals in 2013-16 could not be sent state of democracy in the country.
due to an insufficient police escort. The audit also
noted non-availability of medical officers in test- The audit reports show that most of the states
checked jails, essential for any kind of emergency. formulated AAPs without taking inputs from
Against four sanctioned posts of doctors in these district units which is a requirement. They routinely
jails, the audit could find only one officer posted in lost funds for not taking approvals for strategic plans
central jail, Kanda during (2013-16). Surprisingly, from MHA or underutilised the MPF funds. Assam,
no posts of doctors were earmarked for other jails in HP, MP, Maharashtra, Rajasthan, UP, Uttarakhand,
Himachal Pradesh. West Bengal and Gujarat too defaulted in these
aspects. Bihar could not spend 71 percent of the
The audit also revealed an acute shortage of sleeping funds, Assam used only 68 percent of the released
berths (or raised platform of 6.5 ft. X 2.25 X 1.5 ft.) fund, in Himachal, non-utilisation of funds ranged
for prisoners in all the five barracks of Bilaspur Jail. between 21 and 87 percent. In Maharashtra,
Such berths were not found in 21 barracks, out of there was 88 percent non-utilisation in equipment
42 audited barracks. Thus gross violations of rights component. These are only a few examples of the
of the prisoners as recommended under the Model pattern exposed by the CAG audit reports.
Prison Manual, 2003 was noted.
All the 11 states audited fully have severe shortage of
7.2.4 Karnataka police buildings but despite the availability of funds,
they have failed to build new police stations, barracks
CAG has noted that Karnataka is still managing
and houses. There was almost 99 percent shortage
with an outdated 1978 prison manual and has not
of staff quarters in eight selected districts of Assam,
yet revised it against the BPRD recommendations.
nearly 80 percent in Bihar, 88 percent in Himachal
The audit reports from 2010-15 revealed that many
Pradesh, 69 percent in Rajasthan, 48 percent in
electronic equipment (hand held metal detectors,
Uttar Pradesh, over 89 percent in Uttarakhand, 90
door frame metal detectors, deep search metal
percent in the test checked districts of West Bengal.
detectors, baggage scanners) procured were not
The percentage of utilisation of construction funds
in working condition, in a tech-savvy state which
in Maharashtra and Gujarat was only 8 percent and
prides itself in being India’s Silicon Valley. Worse
26 percent respectively. This indicates that laxity
still is that no action had been initiated to get them
and inefficiency could be as high a culprit for lack of
repaired, the CAG audit has noted.
adequate resources as the actual shortage of funds.
In test checked prisons before 2014, it was found
A substantial chunk of staff quarters in many states,
that out of the 58 cameras installed, 47 were non-
which were somehow made available, were found to
functional, of which 43 cameras were irreparable.
be dilapidated or in bad condition. Same thing can
The medical records of prisoners, such as the
be said about transportation and patrolling vehicles
prescription/case books, medical treatment registers
which need to be procured according to the norms
and hospital rolls were also not maintained regularly.
set up by the BPRD. Among the worst performers
Posts of doctors were not sanctioned in four out of
were Maharashtra and Bihar (70% and 75% shortage
five district prisons or sub-jails. There was no lady
in procurements of all vehicles respectively), UP
Medical Officer in any of the jails test-checked.

(68% shortage of patrolling vehicles) and MP (71% gov.in/sites/default/files/audit_report_files/Assam_
shortage of motorbikes). Report_1_2015_chap_3.pdf (accessed January 10, 2018)
In the forensic science infrastructure, the situation Comptroller & Auditor General of India. Report No. 01
is particularly worrisome where state after state is of 2015- Audit Report (General and Social Sector) for the
year ended 31 March 2014. Government of Rajasthan.
guilty of neglecting this crucial area of investigation.
2015. http://www.cag.gov.in/sites/default/files/audit_
In Bihar almost 80 percent of all positions in report_files/Rajasthan_General_Social_1_2015_Chap_2.
forensic labs are vacant, in West Bengal 88 percent pdf (accessed January 10, 2018)
of lab assistants’ position were vacant. Even in the
Comptroller & Auditor General of India. Report No. 01
states where the forensic infrastructure appears to
of 2016- Audit Report (General and Social Sector) for
be better, such as UP, almost 70 percent positions the year ended March 2015. Government of Karnataka.
are vacant. In the want of proper facilities, most 2016. http://agkar.cag.gov.in/docs/Consolidated%20
states depend on the Central Forensic Science Report%20(GSSA)%202015%20FOR%20PRINT.pdf
Laboratories (CFSLs) located at Chandigarh, (accessed January 10, 2018)
Kolkata, Hyderabad, Bhopal, Pune, Guwahati
Comptroller and Auditor General of India. Report No. 01
and New Delhi which work under the Directorate of 2017- Audit Report on Social, General and Economic
of Forensic Sciences Services (DFSS). As for the (Non- PSUs) Sectors for the year ended 31 March 2016.
preparedness of the police to deal with the rising Government of Assam. 2017. http://www.agasm.cag.
number of cyber-crimes, the less said is better. gov.in/forms/audit_report/2015-16/report1/NPSU.pdf
(accessed January 10, 2018)
The above examples show that virtually no lessons
are being learnt from the CAG audit reports which Comptroller & Auditor General of India. Report No. 01
are painstakingly done in a scientific manner. Under of 2017- Audit Report (Social, General, Revenue and
the circumstances, it seems that the best thing about Economic Sectors) for the year ended 31 March 2016.
Government of Uttarakhand. 2017. http://www.cag.gov.
CAG audits is that they are still taking place. The
in/sites/default/files/audit_report_files/Uttarakhand_
worst thing, of course, is that their meticulous Report_No_1_of_2017.pdf (accessed January 10, 2018)
endeavours do not seem to deter the corrupt or the
inefficient bureaucrats and political leaders from Comptroller & Auditor General of India. Report No. 02
continuing the business as usual at huge cost to the of 2014- Audit Report (General & Social Sector), 2013-
14. Government of West Bengal. 2014. http://www.cag.
nation and its people.
gov.in/sites/default/files/audit_report_files/Chapter_2_
A limitation of the analysis of CAG reports is Performance_Audit_20.pdf (accessed January 10, 2018)
that data on the different states were not found to Comptroller & Auditor General of India. Report No. 02
be across the same parameters, therefore making of 2016- Audit Report (General and Social Sector) for
comparison difficult in many cases. For instance, the year ended 31 March 2015. Government of Gujarat.
information regarding cyber-crime/communication 2016. http://www.cag.gov.in/sites/default/files/
infrastructure was not available for the states of audit_report_files/Gujarat_General_and_Social_Sector_
Rajasthan, West Bengal and MP on comparable Report_2_%202016.pdf (accessed January 10, 2018)
parameters. A detailed comparative CAG data for Comptroller & Auditor General of India. Report No.
all the relevant parameters for 11 states has been 02 of 2017- Audit Report (General, Social & Economic
incorporated in the Appendix 8. Sectors) for the year ended March 2016. Government of
Bihar. 2017. http://www.cag.gov.in/sites/default/files/
References audit_report_files/Bihar_Report_No_2_of_2017_on_
GSES.pdf (accessed January 10, 2018)
Bureau of Police Research and Development, Ministry
of Home Affairs. Model Prison Manual for the
of 2017- Audit Report (General and Social Sector) for the
Superintendence and Management of Prisons in India.
year ended 31 March 2016. Government of Rajasthan.
New Delhi: Government of India. 2003. http://bprd.
2017. http://agraj.cag.gov.in/themes/GSSA%20reports/
nic.in/WriteReadData/userfiles/file/5230647148-
Audit%20Report%20(G&SS)%202015-16-English.pdf
Model%20Prison%20Manual.pdf (accessed 13 September
2017)
of 2017- Audit Report on Social, General and Economic
of 2015- Audit Report on Social, General and Economic
Sectors (Non- PSUs) for the year ended 31 March 2016.
(Non- PSUs) Sectors for the year ended 31 March
Government of Himachal Pradesh. 2017. http://cag.
2014. Government of Assam. 2015. http://www.cag.
gov.in/sites/default/files/audit_report_files/Himachal_

Pradesh_Report_No_3_of_2017_Social_General_ 2017. http://cag.gov.in/sites/default/files/audit_report_
Economic_Sectors_Non_PSUs.pdf (accessed January 10, files/Report_No.4_of_2017_%E2%80%93_General_
2018) and_Social_Sector_Government_of_Maharashtra.pdf
of 2017- Audit Report on the General Social Sectors for Chowdhury, Sagnik. “CCTNS Project to Let Police
the year ended 31 March 2016. Government of Madhya Stations ‘Talk’: Where It Stands, and How It Can help
Pradesh. 2017. http://www.cag.gov.in/sites/default/ Fight Crime.” The Indian Express, November 20, 2015.
files/audit_report_files/Chapter_2_Performance_ http://indianexpress.com/article/explained/cctns-
Audits_MP_Report_3_of_2017.pdf (accessed January 10, project-to-let-police-stations-talk-where-it-stands-and-
2018) how-it-can-help-fight-crime/ (accessed February 07, 2018)
Comptroller & Auditor General of India. Report No. 03 Ministry of Home Affairs. “Modernisation of State Police
of 2017- Audit Report (General and Social Sector) for the Forces (MPF) Scheme.” Government of India. https://
year ended 31 March 2016. Government of Tamil Nadu. mha.gov.in/PMDivMPFScheme (accessed January 05,
2017. http://www.agtn.cag.gov.in/audit1/Reports/ 2018)
GSSA1516.pdf (accessed January 10, 2018)
Ministry of Home Affairs, “Modernisation of Police
Comptroller & Auditor General of India. Report No. Force Scheme,” February 06, 2018, http://pib.nic.in/
03 of 2017- Performance Audit of Modernisation and newsite/PrintRelease.aspx?relid=176238 (accessed
Strengthening of Police Forces for the year ended 31 March March 01, 2018)
2016. Government of Uttar Pradesh. 2017. http://www.
National Crime Records Bureau, “CCTNS Pragati
cag.gov.in/sites/default/files/audit_report_files/Report_
Dashboard as On 31.01.2018.” Government of India.
No.3_of_2017_Performance_Audit_of_Modernisation_
http://www.ncrb.gov.in/BureauDivisions/cctnsnew/
of_Police_Forces_Government_of_Uttar_Pradesh.pdf
CCTNS_Dashboard/Revised%20Consolidated%20
31.01.2018%20ver%202.1%20for%20MHA.pdf (accessed
Comptroller & Auditor General of India. Report No. 04 December 15, 2017)
of 2017- Audit Report (General and Social Sector) for the
year ended March 2016. Government of Maharashtra.

CHAPTER 8:
CONCLUSION
Clash between police and the mob near Tile Wali Masjid, Lucknow
(Credits: Trilochan Singh, Hindustan Times, 7 June 2001, Lucknow)

Conclusion
Public institutions are central to social and political police (Human Rights Watch 2009). At the same
life of modern societies. They play a crucial role in time, inefficient deployment of personnel, lack of
democratic governance and therefore it is expected basic crime investigation equipment and training,
that their functioning will result in goodwill and limited promotional opportunities for junior rank
legitimacy for the state. However, the Indian officers and relegation to menial tasks has left them
experience on the ground leaves much to be desired. demoralized and exhausted. Some of this comes
While institutions such as the Army have gained across rather clearly in the CAG reports on the
popularity and acceptability (SDSA 2007 & 2015), functioning of the state police forces, as detailed in
other formal bodies such as the police, elected Chapter Seven. The police reform agenda has also
representatives and even the judiciary have been failed to address the working and living conditions
viewed with increasing cynicism. In the context of of low-ranking police officers, who are often the
the functioning of the police in India, it is widely perpetrators of abuses, carrying out illegal orders
believed that they do not always cater to the interests or operating within a police culture and command
of common citizens and suffer from a systemic lack structure that condones and facilitates bad behavior.
of accountability. As a result, a perceptible trust In the rigid hierarchy of rank and function between
deficit has developed over the years. the superintendents, inspectors and the constable;
senior police officers drawn from the elite cadre
It is in this light that the Status of Policing in India
of the Indian Police Service have been relatively
Report 2018: A Study of Performance and Perceptions was
unaffected.
conceived. The report is an attempt to understand
the functioning of the police by placing people at Another dimension of this broader context is the
the centre-stage of their assessment. The report is a churning Indian society is witnessing, particularly
conscious effort to move away from a micro analysis in the past quarter of a century. This churning,
and it tries to offer the big picture regarding the characterized by new claims by backward
understanding of satisfaction and trust in the rule of communities and by new aspirations and tensions
law, police excesses, perception of discrimination, generated by the shift in the economic policies,
responsiveness of criminal justice system and has challenged pre-existing hierarchies and
levels of impartiality. The survey data from 22 produced extra-ordinary pressures on India’s public
states provides valuable clues on degrees of public institutions. The increased demands of being more
support, police-community relations and ascertain representative, accountable and efficient have also
the overall impact of the police force on the ground. made the public institutions more fragile. The law &
order machinery is no exception to this overall trend,
8.1 Context but perhaps is in the forefront of these contextual
dynamics placing by the greatest challenges before
Before we proceed to make sense of the empirical
it.
data, it would be useful to reiterate the broader
context of policing in India that informs people’s 8.2. Key findings from the perception
relationship, opinions and experiences. The Indian
policing system as a whole has colonial origins. The survey and official data on police,
British constituted colonial policing with the passage prisons and crime in India
of Indian Police Act, 1861, wherein the intent was
to establish a relationship of control, coercion and A survey of people’s perception of the police
surveillance over a subject population. The onset of warrants a comparative study of people’s attitudes
democracy after independence did little to change with the patterns and findings emerging from the
the structure of the police since the organization statistical data available from official sources. It
and rules established by the repressive colonial is for this juxtaposition that an evaluation of data
power continued business as usual. Numerous available from NCRB and BPRD was carried out.
police reform commissions have recommended While the analysis of official data does not directly
wide-ranging changes, but without effective correlate with the survey findings, both bring out
implementation. Today, organisational, political and nearly similar trends and patterns.
managerial deficiencies of the Indian police system The principal findings in this report have both
have severely inhibited its performance. Decades of challenged conventional notions of policing and
partisan functioning of police machinery, frequent also reiterated and endorsed popular beliefs and
failures to register complaints and investigate crime, generalisations. As discussed in the preceding
arbitrary detentions, torture and killings have chapters, it is clear that people’s perception and
resulted in significant public distrust and fear of the opinion of police are evenly split between positive

and negative assessments. Our analysis revealed lower concentration, the levels of fear were seen to
a peculiar dissonance surrounding policing: be higher. A large plurality of respondents stated
even as citizens’ continue to fear the police, they that the police remain impartial in case of an inter-
simultaneously show considerable satisfaction with community conflict, however, among those who
the rule of law. This puzzling coexistence of despair disagreed, Muslims were most likely to endorse the
and hope is illustrated through many findings: while view of police’s partiality.
a large number of people were willing to approach
Citizens’ responses also revealed an inversely
the police if the need arises, in the same breath they
proportional relationship between a person’s social
also shared that they do not expect police officers to
status, i.e., their class, caste, gender, etc. and their
give them fair and equal treatment.
levels of trust in the police. This perception can be
Presented early on in the report is the finding that rationalized through the findings from the official
a fairly small proportion of citizens contacted the objective data which exhibit a systemic bias against
police in recent past and a majority was satisfied vulnerable sections of the community- SCs, STs,
with the help received. Even those citizens who OBCs, women, children and the Muslims. This bias
experienced negative police contact (such as non- is evident at certain levels, particularly at the level
registration of complaint/FIR and compulsion of recruitment of SCs, STs, OBCs and women in
to pay bribe) did not develop too high a negative the police force. The failure of the states to meet
attitude towards police as one would have expected. the statutory mandate of reservations for SCs,
We also found that police’s initial response to STs and OBCs is made worse by the fact that the
crime; positive police contact and sense of safety representation of these groups is actually decreasing
in neighborhood have a direct bearing on citizens’ over the years in a number of states, contrary to the
levels of satisfaction and trust. Prior positive police popular opinion. Women, who should ideally be
contact was associated with greater optimism and representing half the police force, stood at a meager
openness towards future police contact. 7.3 percent representation in 2016, with none of
the states being able to meet even the 33 percent
8.2.1 Discrimination, minorities and benchmark. Data for the Muslims in the police
vulnerability force, available only till the year 2013, presents an
An important point of overlap between survey data equally bleak picture with poor representation of
and official data pertains to the relationship of the the Muslims in the police in proportion to their
police with vulnerable communities in India. In the population in the states. In contrast, however, there
Common Cause- CSDS survey, the opinion of those is a drastic over-representation of the above groups,
who affirmed the discriminatory attitudes of rule of barring women, in the prisons of India. Of the 22
law revealed core cleavages of caste and community. states studied for a period of five years, there is
Specifically, class emerged as the most significant disproportionately higher representation in prisons
and telling cleavage of discriminatory attitudes, of the SCs in 18 states, of the STs in 19 states and of
with over a half of the respondents confirming the Muslims in all of the 22 surveyed states. Since
that police discriminates between the rich and data for the OBCs is not available in the Census
the poor; followed by caste, gender and religion. 2011, a similar analysis could not be done for them.
Geographically, citizens’ perception of caste-based
8.2.2 Police violence and sympathy for
discrimination by the police was highest in Bihar,
personnel
Karnataka and Uttar Pradesh. The opinion of
Dalits in particular in Bihar, Madhya Pradesh and Another complex finding pertains to the high
Uttar Pradesh on caste based discrimination stood levels of popular agreement on police violence
out compared to other states. against the criminals. While geographically, there
were significant variations across states, the overall
We also found that incidences of fear of police were
pattern showed that a majority of citizens condoned
lowest among people living in metropolitan cities,
it instead of disapproving it. This finding helps us
among people belonging to the upper castes in general
understand excesses by police. The citizens, in their
and the upper castes in rural areas. Communities
eagerness to address crime, are ready to posit trust in
which were vulnerable and fearful in one context
the police and believe that police apprehend only the
may not feel so in another situation: thus, the levels
guilty. Secondly, citizens also seem to care less for
of fear among the Muslims were dependent on their
procedures and the foundational principles of rule
levels of concentration in the area. In states where
of law as far as inquiry into guilt of the suspects and
population concentration of the Muslims is higher,
punishment of the guilty are concerned. Therefore,
their levels of fear were lower whereas in states with
a tendency depicted and glorified by entertainment

industry about police turning themselves into judge Two important components of police infrastructure
and executioner seems to pervade both the self-image that would have a direct bearing on the investigation
of many police personnel and the ordinary citizen. of crime and prosecution duties of the police –
As a result, this study finds that the knowledge of forensics labs and training of police personnel – are
police violence does not agitate the public much; on in equally pathetic condition. Forensic laboratories
the contrary, it earns public approbation. are severely under-staffed, with shortages of over
two-third in UP and over three-fourth in Bihar. Basic
8.2.3 Police infrastructure training equipment were also found to be absent
Our survey sought to evaluate police performance across the states. Firing range was not available
based on their direct contact with the public. The in several training schools in many states, such as
official data, on the other end of the spectrum, Uttarakhand.
examined the infrastructural setup of the police as an
8.2.4 Police autonomy
organisation. Poor utilization of funds is a chronic
issue amongst all government institutions, with the While everyone expects police to be impartial and
police being no exception. In 2015, the utilization efficient, the idea of police autonomy does not
percentage of funds for police modernisation in seem to have much purchase. Neither the political
Andhra Pradesh, West Bengal, Tamil Nadu and executive nor the public in general appear to be
Bihar was shockingly nil. concerned much about police having autonomy in
their functioning. So, police autonomy is a matter
The Comptroller and Auditor General (CAG
often relegated to the backrooms of policy and does
hereafter) of India has done an extensive
not receive much traction in terms of public support
performance audit of police establishments in several
or demand. Yet, people’s perception of police
states and it reports substantial mismanagement of
autonomy appeared to be informed in a significant
modernisation funds, along with infrastructural
way by their overall opinion of police- people who
breakdowns across the 11 states in which the
believed that police is not independent were more
audits were conducted. These reports tell the
likely to hold a negative perception of them. This is
story of inefficiency and maladministration in the
an important finding since it shows that perhaps one
police structure. The poor state of criminal justice
of the biggest roadblocks to improving perceptions
system in India may be symptomatic of this larger
lies in improving police independence. One of the
organizational malfunctioning. Bihar, for instance,
directives of the Supreme Court in the Prakash
failed to spend 71 percent of the modernisation
Singh vs Union of India judgment of 2006 on police
funds due to delays in finalization of tenders for
reforms was the establishment of the State Security
procurement, while in Uttar Pradesh 80 percent of
Commissions (SSC) which would ensure that the
funds for training equipments had to be surrendered
police is able to function independent of unwarranted
due to administrative inefficiency.
government control, pressure or influence. Of the
A similar story of mismanagement and inefficiency 13 state Acts which were studied to understand
in the police infrastructure is documented across the level of compliance to the judgment, all 13
the different parameters evaluated by the CAG have established SSCs, but the composition of its
– be it shortage of buildings, vehicles, arms membership appears to be compromised. Five states
and ammunition, forensic or communication do not have provisions for the Leader of Opposition
infrastructures. More than two-thirds of the police to be a member, which is mandated, while four
staff from Madhya Pradesh could not be provided do not provide for non-political or independent
government accommodation. While the BPRD data members. Worse still, the recommendations of the
may not reveal a shortage of police stations against Commission have been made statutorily binding
its sanctioned number, but when measured against through specific provisions to that effect in only two
the norms set up by BPRD itself, a severe scarcity is states, Kerala and Himachal Pradesh.
disclosed, with 44 percent shortage in Uttar Pradesh
alone. 8.2.5 Incidence of crime
The vehicles and communication infrastructure Another crucial marker for people’s satisfaction
are equally deficient and outdated. The shortage with police in India, as revealed through the survey
of vehicles ranges from 40 percent shortage of findings, is the direct correlation between lower
motorcycles in Rajasthan to 71 percent in Madhya crime rates as perceived by the people with increased
Pradesh. When it comes to weaponry, nearly half satisfaction with police performance. Overall, we
the police force in Uttar Pradesh continues to use found that a greater proportion of citizens said that
outdated arms, while Rajasthan is short of 75 crime had decreased in their locality in the last 2-3
percent of the required modern weaponry. years. A variation in the perception of crime was
observed primarily in urban areas: as the locality

increased in size and urbanity, citizens reported an three states in terms of trust in police, in terms of
increase in the occurrence of crime. satisfaction with the functioning of police in one’s
area, the top three states were Kerala, Himachal
The crime rate from the official data suggests that
Pradesh and Odisha.
when taken as a cumulative score of the total crime
rates, violent crime rates, and rates of crimes against An important parameter for satisfaction with public
SCs, STs, women and children, states such as institutions is their efficiency and promptness in
Punjab, Himachal Pradesh and Jharkhand fare the dealing with cases. For the criminal justice system,
best, signaling the lowest crime rates. An important this can be measured through the disposal of cases-
finding here is that while the total rates of crimes an Index developed using the charge sheeting rate,
have had little fluctuations over the last five years, disposal percentage of cases by police, conviction
the rates of crimes against women, children, SCs rate and disposal percentage of cases by courts.
and STs have increased considerably between 2012 Chhattisgarh, Uttarakhand and Uttar Pradesh
and 2016. figure at the top of the disposal Index, while Delhi,
West Bengal and Assam figure at the bottom. It is
8.2.6 Trust and satisfaction to be noted that disposal of cases by the police is
The survey also illustrated that contrary to uniformly better across all states than disposal of
impressions, the police enjoy a fairly high degree cases by the courts. (For the police the disposal of
of trust, while explicit expression of high distrust a case here denotes a cumulative score of charge
was somewhat limited. However, in relation to sheeting rate and the percentage of cases in which
other public institutions such as the Army and investigation has been completed by the police. For
the judiciary, the police is less trusted and only the courts, the disposal of a case means conviction
fared better when compared to other government rate and the percentage of cases in which trial has
officers. Earlier studies (both rounds of SDSA) been completed by the court) However, here again,
suggest that institutions that are distant from routine discrimination against vulnerable groups is evident –
lives of citizens tend to acquire a halo and greater disposal of cases of crimes against SCs, STs, women
trust whereas institutions that have to continually and children is poorer than the disposal of the total
come into direct contact with citizens accumulate IPC and SLL cases. Whether the disposal is by the
less trust. In the case of the police, this is doubly police or by the courts, this trend is found to be
true: police being an institution that is in constant uniform nearly across all states. For instance, while
interaction with citizens, tends to be more critically the overall conviction rate for IPC and SLL cases in
evaluated and those segments of the police force, India is 75 percent it falls to a poor 21 percent when
such as the higher officials in the police hierarchy it comes to cases of crimes against women.
with whom citizens’ interaction is much more
8.2.7 State comparisons
limited tend to enjoy greater trust than police
personnel in the lower ranks. The Report draws attention to interstate variations
and in some cases, interstate uniformities. We hope
More than a person’s background, trust in police was
that states would learn from this report as far as
to a large extent a function of satisfaction, perception
their state-specific weaknesses are concerned and
and experience with it. Hierarchies of class, caste
address the gaps and limitations. A spirit of healthy
etc. impact one’s vulnerability and as vulnerability
comparison and competition can only enhance the
increases, trust levels in public institutions decrease.
performance of state level institutions such as the
That is, poor and lower classes, the STs and non
police.
literates had the highest levels of distrust in police
than privileged sections of respondents. In a society An overview of the official data shows that Himachal
driven by acute social inequalities, this is bad news Pradesh performed consistently well on multiple
indeed. If the criterion of successful democracy parameters such as crime rate index, police diversity
were satisfaction and trust of those at the bottom index, prison data index and compliance to the
of social hierarchies, then this failure of the police Prakash Singh directives, particularly with respect
becomes central to any discussion of trust in police. to police autonomy. Chhattisgarh, Madhya Pradesh
and Uttar Pradesh performed well in the category
While trust and satisfaction seem to be correlated
of disposal of cases. However, Uttar Pradesh was
when seen in overall terms, people appear to
lagging behind in all other aspects of policing such
understand the two concepts quite differently in
as diversity, infrastructure and prison data indices.
some of the states. We saw a divergent pattern
Delhi performed well on the police infrastructure
wherein states that reported highest trust levels with
index but poorly on the crime and disposal (of total
the police on our scoring scale were not the states
cases, and cases of crimes against SCs, women and
that also reported the highest satisfaction levels.
children) indices. West Bengal also figured as one of
If Assam, Jharkhand and Haryana were the top

the weakly performing states on the diversity index state specific strengths and weaknesses and focus
and along with Assam on the disposal (of total only on the net result in the form of a ranking. Such
crimes, crimes against STs and children) indices. rankings make splash but fail the policy makers and
the concerned actors to identify areas that require
According to the public perception survey, on
more attention. The present report prefers the more
nearly all the parameters, Uttar Pradesh and Bihar
mundane route of detailing to the more dramatic
were the least performing states, and Kerala,
route of ranking with a view to allow for nuance
Himachal Pradesh and Odisha stood out as the best
and hopes for targeted action because it is not the
performing states. A large gap in the performance
ranking but the actual improvement in policing and
of states in both the categories was observed. For
the satisfaction of the citizen which matters most.
example, on parameters such as satisfaction with
the performance of police, if Kerala ranked first, 8.3 Challenge of institutional
on the other end of the spectrum, Bihar scored the
lowest and was ranked at the bottom. The variation rejuvenation
between Himachal Pradesh and Uttar Pradesh on Policing is an unenviable task. Notwithstanding
positive perception of the police was widest at 47 the objectives to help the innocent and the needy,
points. Uttar Pradesh in particular fared poorly on to enhance and protect the civic virtue and to
multiple parameters such as positive perception of strengthen the idea of citizenship, the exercise of
police, trust in the police, sympathy for the working force and coercion not only posits extra-ordinary
conditions of police officers and favorability towards authority in the institutions and personnel involved
women in the police force. Among the larger in policing, it also potentially invites criticism of
states Bihar, Madhya Pradesh and Uttar Pradesh excess and arbitrariness. Any assessment of police
performed poorly on caste based discrimination by therefore is bound to be complex and even disputed.
the police, whereas Kerala, Jharkhand and Andhra More so, when the society is characterized by
Pradesh ranked on top in terms of trust in local diversity and competing claims by groups located at
police. On the parameter of police brutality, while different places in the socio-economic hierarchies.
nearly all the states condoned police violence, states This report seeks to bring out these complexities in
such Himachal Pradesh, Odisha, West Bengal an exercise assessing the performance of police.
and Nagaland stood out with the majority of the
respondents rejecting police violence. The report does not have evidence of full scale
condemnation of police by the public. Contrary
Some states fared better in one domain than others to popular narratives of chronic deficit of trust in
suggesting that there is scope to improve their institutions, our study found that citizens hold
performance with specific targeted interventions. moderate levels of satisfaction and trust in the police.
For example Jharkhand and Telangana fared well This apparently counterintuitive puzzle suggests that
on parameters such as trust in the local police and there is variance in the assessment of institutions
satisfaction with police help. However, incidence such as the police by ordinary citizens on the one
of crime was found to be greatest in Jharkhand and hand, and activists, opinion leaders and political
dissatisfaction with police’s investigation of crime theorists on the other. The difference in assessment
was significantly high in Telangana. is largely due to the nature of yardsticks which are
Juxtaposing evidence from the two data sources applied. There has been an enduring sense of despair
shows that that the overall performance of Kerala among activists, social workers and academics
is not consistent across official data and perception regarding the functioning, accountability and failure
survey. The perception survey illustrated that of public institutions in India. Political theorists
the state performed well on parameters such as in particular rely on theoretical explanations for
positive police contact, incidence of crime, religious understanding trust in institutions and these might
discrimination by the police and police autonomy. be at variance with popular perception of public
However, according to the official data, disposal institutions. Ordinary citizens, on the other hand,
indices (indices based on disposal of cases of crimes appear to be more optimistic and their moderate
against SCs, STs, women and children) highlight levels of trust and distrust need not be understood
that Kerala performed poorly on two of the disposal as absolute and uncritical submission to institutions
indices (disposal of cases of crimes against SCs and of power and authority. Citizens express negative
STs). Apart from the prison Index, the state does not opinions somewhat sparingly. The Indian citizen
figure in the list of top three best performing states is patient and indulgent toward authority. As the
on any of the other indices. preceding chapters show, rather than expressing
views located at the extremes of positive or negative
These variations have dissuaded us from constructing assessment, respondents tend to remain skeptically
a composite index of police performance for all hopeful and hence situate their assessments in the
states. Such an index would distract attention from intermediate terrain. In fact, the negative assessment

that comes to forefront despite these factors is rather police departments and senior officers accountable
striking and is certainly a cause for concern. for their relationship with citizens, and provide
pathways for future research on public institutions
Our analyses suggest that there is a clear relation
in India.
between vicarious experience (in the form of levels
of awareness about police violence), negative In nutshell, the present report shows that the police
encounters with the police and people’s fear and face a critical test today. As an institution, in most
perception of police. That is, the impact of police states of India, people are not exactly happy with
action on public opinion is not limited to the the police but as our survey shows, people have still
individuals that police deal with directly. People’s not given up on the institution. If the dissatisfaction
negative contact (e.g. non registration of FIR, and distrust increase that would have deep impact
complaint and compulsion to pay bribe) with the not merely on the police but on the legitimacy of
police shapes public opinion as much through the Indian state. It would adversely affect not merely
its impact on friends, family and acquaintances popular perceptions of police but also the ability of
as through the citizen directly involved in the democratic institutions to exercise authority with
encounter. Since attempts to improve public opinion care and efficiency. Often, questions of improving
by promoting positive contacts with the police would police as an institution are conveniently laid at the
probably be insufficient on their own, reducing the doors of the politician and the policy maker. But a
number of negative encounters with the police are careful look at the findings here would suggest that
likely to be more consequential. In addition, greater even as the politician and the policy maker need to
accountability of police personnel will also help be persuaded to do their bit, the police leadership
improve citizens’ faith in public institutions and give can also respond to the issues arising from these
them assurance of impartial inquiry. findings, without waiting for larger and systemic
reforms. Sheer professional commitment requires
At the first glance, the reader may also sense a
that the police leadership should address issues of
dichotomy between citizens perceptions and the
limited reform and urgent improvement.
findings from the official data. That is, the average
Indian appears to be satisfied with the day-to- Of course, the larger question that this report should
day functioning of the police machinery despite leave us with is this: Do we allow institutional
critical deficiencies in the police infrastructure; corrosion leading the citizens to more negative
poor performance on parameters such as disposal views thereby risking legitimacy of police; in the
of cases of crimes against SCs, STs, women and process, also risking that citizens will further lose
children; lack of diversity in the police force and hope and confidence in institutions of legitimate
disproportionate representation of minorities in coercion or should there be efforts at stalling the
prisons. But a careful examination reveals that the spiral of negativity that permeates our institutions,
two in fact feed into each other. It is not necessarily including the police? We hope that the dawn of
a paradox that people’s levels of satisfaction with institutional rejuvenation is what the citizens of this
the police are high despite its poor performance on country deserve and this report is a small step in the
several indicators. A clearer understanding of this direction of making this dream possible.
emerges when we examine the levels of sympathy
with the police, and discern it in the context of References:
administrative and state failures in providing SDSA Team. (2008). State of Democracy in India: A
optimum infrastructure for effective policing. Report. New Delhi: Oxford University Press.
This study also highlights the importance of periodic Lokniti Team. (2015). Democracy in India: A Citizens’
evaluation of official data on law enforcement, Perspective. New Delhi .
measuring public opinion and understanding the
Human Rights Watch. (2009). Broken System:
impact of policing through objectively designed Dysfunction, Abuse, and Impunity in the Indian Police.
surveys. This kind of empirical evidence will act as New York.
an intervention to improve public satisfaction, hold

APPENDICES
Students protesting outside Maurice Nagar police station in Delhi, which later ensued in clashes between the police and the students.
(Credits: Burhaan Kinu, Hindustan Times, 22 February 2017, New Delhi)

Appendix 1. Technical Details of Study Design and Sample
Status of Policing in India Report 2018: A Study of Performance and Perceptions is based on a sample
survey of 15563 respondents across 188 assembly constituencies in 22 states of India. The surveyed states
were namely Andhra Pradesh, Assam, Bihar, Gujarat, Haryana, Himachal Pradesh, Karnataka, Kerala,
Madhya Pradesh, Maharashtra, Nagaland, Odisha, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh, West
Bengal, Delhi, Jharkhand, Chhattisgarh, Uttarakhand and Telangana. The survey was conducted by Lokniti-
Programme for Comparative Democracy, Centre for the Study of Developing Societies (CSDS), in the
months of June and July, 2017.
I. Sampling Method
One of the key objectives of the study was to provide state-wise analysis of performance and perception of
policing. Therefore, the sample size for all 22 states was pre-decided based on their size. In big states such
as Uttar Pradesh and Tamil Nadu, 880 interviews were to be targeted. In midsized states such as Odisha
and Karnataka, 720 interviews and in small states such as Delhi and Nagaland, 480 interviews were to be
targeted.
Stage I: Sampling of Assembly Constituencies (ACs)
Based on the pre-decided sample target for each state and with a target of about 80 interviews per seat, the
number of ACs to be sampled per state was arrived at. A total of 188 assembly constituencies were randomly
selected using the Probability Proportionate to Size method.
Stage II: Sampling of Polling Stations (PSs)
The second stage of sampling was the selection of four polling stations within each sampled AC. 752 polling
stations were selected by listing all the PSs within the sampled ACs in the serial order followed by the
Election Commission. The selection of PS was also done using the Systematic Random Sampling Method.
Stage III: Sampling of Respondents
The third and final stage of sampling was selection of the respondents. In every polling station, 35 respondents
were selected from the latest electoral rolls using the Systematic Random Sampling Method. This procedure
ensured that the selected sample was fully representative of the cross-section of voters in the country. In each
sampled polling station, a list of sampled respondents was prepared by listing their name, age, gender and
address.
Table A1: Sampling framework

States Proposed Sample Number of ACs Number of Sampled ACs Achieved Sample
Assam 500 126 6 466
Bihar 900 243 11 814
Gujarat 900 182 11 895
Haryana 500 90 6 478
Karnataka 700 224 9 812
Kerala 700 140 9 613
Nagaland 500 60 6 549
Odisha 700 147 9 809
Punjab 500 117 6 488
Rajasthan 800 200 10 764
Tamil Nadu 900 234 11 1011

States Proposed Sample Number of ACs Number of Sampled ACs Achieved Sample
West Bengal 900 294 11 834
Delhi 500 70 6 503
Telangana 700 119 9 811
Total 15300 15563
II. Research Instrument

a) Preparation of the Questionnaire: The English questionnaire was designed after a rigorous dialogue
in a series of meetings and discussions. The main objective of the survey was to study citizens’ experience
with the police, their satisfaction and trust in the rule of law, police excesses, perception of discrimination,
responsiveness of criminal justice system and levels of impartiality. Most questions in the questionnaire were
structured, i.e., close-ended. However there were some that were kept open-ended in order to find out the
respondent’s spontaneous feelings about an issue without giving her/him a pre-decided set of options.
Pre-testing and Finalizing the Questionnaire: To check the accuracy and credibility of the questions set in
the questionnaire, it was necessary to administer it in the field. A pilot fieldwork was conducted on April
17-18, 2017 in poor & lower income settlements, middle & upper income settlements and urban villages of
Delhi. No sampling of any sort was carried out during pre-testing. The pre-testing was conducted by the
research team at Lokniti and Common Cause which was involved in the questionnaire designing. After
getting inputs from the researchers, the questions were reframed, omitted and added. This process also gave
insights to determine the length of questionnaire, writing instructions for field investigators and adding and
omitting some new options in answer categories.
Translation: It would not be justifiable to use a single language questionnaire in a multi lingual country like
India. Therefore, translation was done for each state by the regional team which was familiar with the lan-
guage of each region before administering the questionnaire in field. The questionnaire was translated in
eleven (Assamese, Bangla, Hindi, Gujarati, Kannada, Telugu, Tamil, Malayalam, Marathi, Oriya and Pun-
jabi) languages.
Training Workshop: A two-day training workshop was organized in each state before the survey fieldwork
began in order to train the field investigators (FIs) and supervisors who carry out the fieldwork operations.
The trainers conducted an intensive and interactive workshop wherein investigators underwent an orientation
programme and were trained rigorously on survey method, interviewing techniques and communication with
the respondents. A comprehensive and detailed interviewing guide, designed on the basis of the questionnaire
and survey methodology, was prepared for the interviewers. For a better understanding of the questionnaire,
mock interviews were also conducted by the interviewers.

b) Fieldwork: The fieldwork of the survey took place in the months of June and July, 2017. Field investigators,
who were mainly students of social sciences belonging to colleges and universities in different parts of the
country, were selected to carry out the field work. They conducted face to face interviews with the respondents
at the place of residence of the respondent using a standardized questionnaire in the language spoken and
understood by her/ him. They were instructed to interview only those respondents whose names were given
to them. At some locations the non-availability of sampled respondents or difficulty in locating households
necessitated replacements or substitutions. A total of 15562 interviews could be completed across the 22
States.
c) Data checking and analysis: All questionnaires were manually screened for consistency and quality
check. The questionnaire had codes (of pre-coded questions) that were used for data punching. A team was
constituted for checking the codes and making corrections if there were any mistakes. The checking and the
subsequent data entry took place at the Lokniti-CSDS office in Delhi. The analyses presented in this report
have been done using Statistical Package for the Social Sciences (SPSS). In order to be representative at the
state level, the achieved sample of every state was weighted by locality, religion, caste group and gender
based on Census 2011 figures.
Table A2: Profile of the achieved sample

Raw survey Weighted survey Profile of 22 States All-India population profile
sample (22 states) sample (22 states) based on Census 2011 based on Census 2011
Women 45.5 48.3 48.3 48.6
Urban 26.3 32.8 32.8 34.7
Scheduled Castes 18.6 16.0 16.1 16.2
Scheduled Tribes 10.8 11.3 11.4 9.7
Muslims 9.9 11.7 11.7 14.2
Christian 5.9 5.4 5.4 2.3
Sikhs 2.8 2.2 2.2 1.7

Appendix 2: Questionnaire
State ID Official No. of A.C. Official No. of P.S. Respondent No.

As in voter list
POLICE STUDY - 2017
LOKNITI, CSDS-COMMON CAUSE STUDY
F1.
State Name:
F2. A.C. Name: ___________________________________________________________
F3. P.S. Name: ____________________________________________________________
F4. Name of the Respondent: _________________________________________________
F5. Address of the respondent (Give landmark): __________________________________
F6. Date of interview (dd/mm/yyyy): ___________________________________________
F7. Name of the Investigator (Code Roll No.): ____________________________________
INVESTIGATOR’S INTRODUCTION AND STATEMENT OF INFORMED CONSENT
My name is __________________and I have come on behalf of Centre for the Study of Developing Societies
(also give your university’s reference), a social science research organization and Common Cause, an NGO
in Delhi. We are conducting a survey on people’s perception and experience of dealing with the police. Every
person over the age of 18 has an equal chance of being included in this study. You have been selected by
chance. There is no risk and also no benefit in participating in this survey and your participation is voluntary.
This survey is an independent study and is not linked to any political party or government agency. Your
identity and the information you provide will be kept strictly confidential. Participation in this survey is
voluntary. We hope that you will take part in this survey since your participation is important. It usually
takes 30 to 35 minutes to complete this interview. Please spare some time for the interview and help me in
sucessfully completing the survey.
F8. May I begin the interview now? 1. Respondent agrees to be interviewed

2. Respondent does
not agree to be interviewed INTERVIEW BEGINS:

Z1. Let us begin by talking about this village/ town you live in. How long have you lived here? (Number
of years) _______________________________ 97. Entire life 98. Don’t know
Z1a. Which state do your ancestors belong to? (Record state name & consult Codebook for coding)
Name of State: ________________________________________ 98. Don’t know
Z2. What is your age? (in completed years) ______ 98. No Response (Code 95 for 95 yrs & above)
Z3. Gender: 1. Male 2. Female 3. Other
Z4. Up to what level have you studied? (Record exactly and consult code book)
_______________________________________ 9. No response
Q1. How often do incidents such as burglary, murder, physical assault, chain snatching occur in your
locality very often, sometimes, rarely or never? 1. Very often 2. SomeTimes
3. Rarely 4. Never 8. DK
Q1a. In your opinion, is the police’s investigation of such incidents proper and satisfactory or is it
temporary investigation?
1. Investigation is proper and satisfactory 2. Temporary investigation
3. Police harasses people during investigation 4. Does not carry out the needful investigation 8. DK

Q2. What kind of police presence would you like to see in your village/area- greater, less or no change?
1. Greater 2. Less 3. No change, same as before 8. DK
Q3. From time to time, for different purposes, people have some kind of contact with the police. In the
last 4-5 years, have you or your family member had any kind of contact with the police?
2. Yes 1. No 8. DK
(If in Q3 the respodent’s answer is NO or DK, then please do no ask questions Q3a to Q8a and move to Q9)
Q3a. If in Q3, answer is yes) So did you or someone from your family contact the police or the police
contactedyou?
1. I contacted the police 2. Police contacted me 3. Both 8. DK 9. NA
Q3b. (If in Q3, answer is yes) What was the reason for contacting the police or the police
contacting you?
(Investigator can write down upto two reasons for police contact and alongside each reason, kindly
ask the respondent whether she or he was an accused, victim or witnessed a crime.)
Q3ba.Reason for police contact
a. _______________________________________________________98. DK/CS 99.NA
1. Accused 2.Victim 3. Witnessed a crime 4.Other____________8.DK 9. NA
Q3bb. b. ________________________________________________________98. DK/CS 99.NA
1. Accused 2.Victim 3. Witnessed a crime 4.Other____________8.DK 9. NA
Q4. How did you first contact the police- over the phone, visited the police station, via internet/online or
police visited residence or workplace?
1. Over the phone 2. Visited the police station 3. Via internet/ Online
4. Police visited the residence/workplace of the respondent 5. Other______ 8. DK 9.NA
Q5. Who assisted you in contacting the police or visiting the police station?
1. Family member 2. Influential person 3. Neighbour / friend
4. Any other person_________ 5. Nobody, went alone 8.DK 9.NA
Q6. On contacting the police, was your complaint/ FIR registered? 2.Yes 1. No
8. Can’t say 9. N.A.
Q6a. (If no in Q6) So in that case, why did the police not file your complaint/ FIR? (Record answer and
consult codebook) ______________________________________ 98.DK 99.NA
Q6b. (If yes in Q6) How was the FIR registered-was it read out, written or via mail?
1. Oral/ read out 2. Written 3. Email/via internet
4. Other_____________________ 8. DK 9. N.A.
Q6c. (If yes in Q6) Did you get a copy of the FIR? 2.Yes 1. No 8. DK 9. N.A.
Q7. During the last 4-5 years, whenever you contacted a police officer or visited the police station, did you
have to pay a bribe to get your work done? 2.Yes 1. No 8. DK 9.NA
Q8. How satisfied or dissatisfied were you with the help provided at the police station? (If satisfied or
dissatisfied, probe further whether fully or somewhat)
1. Very satisfied 2. Somewhat satisfied
3. Somewhat dissatisfied 4. Very dissatisfied 8. DK
9. NA
Q8a. (If somewhat or fully dissatisfied in Q8), What was the main reason for your dissatisfaction?
(Record answer and consult codebook)________________________ 98.DK 99.NA
Q9. In the future, if you have a problem that requires police help, would you go to the police?
2.Yes 1. No 3. Probably 4. Have no other option 8. DK

Q10. How unsafe do you feel in your village/ neighbourhood- very, somewhat, not very or not at all?
Very unsafe Somewhat Not very unsafe Not at all unsafe DK
a a. Early morning 1 2 3 4 8
b b. During the day 1 2 3 4 8
c c. At night 1 2 3 4 8
Q11. Usually both men and women work in the police force. In your opinion who is more:
Police woman Police man Both Neither DK
a a. Honest 1 2 3 4 8
b b. Hardworking 1 2 3 4 8
And........
c c. Whom would you approach for help 1 2 3 4 8
Q12. Listed below are a number of institutions. Please tell me how much trust do you have in each of
them-a lot, somewhat, not much or not at all? A lot Somewhat Not much Not at all DK
a a. Local police like police inspector, 1 2 3 4 8
Sub inspector, SHO
b b. Senior police officer like SP, DCP 1 2 3 4 8

c c. Traffic police 1 2 3 4 8
d d. Army/Paramilitary 1 2 3 4 8
e e. Court 1 2 3 4 8
f f. Government official 1 2 3 4 8
Q13. Often women and young girls are scared to seek help from the police or visit the police station. In
your opinion, what is the main reason for this? (Record answer and consult the codebook for
coding).________________________________________________________________ 98. DK
Q14. On a scale of 10 points where the 1st point at left stands for extremely ineffective and the10th point at
the right stands for extremely effective. In your opinion, where would you place the following in
terms of effectiveness to get work done from the police? SHOW THE SCALE AND EXPLAIN
(If no answer is given, then code 98) Extremely ineffective Effectiveness
a. a. Political connection 01 02 03 04 05 06 07 08 09 10
b. b. Money 01 02 03 04 05 06 07 08 09 10
c. c. Personal connections in the police 01 02 03 04 05 06 07 08 09 10
d. d. Seeking help from local goon 01 02 03 04 05 06 07 08 09 10
Q15. In the last 2-3 years, has crime in your locality increased or decreased?
1. Increased 2. Decreased 3. Remained the same 8. DK
Q16. Now I will read out two statements. Please tell me which statement would you agree the most with?
Statement 1: Police is not able to function properly due to lack of training and other resources.
Statement 2: It is not that the police lacks resources, they are in fact lazy and not motivated to serve
people.
1. Agree with statement 1 2. Agree with statement 2 8. DK
Q17. Do you think the police intentionally implicates people under false charges?
2.Yes 1. No 3.May be 8. Can’t say
Q18. In an area, whenever there is an instance of fight between people from two religious communities,
do you think the police sides with any particular religious community or remains impartial ?
1. Sides with a particular religious community 2. Remains impartial 8. DK

Q18a. (If answer in Q18 is police sides with a particular religious community), In your opinion, which
religious community does the police take sides with? (Record answer and consult codebook)
______________________________________________________________ 98. DK 99. NA
Q19. How satisfied are you with police performance and their work in your area? (If satisfied or dissatisfied,
probe further whether fully or somewhat.) 1. Fully satisfied 2. Somewhat satisfied
3. Somewhat dissatisfied 4. Fully dissatisfied 8. DK
Q20. Do you know of anyone who....? Yes No DK/ CS
a a. Died in mysterious circumstances under police custody 2 1 8
b b. Been taken into unlawful detention by police/army 2 1 8
c c. A woman who is a victim of sexual harassment or eve teasing by the police. 2 1 8
d d. A victim of fake encounter 2 1 8
e e. A victim of police firing, lathi charge. 2 1 8
f f. A victim of police torture. 2 1 8
Q21. Very often , even in times of need, people are hesitant to visit the police or seek help. What is the single
most important reason for this? ________________________________________ 98. DK
Q22. Now I will read out some statements. Please tell me whether you would agree or disagree with the
following?
(Probe further whether ‘fully’ or ‘somewhat’ agree or disagree) Agree Disagree NR
Fully Somewhat Somewhat Fully
a a. Police is blamed unnecessarily even when it does 1 2 3 4 8

its job well
b b. There is nothing wrong in politicians interfering in
the transfer and posting of police officers. 1 2 3 4 8
c c. There is nothing wrong in the police being violent

towards criminals. . 1 2 3 4 8
Q23. Looking at the present number of these communities in the police force, are there adequate numbers
of the following groups in the police ? Adequate Less than adequate More than adequate DK
a a. Scheduled Caste such as Dalits 1 2 3 8

b b. OBC 1 2 3 8
c c. Scheduled Tribes such as Adivasis 1 2 3 8
And what about the following?
d d. Muslims 1 2 3 8
e e. Women 1 2 3 8
Q24. On a scale of 10 points where the 1st point at left stands for extremely corrupt and the10th point at
the right stands for not at all corrupt. In your opinion, where would you place the following
institutions in terms of corruption? SHOW THE SCALE AND EXPLAIN (If no answer is given,
then code 98) Extremely corrupt Not at all corrupt
a a. Local police- police inspector,
Sub inspector, SHO, Beat constable 01 02 03 04 05 06 07 08 09 10

b b. Senior police officer- SP, DCP 01 02 03 04 05 06 07 08 09 10
c c. Traffic police 01 02 03 04 05 06 07 08 09 10

d d. Army/Paramilitary 01 02 03 04 05 06 07 08 09 10
e e. Court 01 02 03 04 05 06 07 08 09 10
f f. Government official 01 02 03 04 05 06 07 08 09 10

Q25. Now I will read out a few statements. Please tell me whether you agree or disagree with these
statements?
(Probe further whether strongly or somewhat agree or disagree) Agree Disagree NR
a a. Police as a profession is better than other occupations
because of easy access to power and security. 1 2 3 4 8
b b. Compared to other professions, it is difficult to work in
the police force because of high stress levels and long 1 2 3 4 8
working hours.
c c. Police is corrupt- it does not do its job without a bribe. 1 2 3 4 8
Q26 Often people are scared of police due to different reasons. What about you- how scared are you of
the following- a lot, somewhat, not much or not at all? Very Somewhat Not much Not at all DK
a a. Fear of being beaten up by the police 1 2 3 4 8
b b. Fear of being arrested by the police for no reason 1 2 3 4 8
c c. Fear of the police coming to your house 1 2 3 4 8
d d. Fear of being falsely implicated in police cases. 1 2 3 4 8
e e. Fear of sexual harassment or eve teasing by the police.1 2 3 4 8
Q27. Have you ever seen a police officer violating the law? 2.Yes 1. No 8. DK
Q27a. (If answer is yes in Q27) Then did you file a complaint? 2.Yes 1. No 8. DK 9. NA
Q27b. Do you think there should be a separate body for filing complaint against police officers who violate
the law?
2.Yes 1. No 8. DK
Q28. Often the police says that it is not allowed to work independently and free from political interference.
Please tell me how much interference do these groups have in the functioning of the police in your
locality- a lot, somewhat, not much or not at all? A lot Somewhat Not much Not at all DK
a a. Political party & politicans 1 2 3 4 8
b b. Local goons / gundas 1 2 3 4 8
c c. Senior police officers 1 2 3 4 8
d d. Economically powerful groups 1 2 3 4 8
e e. Influential and dominant caste of your area 1 2 3 4 8
Q29 Do you know of someone or have heard of a case of domestic violence in your village/ locality?
2.Yes 1. No 8. DK
Q29a. (If answer is yes in Q29) Do you know if the victim of domestic violence filed a complaint or not?
2.Yes 1. No 8. DK 9.NA
Q29b. In cases of domestic violence, is the police helpful to the victim? 2.Yes 1. No 8. DK
Q30. It is often argued that police harasses certain groups of people. Have you seen the police harassing
the following communities? Q30a. Is the police right or wrong in taking action against these
communities?
Yes
Q30 Q30a
No DK
Q30a.
Right
Wrong
DK
a a. Rickshaw pullers 2 1 8 2 1 8
b b. street vendors 2 1 8 2 1 8
c c. Nat dancers/street performers/Madaris/ 2 1 8 2 1 8
Saperas (specify Denotified and Nomadic
Tribes in every state)

d d. Beggars 2 1 8 2 1 8
e e. Hijras/ Kinnars/ Kothi 2 1 8 2 1 8
Q31. According to you, in the police force of your village/ locality? (Record answer and consult codebook
for coding)
a. Which religious community’s members are more in number?_________________________ 98. DK
b. Which caste’s members are more in number? ___________________________________ 998. DK
Q32. It is widely believed that police discriminates between people on the basis of different things. In your
opinion, does the police discriminate? Yes No DK/CS
a a. On the basis of caste. 2 1 8
b b. On the basis of religion. 2 1 8
And does it also discriminate between
c c. Rich and poor. 2 1 8
d d. Women and men. 2 1 8
e e. People from another state. 2 1 8
Q33. Many people argue that working in the police is not appropriate for women. Now I am going to read
out some such arguments. Please tell me whether these arguments are justified or not? (Probe further
whether very or somewhat justified or unjustified). Justified Unjustified DK
Very Somewhat Somewhat Very
a a. Being in the police requires physical strength and

aggressive behavior which women lack. 1 2 3 4 8
b b. A woman should prioritise managing home

instead of joining the police force. 1 2 3 4 8
c c. Women police are incapable of handling high

intensity crimes and cases. 1 2 3 4 8
d d. Because of inflexible working hours, it is difficult

for women to work in the police force. 1 2 3 4 8
Q34. Please tell me whether the following measures in your locality/ area have been introduced or not?
Introduced Not introduced CS
a a. All women police station. 1 2 8
b b. PCR van patrolling in your locality. 1 2 8
c c. Senior citizen helpline. 1 2 8
d d. Child helpline number. 1 2 8
e e. Helpline for people from North Eastern part of India 1 2 8
(To be asked only in cities)
f f. Help desk for SC & ST in police station. 1 2 8
Q35. If your daughter/son was to be the victim of any crime, would you allow her/him to visit the police
station alone to file a complain? (To be asked to everyone including unmarried respondents)
2.Yes 1. No 8. DK
Q35a. (If yes in Q35) Would you allow your daughter to visit the police station as much as you would
allow your son?
2. Yes 1. No 8. DK 9. NA
Q36. In your locality, members from which caste/community are more in number? (Record answer and
consult codebook) ______________________________ 998. DK

Q37. In an area, whenever there is an instance of fight between people from two caste groups, do you think
the police sides with any particular caste group or remains impartial ?
1. Sides with a particular caste group 2. Remains impartial 8. DK
Q37a. (If answer in Q37 is police sides with a particular caste group), In your opinion, which caste
group does the police take sides with? (Record answer and consult codebook)
______________________________________________________________ 998. DK 999. NA
Q38. Now I will read out three statements. Please tell me which of these would you agree with? (Probe
further whether ‘fully’ or ‘somewhat’ agrees or disagrees). Agree Disagree NR
a a. Often members of backward castes such as Dalits
are falsely implicated in petty crimes such as theft,
robbery, dacoity by the police. 1 2 3 4 8
b b. Often tribals are falsely implicated on Maoist charges

by the police. 1 2 3 4 8
c c. Often Muslims are falsely implicated in terrorism
related cases by the police. 1 2 3 4 8
Q39. In any of the above cases, if the person is absolved of charges by the court, should action be initiated
against policemen who implicated them? 2.Yes 1. No 8. DK
Q40. In the last 2-3 years, have you tried calling police (100 number) on phone in an emergency situation?
2.Yes 1. No 8. DK
Q40a. (Only for women) Have you ever called on the women’s helpline number?
2.Yes 1. No 8. DK 9. NA
Q40b. (If yes in Q40) Then on calling police did you receive any help?
2. Yes 1. No 3. No one took the call
4. Number was not in use 5. Other _________ 8. DK 9.NA
Q40c. (If yes in Q40) In your experience, to what extent has access to 100 number improved in the last 2-3
years- a lot, somewhat, not much or not at all? 1. A lot 2. Somewhat 3. Not much
4. Not at all 8.DK 9. NA
Q41. In comparison to other jobs, are the working hours of the police greater or less?
1. Greater 2. Less 3. Same as other jobs 8.DK
Q42. We know that the process of justice often gets delayed and numerous cases remain pending for
several years. According to you, which institution is responsible for this delay?
1. Police 2. Court 3. Both 4. Neither
5. Other________________________ 8. DK
Q43. Often, police harasses those girls and boys who hang out and roam around in public places. In your
opinion is this right, somewhat right or wrong?
1. Right 2. Somewhat Right 3. Wrong 8. DK
Q44. In your locality, have you seen the police preventing girls and boys from meeting and hanging out in
public area?
2. Yes 1. No 8. DK

BACKGROUND DATA
Personal Information
Z4a. Up to what level have your father and your mother studied?
Z4a. Father:____________________ Z4b.Mother:_____________________ 9. No response
Z5. What is your main occupation? (Record exactly and consult codebook & if retired, try to ascer
tain his/her previous occupation. If student or housewife, then note down that as well)
_______________________________________ 98. No response
Z5a. Are you the main earner of your household? 2. Yes 1. No
Z5b. (If No in Z5a) What is the occupation of the main earner of your household? (Record exactly and
consult codebook) _____________________________________99. NA
Z6. How far is the nearest police station/chowki from your village/locality? (Record answer in
kilometer. If answer is more than 100 kilometer then code 96) _________________ 98. Can’t say
Z7. Are you married? 1. Married 2. Married (Gauna not performed, not started living together)
3. Widowed 4. Divorced 5. Separated 6. Deserted
7. Unmarried/Single 8. Live with partner but not married 9. NR
Z7a. ( If married) Do you have a boy or a girl?
1. Boy 2. Girl 3. Both 4. None 8. NR 9. NA
Z8. What is your Caste/Jati-biradari/Tribe name?(Consult code book for code)
______________________________________
Z8a. And what is your caste group? (Double check and consult code book)
1. Scheduled Caste (SC) 2. Scheduled Tribe (ST) 3. Other Backward Classes (OBC) 4. Other
Z9. What is your religion? 1. Hindu 2. Muslim 3. Christian 4. Sikh 5. Buddhist/Neo Buddhist
6. Jain 7. Parsi 8. No religion 9. Others (Specify) ___________
Z10. Generally, which language is spoken in your house?(Consult code book for coding)
_________________________ 98. No response
Z11. What kind of mobile phone do you have – a normal phone or a smart phone with a touch screen?
1. Normal phone 2. Smart phone 3. Don’t have a phone 8. No answer
Z11a. (If respondent has a mobile phone) Does your phone have an internet connection?
2.Yes 1. No 8. No Answer 9. Not Applicable
Z12. Do you have an Aadhaar Card? 2.Yes 1. No
Z13. Locality: 1. Village 2. Town (50,000 to 1 lakh population)
3. Small City (1-5 lakh) 4. Big City (5-10 lakh) 5. Metropolitan City (Above 10 lakh)
Z13a. (If Town/Small City/Big City/Metropolitan City) Type of house where the respondent lives
1. House/Flat/Bunglow 2. House/Flat with 5 or more rooms
3. House/Flat with 4 rooms 4. Houses/Flat with 3 rooms 5. Houses/Flat with 2 rooms
6. House with 1 room 7. Mainly Kutcha house 8. Slum/Jhuggi Jhopri
9. NA.
Z13b. (If Village) Type of house where the respondent lives
1. Pucca (both wall and roof made of pucca material)
2. Pucca-Kutcha (Either wall or roof is made of pucca material and other of kutcha material)
3. Kutcha/Mud houses (both wall and roof are made of kutcha material )
4. Hut (both wall and roof made of grass, leaves, un-burnt brick or bamboo) 9. NA.

Household Information
Z14. Total No. of family members living in the household:

Above 18 years:_____ Below 18 years :___ (If more than 9, Code 9)
Z15. Total agricultural land including orchard and plantation owned by your household (as
on date of survey):
_____________(Ask in local units, but record in standard acres. If more than 99, Code 99)
Z16. In normal circumstances, what is your monthly household expenditure?
(Record in Rupees, if respondent gives no answer, fill 000000 in the box)______________
Z17. Do you or members of your household have the following: Yes No
a a. Car/Jeep/Van 1 2
b b. Scooter/Motorcycle/Moped 1 2
c d1 c. Airconditioner 1 2
d d. Computer/laptop/Ipad 1 2 (d1. If resp. has computer/
laptop)
e e. Fan/Cooler 1 2 1.With Internet 2.Without
Net 9. NA
f f. Washing machine/Microwave 1 2
g g. Fridge 1 2
h h. TV 1 2
i i. Bank/Post office account 1 2
j j. ATM/Debit/Credit card 1 2
k k. LPG gas 1 2
l l. Toilet inside the house 1 2
m m. Motorised pumping set 1 2
n n. Tractor 1 2
o o. Handpump inside the house 1 2
Z18.
Livestock: Total Number
a. Goat /sheep/pig: ___________
b. Cow/Oxen /buffalo/Camel: ___________
c. Any other: ___________
Z19. Total monthly household income - putting together the income of all members of the
household?
_____________________________________________________________
(Record exact amount in Rupees. If respondent does not give any amount then record 000000)
Z20. Mobile/Telephone number of the respondent _____________________

FILL AFTER COMPLETING INTERVIEW
E1. Were there any other people immediately present who might be listening during the interview?
1. No one 2. Husband 3. Other adult male family members 4. Adult female family members
5. Any male from neighborhood 6. Any Female from neighborhood
7. Small crowd 8. Any Other
E2. In how many questions did the respondent check with others for information to answer for
questions?
1. None 2. One or two 3. Three to five 4. Six to ten 5. More than 10
E3. At some stage did you notice something that made you feel that the respondent was answering
under some fear or pressure? 1. Yes 2. No 3. Not sure
E4. Which caste community was more in number in the locality you visited? (Record answer and
consult codebook) _______________________________________________________________
E5. Which religious community was more in number in the locality you visited? (Record answer and
for coding refer to Z9 Codes. ) _____________________________________________________
E6. Overall was the respondent cooperative? 1. Yes, very much 2. Somewhat 3. Not at all
E7. Investigator’s signature (Sign in box):
E8. Name of the Supervisor: ________________________________________________________
E9. Checked by the Supervisor: 1. Yes 2. No
E10. Supervisor’s signature (Sign in box):
Any additional notes/comments
__________________________________________________________________________________________
__________________________________________________________________________________________
__________________________________________________________________________________________

Appendix 3: Details of How The Indices Were Constructed
Index of fear of police
The Index was constructed by taking into account 5 questions asked in the survey. They are –
Q26. Often people are scared of police due to different reasons. What about you - how scared are you of the
following?
Q26a. Fear of being arrested by the police for no reason
Q26b. Fear of being arrested by the police for no reason
Q26c. Fear of the police coming to your house
Q26d. Fear of being falsely implicated in police cases
Q26e. Fear of sexual harassment or eve teasing by the police
In each question, the response options offered to the respondent were ‘a lot’, ‘somewhat’, ‘not much’ or ‘not
at all’. Across all the questions, a no response category was also provided, in case the respondent refused to
answer the question.
Step 1: An ‘a lot’ answer was scored as 4, a ‘somewhat’ answer was scored as 3, a ‘not much’ answer was
scored as 2, a ‘not at all’ answer was scored as 1 and a no response was scored as 0.
Step 2: The scores of all questions were summed up. The summated scores of all questions ranged from 0
to 20.
Step 3: These summated scores were then distributed across five newly created categories that indicated
different degrees of fear of police among the people –
A total score ranging from 16 to 20 was categorised as ‘Highly fearful’.
A total score ranging from 11 to 15 was categorised as ‘Somewhat fearful’.
A total score ranging from 6 to 10 was categorised as ‘Not much fearful’.
A total score ranging from 1 to 5 was categorised as ‘Not at all fearful’.
A total score of 0 was treated as being ‘Non-committal’.
Index of awareness about police excesses
The Index was constructed by taking into account 6 questions asked in the survey. They are -
Q20a. Do you know of anyone who died in mysterious circumstances under police custody?
Q20b. Do you know of anyone who had been taken into unlawful detention by police/army?
Q20c. Do you know a woman who is a victim of sexual harassment or eve teasing by the police?
Q20d. Do you know anyone who is a victim of fake encounter?
Q20e. Do you know anyone who is a victim of police firing, lathi charge?
Q20f. Do you know anyone who is a victim of police torture?
In each question, the possible response options were ‘yes’ and ‘no’.
Step 1: A ‘yes’ answer was scored as 1 and a ‘no’ answer or no response was scored as 0. Across all the
questions, a no response category was also provided, in case the respondent refused to answer the question.
to 6.
Step 3: These summated scores were then distributed across four newly created categories that indicated
different levels of awareness about police excesses –
A total score of 5 or 6 was categorised as ‘Know of many cases’.
A total score of 3 or 4 was categorised as ‘Know of quite a few cases’.

A total score of 1 or 2 was categorised as ‘Know of a few cases’.
A total score of 0 was categorised as ‘Don’t know of any case’.
Index of perception about the police
Q12a. Please tell me how much trust do you have in local police like police inspector, Sub inspector, SHO?
Q12b. Please tell me how much trust do you have in senior police officer like SP, DCP?
Q12c. Please tell me how much trust do you have in traffic police?
Q17. Do you think the police intentionally implicate people under false charges?
Q22a. Please tell me whether you would agree or disagree with - Police is blamed unnecessarily even when
it does its job well? (Probe further whether ‘fully’ or ‘somewhat’ agree or disagree)
Q24a On a scale of 10 points where the 1st point on the left stands for extremely corrupt and the 10th point
on the right stands for not at all corrupt, where would you place the local police, i.e, police inspector, Sub
inspector, SHO, beat constable, in terms of corruption?
Q24b. On a scale of 10 points where the 1st point on the left stands for extremely corrupt and the 10th point on
the right stands for not at all corrupt, where would you place the senior police officer, i.e., SP, DCP in terms
of corruption?
Q24c. On a scale of 10 points where the 1st point on the left stands for extremely corrupt and the 10th point
on the right stands for not at all corrupt, where would you place the traffic police in terms of corruption?
Q25c. Please tell me whether you agree or disagree with this statement - Police is corrupt, it does not do its
job without a bribe? (Probe further whether ‘fully’ or ‘somewhat’ agree or disagree)
In Q12a, Q12b, Q12c, the response options offered were ‘a lot’, ‘somewhat’, ‘not much’ and ‘not at all’. In
Q17, the possible responses were ‘yes’, ‘no’ and ‘maybe’. In Q22a, the response options offered were ‘fully
agree’, ‘somewhat agree’, “somewhat disagree’ and ‘fully disagree’. In Q24a, Q24b and Q24c, the response
options offered were any number between 1 and 10 where 1 stood for extremely corrupt and 10 stood for
not at all corrupt. In Q25c the response options offered were ‘fully agree’, ‘somewhat agree’, “somewhat
disagree’, and ‘fully disagree’. Across all the questions, a no response category was also provided, in case the
respondent refused to answer the question.
Step 1: In Q12a, Q12b and Q12c, an ‘a lot’ answer was scored as 4, a ‘somewhat’ answer was scored as 3, a
‘not much’ answer was scored as 2, a ‘not at all’ answer was scored as 1 and a no response was scored as 0. In
Q17, a ‘no’ answer was scored as 4, a ‘maybe’ answer was scored as 3, a ‘yes’ answer was scored as 2 and a
no response was scored as 0. In Q24a, Q24b and Q24c, answers ‘8’ ‘9’ and ‘10’ were scored as 4, answers ‘6’
and ‘7’ were scored as 3, answers ‘4’ and ‘5’ were scored as 2, answers ‘1’ ‘2’ and ‘3’ were scored as 1 and a
no response was scored as 0. In Q22a, a ‘fully agree’ answer was scored as 4, a ‘somewhat agree’ answer was
scored as 3, a ‘somewhat disagree’ answer was scored as 2, a ‘fully disagree’ answer was scored as 1 and a ‘no
response’ answer was scored as 0. In Q22a, a ‘fully disagree’ answer was scored as 4, a ‘somewhat disagree’
answer was scored as 3, a ‘somewhat agree’ answer was scored as 2, a ‘fully agree’ answer was scored as 1
and a no response was scored as 0.
to 36.
Step 3: These summated scores were then distributed across five newly created categories that indicated how
negative or positive people’s perception of police is -
A total score ranging from 25 to 36 was categorised as ‘Very positive’.
A total score ranging from 19 to 24 was categorised as ‘Somewhat positive’.
A total score ranging from 16 and 18 was categorised as ‘Somewhat negative’.
A total score ranging from 1 to 15 was categorised as ‘Very negative’.

Index of feeling safe
Q10a. Early morning, how unsafe do you feel in your village/ neighbourhood?
Q10b. During the day, how unsafe do you feel in your village/ neighbourhood?
Q10c. At night, how unsafe do you feel in your village/ neighbourhood?
In each question, the response options offered to the respondent were ‘very’, ‘somewhat’, ‘not much’ or ‘not
at all’. Across all the questions, a no response category was also provided, in case the respondent refused to
answer the question.
Step 1: A ‘very’ answer was scored as 4, a ‘somewhat’ answer was scored as 3, a ‘not much’ answer was
to 12.
different degrees of feeling unsafe -
A total score ranging from 9 to 12 was categorised as ‘Feel highly unsafe’.
A total score ranging from 6 to 8 was categorised as ‘Feel unsafe’.
A total score of 4 or 5 was categorised as ‘Feel somewhat safe’.
A total score ranging from 1 to 3 was categorised as ‘Feel highly unsafe’.
Index of trust in police
Q12a. Please tell me how much trust do you have in local police like police inspector, Sub inspector, SHO?
Q12b. Please tell me how much trust do you have in a senior police officer like SP, DCP?
Q35. If your daughter/son was to be the victim of any crime, would you allow her/him to visit the police
station alone to file a complaint?
In Q12a and Q12b the response options offered were ‘a lot’. ‘somewhat’, ‘not much’, and ‘not at all’. In Q35,
the possible response options were ‘yes’ and ‘no’. Across all the questions, a no response category was also
provided, in case the respondent refused to answer the question.
Step 1: In Q12a and Q12b, an ‘a lot’ answer was scored as 4, a ‘somewhat’ answer was scored as 3, a ‘not
much’ answer was scored as 2, a ‘not at all’ answer was scored as 1 and no response was scored as 0. In Q35,
a ‘yes’ answer was scored as 4, a ‘no’ answer was scored as 1 and a no response was scored as 0.
to 12.
different degrees of trust in the police –
A total score ranging from 10 to 12 was categorised as ‘Highly trust’.
A total score ranging from 7 to 9 was categorised as ‘Somewhat trust’.
A total score ranging from 4 to 6 was categorised as ‘Somewhat distrust’.
A total score ranging from 1 to 3 was categorised as ‘Highly distrust’.

Index of perception about police independence
Q28a. Please tell me how much interference do the political party & politicians have in the functioning of
the police in your locality?
Q28b. Please tell me how much interference do the local goons/goondas have in the functioning of the police
in your locality?
Q28c. Please tell me how much interference do the senior police officers have in the functioning of the police
in your locality?
Q28d. Please tell me how much interference do the economically powerful groups have in the functioning of
the police in your locality?
Q28e. Please tell me how much interference do the influential and dominant castes of your area have in the
functioning of the police in your locality?
In all the questions the response options offered were ‘a lot’. ‘somewhat’, ‘not much’, and ‘not at all’. Across
all the questions, a no response category was also provided, in case the respondent refused to answer the
question.
Step 1: An ‘a lot’ answer was scored as 1, a ‘somewhat’ answer was scored as 2, a ‘not much’ answer was
to 20.
Step 3: These summated scores were then distributed across five newly created categories that indicated how
negative or positive people’s perception of independence of police is –
A total score ranging from 15 to 20 was categorised as ‘Highly independent’.
A total score ranging from 10 to 14 was categorised as ‘Somewhat independent’.
A total score ranging from 5 to 9 was categorised as ‘Not much independent’.
A total score ranging from 1 to 4 was categorised as ‘Not independent at all’.
Index of perception of discrimination by the police
Q18. In an area whenever there is an instance of fight between people from two religious communities, do
you think the police sides with any particular religious community or remains partial?
Q32. It is widely believed that police discriminates between people on the basis of different things. In your
opinion…
Q32a: Does the police discriminate on the basis of caste?
Q32b: Does the police discriminate basis of religion?
Q32c: Does the police also discriminate between rich and poor?
Q32d: Does the police also discriminate between women and men?
Q32e: Does the police also discriminate between people from another state?
Q37. In an area, whenever there is an instance of a fight between people from two caste groups, do you think
the police sides with any particular group or remains impartial?
In Q32a and Q32e the possible response options were ‘yes’ and ‘no’. In Q18 and Q37, the response options
were ‘sides with a particular caste group’ and ‘remains impartial’. Across all the questions, a no response
category was also provided, in case the respondent refused to answer the question.
Step 1: In Q32a Q32b, Q32c, Q32d, and Q32e, a ‘yes’ answer was scored as 2, a ‘no’ answer was scored as 1
and a no response was scored as 0. In Q18and Q37, a ‘sides with a particular caste group’ answer was scored

as 2, a ‘remains impartial’ answer was scored as 1 and a no response was scored as 0.
to 14.
Step 3: These summated scores were then distributed across five newly created categories that indicated the
extent to which people thought the police were discriminatory –
A total score ranging from 11 to 14 was categorised as ‘Highly discriminates’.
A total score ranging from 6 to 10 was categorised as ‘Somewhat discriminates’.
A total score ranging from 3 to 5 was categorised as ‘Rarely discriminates’.
A total score of 1 or 2 was categorised as ‘Very rarely discriminates’.
Index of perception of police diversity
Q23a. Looking at the present number of Scheduled Castes (Dalits) in the police force, are there adequate
numbers of the community in the police?
Q23b. Looking at the present number of OBC in the police force, are there adequate numbers of the
community in the police?
Q23c. Looking at the present number of Scheduled Tribes (Adivasis) in the police force, are there adequate
numbers of the community in the police?
Q23d. Looking at the present number of Muslims in the police force, are there adequate numbers of the
community in the police?
In all the questions the possible response options were ‘yes, adequate’, ‘no, less than adequate’, and ‘more
than adequate’. Across all the questions, a no response category was also provided, in case the respondent
refused to answer the question.
Step 1: A ‘yes, adequate’ answer and a ‘more than adequate’ answer were scored as 2, a ‘no, less than
adequate’ answer was scored as 1 and a no response was scored as 0.
to 10.
extent to which people saw the police as being diverse/representative –
A total score of 9 or 10 was categorised as ‘View police as highly diverse/representative’.
A total score of 7 or 8 was categorised as ‘View police as somewhat diverse/representative’.
A total score ranging from 4 to 6 was categorised as ‘View police as not much diverse/representative’.
A total score ranging from 1 to 3 was categorised as ‘View police as not at all diverse/representative’.
Index of sympathetic attitude towards police working conditions
Q16. Now I will read out two statements. Please tell me which statement would you agree most with?
Statement 1 - Police is not able to function properly due to lack of training and other resources.
Statement 2 - It is not that the police lack resources; they are in fact lazy and not motivated to serve people.
Q25b. Please tell me whether you agree or disagree with - Compared to other professions, it is difficult to
work in the police force because of high stress levels and long working hours? (Probe further whether fully or
somewhat agree or disagree)
Q41. In comparison to other jobs, are the working hours of the police greater or less?

In Q16, the response options were ‘agree with statement 1’ or ‘agree with statement 2’. In Q25b, the response
options were ‘fully agree’, ‘somewhat agree’, ‘somewhat disagree’ and ‘fully disagree’. In Q41, the response
options were ‘greater’, ‘less’ and ‘same as other jobs’. Across all the questions, a no response category was
also provided, in case the respondent refused to answer the question.
Step 1: In Q16 an ‘agree with statement 1’ answer was scored as 4, an ‘agree with statement 2’ answer was
scored as 1 and a no response was scored as 0. In Q25b, a ‘fully agree’ answer was scored as 4, a ‘somewhat
agree’ answer was scored as 3, a ‘somewhat disagree’ answer was scored as 2, a ‘fully disagree’ answer was
scored as 1 and a no response was scored as 0. In Q41, a ‘greater’ answer was scored as 4, a ‘less’ or ‘same as
other jobs’ answer was scored as 1 and a no response was scored as 0.
to 12.
Step 3: These summated scores were then distributed across four newly created categories that indicated the
extent of people’s sympathy with police working conditions –
A total score ranging from 8 to 12 was categorised as ‘Strongly believes that police works under stressful
conditions’.
A total score ranging from 4 to 7 was categorised as ‘Somewhat believes that police works under stressful
conditions’.
A total score ranging from 1 to 3 was categorised as ‘Does not believe police works under stressful conditions’.
A total score of score of 0 was treated as being ‘Non-committal’.
Index of favorability towards women in police
Q33a. Being in the police requires physical strength and aggressive behavior which women lack. Please tell
me whether this argument is justified or not? (Probe further whether very or somewhat justified or unjustified).
Q33b. A woman should prioritize managing home instead of joining the police force. Please tell me whether
this argument is justified or not? (Probe further whether very or somewhat justified or unjustified).
Q33c. Women police are incapable of handling high intensity crimes and cases. Please tell me whether this
argument is justified or not? (Probe further whether very or somewhat justified or unjustified).
Q11a. Usually both men and women work in the police force. In your opinion who is more honest?
Q11b. Usually both men and women work in the police force. In your opinion who is more hardworking?
Q11c. And whom would you approach for help?
In Q33a, Q33b, and Q33c, the response options offered were ‘fully justified’, ‘somewhat justified’, ‘somewhat
unjustified’ and ‘fully unjustified’. In Q11a, Q11b, Q11c, and Q11d. the response options offered were ‘police
women’, ‘police man’, ‘both’ and ‘neither’. Across all the questions, a no response category was also provided,
in case the respondent refused to answer the question.
Step 1: In Q33a, Q33b and Q33c, a ‘fully justified’ answer was scored as 1, a ‘somewhat justified’ answer was
scored as 2, a ‘somewhat unjustified’ answer was scored as 3, a ‘fully unjustified’ answer was scored as 4 and a
no response was scored as 0. In Q11a, Q11b, Q11c and Q11d, a ‘police woman’ or ‘both’ answer was scored as
4, a ‘police man’ or ‘neither’ answer was scored as 1 and a no response was scored as 0.
Step 2: The scores of all questions were summed up. The summated scores of all questions ranged from 0 to 24.
extent of people’s favorability to women in police –
A total score ranging from 20 to 24 was categorised as ‘Very favourable’.
A total score ranging from 14 to 19 was categorised as ‘Somewhat favourable’.
A total score ranging from 6 to 13 was categorised as ‘Not much favourable’.
A total score ranging from 1 to 5 was categorised as ‘Not at all favourable’.

Appendix 4: Details of State Rankings
Index of trust in police
The state rankings for the Index of trust in police are based on summated scores that were arrived at after
weighting each Index category. The ‘highly distrust’ category was weighted as -0.2, the ‘somewhat distrust’
category was weighted as -0.1, the ‘somewhat trust’ category was weighted as 0.1, and the ‘highly trust’
category was weighted as 0.2. The category of non-committal (those who did not answer any question that
went into making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher
summated score here indicates a greater trust.
State Highly After Somewhat After Somewhat After Highly After Non After Score Rank
distrust W1 distrust W2 trust W3 trust W4 committal W5
(%) (%) (%) (%) (%)
Assam 1.7 -0.3 14.8 -1.5 25.9 2.6 57.0 11.4 0.6 0.0 12.2 1
Jharkhand 3.4 -0.7 8.0 -0.8 44.5 4.5 44.1 8.8 0.0 0.0 11.8 2
Haryana 1.0 -0.2 9.9 -1.0 54.5 5.5 34.6 6.9 0.0 0.0 11.2 3
Himachal 3.2 -0.6 14.5 -1.5 35.9 3.6 43.0 8.6 3.4 0.0 10.1 4
Pradesh
West Bengal 2.6 -0.5 22.4 -2.2 25.0 2.5 47.2 9.4 2.8 0.0 9.2 5
Uttarakhand 3.2 -0.6 16.7 -1.7 57.4 5.7 22.2 4.4 0.5 0.0 7.9 6
Kerala 7.4 -1.5 14.1 -1.4 47.4 4.7 29.9 6.0 1.3 0.0 7.8 7
Tamil Nadu 9.8 -2.0 17.6 -1.8 43.1 4.3 29.2 5.8 0.4 0.0 6.4 8
Odisha 10.6 -2.1 15.2 -1.5 46.5 4.7 26.3 5.3 1.4 0.0 6.3 9
Karnataka 4.6 -0.9 20.9 -2.1 57.3 5.7 17.1 3.4 0.1 0.0 6.1 10
Bihar 6.0 -1.2 23.8 -2.4 43.4 4.3 26.6 5.3 0.1 0.0 6.1 11
Nagaland 2.9 -0.6 34.7 -3.5 31.5 3.2 30.7 6.1 0.2 0.0 5.2 12
Andhra 12.0 -2.4 17.6 -1.8 49.5 5.0 19.7 3.9 1.3 0.0 4.7 13
Pradesh
Madhya 3.5 -0.7 27.4 -2.7 55.8 5.6 12.9 2.6 0.3 0.0 4.7 14
Pradesh
Maharashtra 8.8 -1.8 23.6 -2.4 51.5 5.2 14.6 2.9 1.5 0.0 4.0 15
Chhattisgarh 7.2 -1.4 25.3 -2.5 43.8 4.4 17.7 3.5 6.1 0.0 4.0 16
Gujarat 9.8 -2.0 24.9 -2.5 47.3 4.7 16.5 3.3 1.5 0.0 3.6 17
Delhi 12.1 -2.4 23.3 -2.3 44.9 4.5 19.1 3.8 0.6 0.0 3.6 18
Telangana 6.5 -1.3 24.6 -2.5 58.6 5.9 7.2 1.4 3.1 0.0 3.5 19
Punjab 8.2 -1.6 41.3 -4.1 31.9 3.2 17.4 3.5 1.2 0.0 0.9 20
Uttar 10.7 -2.1 35.7 -3.6 42.5 4.3 8.3 1.7 2.7 0.0 0.2 21
Pradesh
Rajasthan 17.0 -3.4 38.4 -3.8 30.8 3.1 9.9 2.0 3.9 0.0 -2.2 22

Index of fear of police
The state rankings for the Index of pear of police are based on summated scores that were arrived at after
weighting each Index category. The ‘highly fearful’ category was weighted as -0.2, the ‘somewhat fearful’
category was weighted as -0.1, the ‘not much fearful’ category was weighted as 0.1, and the ‘not at all fearful’
category was weighted as 0.2. The category of non-committal (those who did not answer any question that
went into making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher
summated score here indicates a positive assessment, i.e., lesser fear.
State Not After Not After Somewhat After Highly After Non- After Score Rank
at all W1 much W2 fearful W3 fearful W4 committal W5
fearful fearful (%) (%) (%)
(%) (%)
Himachal 83.2 16.6 9.1 0.9 2.9 -0.3 0.2 -0.04 4.5 0.0 17.2 1
Pradesh
Uttarakhand 71.0 14.2 17.6 1.8 9.6 -1.0 1.4 -0.28 0.5 0.0 14.7 2
Haryana 53.7 10.7 39.9 4.0 4.0 -0.4 2.3 -0.46 0.2 0.0 13.9 3
Kerala 51.9 10.4 27.1 2.7 12.9 -1.3 3.8 -0.76 4.4 0.0 11.0 4
Delhi 58.0 11.6 19.3 1.9 14.9 -1.5 6.2 -1.24 1.6 0.0 10.8 5
Rajasthan 41.6 8.3 25.0 2.5 18.7 -1.9 2.2 -0.44 12.4 0.0 8.5 6
Maharashtra 39.1 7.8 30.5 3.1 23.7 -2.4 4.6 -0.92 2.1 0.0 7.6 7
Nagaland 27.7 5.5 39.5 4.0 24.0 -2.4 4.4 -0.88 4.4 0.0 6.2 8
Chhattisgarh 29.5 5.9 30.9 3.1 23.2 -2.3 4.2 -0.84 12.2 0.0 5.8 9
Assam 25.5 5.1 32.5 3.3 31.9 -3.2 6.9 -1.38 3.2 0.0 3.8 10
Gujarat 27.0 5.4 31.1 3.1 25.8 -2.6 14.0 -2.80 2.1 0.0 3.1 11
West Bengal 27.0 5.4 29.5 3.0 26.3 -2.6 13.8 -2.76 3.5 0.0 3.0 12
Bihar 26.8 5.4 24.5 2.5 43.4 -4.3 4.8 -0.96 0.5 0.0 2.5 13
Madhya 11.0 2.2 35.3 3.5 43.7 -4.4 8.0 -1.60 2.0 0.0 -0.2 14
Pradesh
Odisha 24.5 4.9 17.3 1.7 26.2 -2.6 23.2 -4.64 8.8 0.0 -0.6 15
Jharkhand 19.0 3.8 17.6 1.8 56.6 -5.7 6.4 -1.28 0.4 0.0 -1.4 16
Uttar 15.4 3.1 20.4 2.0 41.5 -4.2 17.8 -3.56 4.9 0.0 -2.6 17
Pradesh
Telangana 9.5 1.9 22.4 2.2 54.4 -5.4 8.9 -1.78 4.8 0.0 -3.1 18
Andhra 9.0 1.8 17.1 1.7 43.4 -4.3 25.1 -5.02 5.3 0.0 -5.9 19
Pradesh
Tamil Nadu 4.8 1.0 14.8 1.5 33.5 -3.4 39.2 -7.84 7.6 0.0 -8.8 20
Karnataka 3.8 0.8 14.7 1.5 44.1 -4.4 33.9 -6.78 3.6 0.0 -9.0 21
Punjab 6.1 1.2 9.8 1.0 20.9 -2.1 46.7 -9.34 16.4 0.0 -9.2 22

Index of perception about the police
The state rankings for the Index of perception about the police are based on summated scores that were arrived
at after weighting each Index category. The ‘very negative’ category was weighted as -0.2, the ‘somewhat
negative’ category was weighted as -0.1, the ‘somewhat positive’ category was weighted as 0.1, and the
‘very positive’ category was weighted as 0.2. The category of non-committal (those who did not answer any
question that went into making the Index) was weighted as 0 and hence excluded from the ranking analysis.
A higher summated score here indicates a positive assessment, i.e., a better perception of the police.
State Very After Somewhat After Somewhat After Very After Non After Score Rank
negative W1 negative W2 positive W3 positive W4 committal W5
(%) (%) (%) (%) (%)
Haryana 2.9 -0.6 3.8 -0.4 22.4 2.2 70.9 14.2 0.0 0.0 15.5 1
Himachal 3.6 -0.7 4.5 -0.5 22.0 2.2 69.8 14.0 0.0 0.0 15.0 2
Pradesh
Jharkhand 5.6 -1.1 4.6 -0.5 42.5 4.3 46.7 9.3 0.6 0.0 12.0 3
Nagaland 9.5 -1.9 6.4 -0.6 47.0 4.7 37.2 7.4 0.0 0.0 9.6 4
Kerala 9.5 -1.9 10.3 -1.0 37.7 3.8 41.4 8.3 1.1 0.0 9.1 5
Bihar 8.4 -1.7 13.0 -1.3 48.3 4.8 30.3 6.1 0.0 0.0 7.9 6
Uttarakhand 12.0 -2.4 14.1 -1.4 43.6 4.4 29.6 5.9 0.6 0.0 6.5 7
Andhra 14.7 -2.9 15.3 -1.5 44.5 4.5 24.9 5.0 0.5 0.0 5.0 8
Pradesh
Karnataka 17.2 -3.4 12.3 -1.2 44.7 4.5 25.7 5.1 0.0 0.0 4.9 9
Assam 16.5 -3.3 13.7 -1.4 45.7 4.6 24.0 4.8 0.0 0.0 4.7 10
Maharashtra 16.6 -3.3 15.9 -1.6 40.0 4.0 25.8 5.2 1.7 0.0 4.3 11
Telangana 15.8 -3.2 16.9 -1.7 44.8 4.5 19.9 4.0 2.7 0.0 3.6 12
Odisha 23.9 -4.8 11.2 -1.1 37.7 3.8 26.1 5.2 1.1 0.0 3.1 13
Madhya 14.2 -2.8 25.4 -2.5 41.8 4.2 18.7 3.7 0.0 0.0 2.5 14
Pradesh
Gujarat 21.7 -4.3 18.5 -1.9 34.2 3.4 25.6 5.1 0.0 0.0 2.4 15
Delhi 21.8 -4.4 20.8 -2.1 34.9 3.5 22.2 4.4 0.2 0.0 1.5 16
Tamil Nadu 17.8 -3.6 26.0 -2.6 40.2 4.0 15.9 3.2 0.1 0.0 1.0 17
Rajasthan 29.5 -5.9 12.4 -1.2 39.7 4.0 14.5 2.9 3.9 0.0 -0.3 18
Chhattisgarh 31.5 -6.3 16.8 -1.7 32.8 3.3 18.3 3.7 0.6 0.0 -1.0 19
West Bengal 31.9 -6.4 13.9 -1.4 35.6 3.6 15.7 3.1 2.9 0.0 -1.1 20
Uttar 26.2 -5.2 27.6 -2.8 37.2 3.7 8.2 1.6 0.8 0.0 -2.6 21
Pradesh
Punjab 34.6 -6.9 17.8 -1.8 37.8 3.8 9.4 1.9 0.4 0.0 -3.0 22

Index of perception about police independence
The state rankings for the Index of perception about police independence are based on summated scores that
were arrived at after weighting each Index category. The ‘not at all independent’ category was weighted as
-0.2, the ‘not much independent’ category was weighted as -0.1, the ‘somewhat independent’ category was
weighted as 0.1, and the ‘highly independent’ category was weighted as 0.2. The category of non-committal
(those who did not answer any question that went into making the Index) was weighted as 0 and hence
excluded from the ranking analysis. A higher summated score here indicates a positive assessment, i.e., a
better perception of the police’s independence.
State Not at all After Not much After Somewhat After Highly After Non After Score Rank
independent W1 independent W2 independent W3 independent W4 committal W5
(%) (%) (%) (%) (%)
Kerala 2.1 -0.4 12.4 -1.2 14.4 1.4 64.8 13.0 6.4 0.0 12.7 1
Assam 2.4 -0.5 17.3 -1.7 52.2 5.2 20.6 4.1 7.5 0.0 7.1 2
Nagaland 7.5 -1.5 19.5 -2.0 42.1 4.2 24.0 4.8 6.9 0.0 5.6 3
Himachal 0.7 -0.1 22.5 -2.3 52.5 5.3 10.2 2.0 14.1 0.0 4.9 4
Pradesh
Gujarat 2.3 -0.5 30.3 -3.0 42.6 4.3 17.1 3.4 7.7 0.0 4.2 5
Haryana 0.2 0.0 36.5 -3.7 43.2 4.3 17.0 3.4 3.1 0.0 4.0 6
Maharashtra 3.3 -0.7 28.3 -2.8 52.2 5.2 10.8 2.2 5.5 0.0 3.9 7
Rajasthan 1.4 -0.3 29.6 -3.0 45.0 4.5 9.8 2.0 14.1 0.0 3.2 8
Bihar 1.2 -0.2 41.7 -4.2 42.7 4.3 11.8 2.4 2.6 0.0 2.2 9
Jharkhand 6.2 -1.2 32.0 -3.2 52.0 5.2 7.0 1.4 2.8 0.0 2.2 10
Andhra 7.3 -1.5 27.6 -2.8 51.2 5.1 6.3 1.3 7.5 0.0 2.2 11
Pradesh
Karnataka 1.7 -0.3 43.8 -4.4 37.9 3.8 12.8 2.6 3.8 0.0 1.6 12
Uttarakhand 2.9 -0.6 40.7 -4.1 44.9 4.5 7.9 1.6 3.6 0.0 1.4 13
Madhya 1.8 -0.4 42.9 -4.3 51.0 5.1 3.7 0.7 0.6 0.0 1.2 14
Pradesh
Tamil Nadu 0.3 -0.1 44.9 -4.5 32.4 3.2 10.7 2.1 11.7 0.0 0.8 15
Telangana 3.6 -0.7 39.6 -4.0 42.1 4.2 6.4 1.3 8.3 0.0 0.8 16
West Bengal 11.4 -2.3 27.3 -2.7 32.8 3.3 12.0 2.4 16.5 0.0 0.7 17
Uttar 2.7 -0.5 43.6 -4.4 39.3 3.9 8.0 1.6 6.3 0.0 0.6 18
Pradesh
Chhattisgarh 13.9 -2.8 23.2 -2.3 36.3 3.6 7.2 1.4 19.4 0.0 0.0 19
Punjab 4.5 -0.9 43.2 -4.3 42.8 4.3 3.7 0.7 5.7 0.0 -0.2 20
Delhi 4.8 -1.0 48.6 -4.9 33.9 3.4 5.4 1.1 7.3 0.0 -1.4 21
Odisha 9.8 -2.0 40.4 -4.0 29.4 2.9 3.1 0.6 17.3 0.0 -2.4 22

Index of perception of discrimination by the police
The state rankings for the Index of perception of discrimination by the police are based on summated scores
that were arrived at after weighting each Index category. The ‘highly discriminates’ category was weighted
as -0.2, the ‘somewhat discriminates’ category was weighted as -0.1, the ‘rarely discriminates’ category was
weighted as 0.1, and the ‘very rarely discriminates’ category was weighted as 0.2. The category of non-
committal (those who did not answer any question that went into making the Index) was weighted as 0 and
hence excluded from the ranking analysis. A higher summated score here indicates a positive assessment, i.e.,
weaker perception of discrimination.
State Very rarely After Rarely After Somewhat After Highly After Non After Score Rank
discriminates W1 discriminates W2 discriminates W3 discriminates W4 committal W5
(%) (%) (%) (%) (%)
West Bengal 19.9 4.0 17.3 1.7 47.5 -4.8 4.7 -0.9 10.7 0.0 0.0 1
Chhattisgarh 17.1 3.4 15.6 1.6 57.1 -5.7 0.6 -0.1 9.7 0.0 -0.9 2
Odisha 15.0 3.0 15.3 1.5 60.8 -6.1 1.2 -0.2 7.7 0.0 -1.8 3
Rajasthan 16.4 3.3 16.0 1.6 55.2 -5.5 8.4 -1.7 4.1 0.0 -2.3 4
Nagaland 4.9 1.0 26.7 2.7 65.3 -6.5 1.1 -0.2 2.0 0.0 -3.1 5
Uttarakhand 8.3 1.7 18.2 1.8 66.9 -6.7 4.1 -0.8 2.4 0.0 -4.0 6
Madhya 4.3 0.9 24.4 2.4 66.4 -6.6 3.8 -0.8 1.0 0.0 -4.1 7
Pradesh
Andhra 8.9 1.8 16.3 1.6 69.8 -7.0 2.8 -0.6 2.2 0.0 -4.1 8
Pradesh
Himachal 4.1 0.8 18.1 1.8 70.5 -7.1 0.2 0.0 7.0 0.0 -4.5 9
Pradesh
Uttar 9.4 1.9 15.1 1.5 62.7 -6.3 9.6 -1.9 3.1 0.0 -4.8 10
Pradesh
Kerala 8.5 1.7 13.4 1.3 71.9 -7.2 3.3 -0.7 2.9 0.0 -4.8 11
Gujarat 7.7 1.5 13.1 1.3 72.3 -7.2 4.2 -0.8 2.7 0.0 -5.2 12
Punjab 4.7 0.9 3.7 0.4 66.6 -6.7 3.9 -0.8 21.1 0.0 -6.1 13
Telangana 2.6 0.5 13.9 1.4 77.6 -7.8 3.6 -0.7 2.3 0.0 -6.6 14
Assam 5.4 1.1 10.9 1.1 77.0 -7.7 5.8 -1.2 0.9 0.0 -6.7 15
Maharashtra 4.0 0.8 8.7 0.9 73.4 -7.3 12.0 -2.4 1.9 0.0 -8.1 16
Tamil Nadu 4.7 0.9 9.6 1.0 56.9 -5.7 21.8 -4.4 7.0 0.0 -8.2 17
Karnataka 3.6 0.7 7.7 0.8 65.1 -6.5 17.8 -3.6 5.8 0.0 -8.6 18
Delhi 5.6 1.1 7.2 0.7 68.2 -6.8 18.5 -3.7 0.6 0.0 -8.7 19
Haryana 4.6 0.9 0.0 0.0 86.4 -8.6 8.4 -1.7 0.6 0.0 -9.4 20
Jharkhand 0.4 0.1 3.6 0.4 88.6 -8.9 7.0 -1.4 0.4 0.0 -9.8 21
Bihar 0.2 0.0 0.7 0.1 66.8 -6.7 31.8 -6.4 0.4 0.0 -12.9 22

Index of sympathetic attitudes towards police working conditions
The state rankings for the Index of sympathetic attitude towards police working conditions are based on
summated scores that were arrived at after weighting each Index category. The ‘no sympathy’ category was
weighted as 2, the ‘somewhat sympathy’ category was weighted as 3, the ‘strong sympathy category was
weighted as 5 and the category of non-committal (those who did not answer any question that went into
making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher summated
score here indicates more sympathy.
State No After Moderate After Strong After Non After Score Rank
sympathy W1 sympathy W2 sympathy W3 committal W4
(%) (%) (%) (%)
Telangana 3.0 0.6 24.8 7.4 70.6 35.3 1.7 0.0 43.3 1
Andhra 6.6 1.3 30.2 9.1 60.2 30.1 3.0 0.0 40.5 2
Pradesh
Karnataka 3.0 0.6 42.1 12.6 54.1 27.1 0.9 0.0 40.3 3
Tamil Nadu 5.3 1.1 36.0 10.8 56.7 28.4 2.0 0.0 40.2 4
Assam 6.4 1.3 36.6 11.0 55.7 27.9 1.3 0.0 40.1 5
Kerala 7.0 1.4 35.1 10.5 54.3 27.2 3.6 0.0 39.1 6
Maharashtra 5.8 1.2 34.1 10.2 55.2 27.6 4.8 0.0 39.0 7
Bihar 9.0 1.8 40.7 12.2 49.6 24.8 0.7 0.0 38.8 8
Nagaland 9.1 1.8 42.7 12.8 45.5 22.8 2.7 0.0 37.4 9
Delhi 14.3 2.9 41.0 12.3 44.2 22.1 0.4 0.0 37.3 10
Punjab 7.0 1.4 52.7 15.8 38.1 19.1 2.3 0.0 36.3 11
Uttarakhand 16.1 3.2 43.9 13.2 37.4 18.7 2.6 0.0 35.1 12
Jharkhand 4.4 0.9 63.4 19.0 29.2 14.6 3.0 0.0 34.5 13
Odisha 20.6 4.1 43.5 13.1 30.8 15.4 5.1 0.0 32.6 14
Gujarat 23.6 4.7 44.2 13.3 27.8 13.9 4.4 0.0 31.9 15
Rajasthan 23.7 4.7 34.2 10.3 33.5 16.8 8.6 0.0 31.8 16
Madhya 20.2 4.0 38.1 11.4 32.3 16.2 9.5 0.0 31.6 17
Pradesh
Haryana 15.3 3.1 68.2 20.5 15.5 7.8 1.0 0.0 31.3 18
West Bengal 13.2 2.6 37.4 11.2 34.7 17.4 14.7 0.0 31.2 19
Himachal 32.5 6.5 32.3 9.7 30.0 15.0 5.2 0.0 31.2 20
Pradesh
Chhattisgarh 16.4 3.3 42.3 12.7 29.7 14.9 11.6 0.0 30.8 21
Uttar 27.4 5.5 53.0 15.9 14.1 7.1 5.5 0.0 28.4 22
Pradesh

Index of favorability towards women in the police
The state rankings for the Index of favorability towards women in the police are based on summated scores that were
arrived at after weighting each Index category. The ‘not at all favourable’ category was weighted as -0.2, the ‘not
much favourable’ category was weighted as -0.1 the ‘somewhat favourable’ category was weighted as 0.1 and the ‘very
favourable’ category as 0.2. The category of non-committal (those who did not answer any question that went into
making the Index) was weighted as 0 and hence excluded from the ranking analysis. A higher summated score here
indicates more favorability.
Not at all After Not much After Somewhat After Very After Non After Score Rank
favourable W1 favourable W2 favourable W3 favourable W4 committal W5
(%) (%) (%) (%) (%)
Uttarakhand 1.7 -0.3 11.1 -1.1 39.8 4.0 47.4 9.5 0.0 0.0 12.0 1
Haryana 1.5 -0.3 13.0 -1.3 37.7 3.8 47.8 9.6 0.0 0.0 11.7 2
Maharashtra 1.5 -0.3 14.9 -1.5 37.0 3.7 45.2 9.0 1.4 0.0 11.0 3
Nagaland 0.0 0.0 15.8 -1.6 43.3 4.3 40.7 8.1 0.2 0.0 10.9 4
Delhi 2.4 -0.5 20.1 -2.0 40.6 4.1 36.4 7.3 0.6 0.0 8.9 5
Jharkhand 1.6 -0.3 15.0 -1.5 64.0 6.4 19.4 3.9 0.0 0.0 8.5 6
Kerala 2.0 -0.4 24.1 -2.4 32.7 3.3 39.3 7.9 2.0 0.0 8.3 7
Punjab 1.6 -0.3 21.5 -2.2 53.5 5.4 23.2 4.6 0.2 0.0 7.5 8
Odisha 4.0 -0.8 21.9 -2.2 40.4 4.0 30.4 6.1 3.3 0.0 7.1 9
Tamil Nadu 2.0 -0.4 30.5 -3.1 44.8 4.5 21.7 4.3 1.1 0.0 5.4 10
Madhya 0.3 -0.1 34.2 -3.4 43.7 4.4 21.5 4.3 0.2 0.0 5.2 11
Pradesh
Gujarat 1.9 -0.4 29.0 -2.9 49.8 5.0 17.4 3.5 1.9 0.0 5.2 12
Himachal 0.9 -0.2 43.6 -4.4 12.3 1.2 39.1 7.8 4.1 0.0 4.5 13
Pradesh
Chhattisgarh 2.7 -0.5 34.9 -3.5 31.6 3.2 26.7 5.3 4.0 0.0 4.5 14
Assam 3.6 -0.7 35.1 -3.5 38.3 3.8 21.8 4.4 1.1 0.0 4.0 15
Bihar 1.2 -0.2 33.7 -3.4 57.4 5.7 7.4 1.5 0.4 0.0 3.6 16
Rajasthan 3.9 -0.8 37.1 -3.7 35.6 3.6 19.9 4.0 3.4 0.0 3.1 17
Andhra 4.5 -0.9 37.7 -3.8 46.5 4.7 9.4 1.9 1.9 0.0 1.9 18
Pradesh
West Bengal 6.8 -1.4 38.2 -3.8 33.1 3.3 16.9 3.4 4.9 0.0 1.5 19
Karnataka 3.8 -0.8 44.0 -4.4 42.2 4.2 9.5 1.9 0.5 0.0 1.0 20
Uttar 6.2 -1.2 42.6 -4.3 38.3 3.8 10.2 2.0 2.7 0.0 0.4 21
Pradesh
Telangana 5.1 -1.0 48.1 -4.8 33.7 3.4 9.9 2.0 3.3 0.0 -0.5 22

Q12a
The state rankings for Q12a (Please tell me how much trust do you have in local police like police inspector, Sub
inspector, SHO - a lot, somewhat, not much, or not at all?) are based on summated scores that were arrived at after
weighting each response option. An ‘a lot’ answer was weighted as 0.2, a ‘somewhat’ answer was weighted as
0.1, a ‘not much’ answer was weighted as -0.1, and a ‘not at all’ answer was weighted as -0.2. The category of
no response (those who did not answer the question) was weighted as 0 and hence excluded from the ranking
analysis. A higher summated score here indicates a more positive assessment.
State A lot After Somewhat After Not much After Not at After No response After Score Rank
(%) W1 (%) W2 (%) W3 all (%) W4 (%) W5
Andhra 35.9 7.2 42.1 4.2 5.7 -0.6 3.6 -0.7 12.7 0.0 10.1 1
Pradesh
Jharkhand 26.3 5.3 58.3 5.8 9.2 -0.9 0.8 -0.2 5.4 0.0 10.0 2
Telangana 26.4 5.3 55.9 5.6 5.4 -0.5 1.8 -0.4 10.5 0.0 10.0 3
Kerala 43.1 8.6 32.6 3.3 11.3 -1.1 5.1 -1.0 8.0 0.0 9.7 4
Odisha 33.1 6.6 40.3 4.0 6.7 -0.7 4.2 -0.8 15.7 0.0 9.1 5
Himachal 13.9 2.8 72.0 7.2 6.1 -0.6 2.3 -0.5 5.7 0.0 8.9 6
Pradesh
Uttarakhand 23.7 4.7 57.6 5.8 10.0 -1.0 6.2 -1.2 2.4 0.0 8.3 7
Tamil Nadu 30.6 6.1 45.1 4.5 11.2 -1.1 6.6 -1.3 6.5 0.0 8.2 8
Madhya 25.3 5.1 53.4 5.3 15.1 -1.5 4.0 -0.8 2.3 0.0 8.1 9
Pradesh
Karnataka 26.1 5.2 49.7 5.0 12.9 -1.3 5.2 -1.0 6.0 0.0 7.9 10
Haryana 11.3 2.3 72.0 7.2 14.0 -1.4 2.1 -0.4 0.6 0.0 7.6 11
Chhattisgarh 22.6 4.5 46.0 4.6 11.8 -1.2 3.8 -0.8 15.8 0.0 7.2 12
Assam 18.5 3.7 56.1 5.6 14.4 -1.4 4.7 -0.9 6.2 0.0 6.9 13
Maharashtra 20.5 4.1 49.5 5.0 11.9 -1.2 8.6 -1.7 9.6 0.0 6.1 14
Gujarat 22.8 4.6 47.0 4.7 16.2 -1.6 7.8 -1.6 6.1 0.0 6.1 15
Punjab 25.1 5.0 37.4 3.7 19.1 -1.9 8.4 -1.7 10.1 0.0 5.2 16
Delhi 13.9 2.8 50.5 5.1 16.7 -1.7 9.7 -1.9 9.1 0.0 4.2 17
West Bengal 18.1 3.6 39.9 4.0 14.1 -1.4 10.2 -2.0 17.6 0.0 4.2 18
Nagaland 16.8 3.4 47.0 4.7 23.7 -2.4 9.5 -1.9 3.1 0.0 3.8 19
Bihar 16.3 3.3 44.0 4.4 27.4 -2.7 10.9 -2.2 1.3 0.0 2.7 20
Uttar 9.4 1.9 48.4 4.8 19.4 -1.9 15.7 -3.1 7.1 0.0 1.6 21
Pradesh
Rajasthan 6.2 1.2 43.5 4.4 27.7 -2.8 6.8 -1.4 15.9 0.0 1.5 22

Q12b.
The state rankings for Q12b (Please tell me how much trust do you have in a senior police officer like SP, DCP - a
lot, somewhat, not much, or not at all?) are based on summated scores that were arrived at after weighting
each response option. An ‘a lot’ answer was weighted as 0.2, a ‘somewhat’ answer was weighted as 0.1, a
‘not much’ answer was weighted as -0.1, and a ‘not at all’ answer was weighted as 0.2. The category of no
response (those who did not answer the question) was weighted as 0 and hence excluded from the ranking
analysis. A higher summated score here indicates a more positive assessment.
State A lot After Somewhat After Not much After Not at After No response After Score Rank
(%) W1 (%) W2 (%) W3 all (%) W4 (%) W5
Haryana 31.0 6.2 60.3 6.0 6.9 -0.7 1.0 -0.2 0.8 0.0 11.3 1
Himachal 25.2 5.0 63.0 6.3 3.9 -0.4 2.0 -0.4 5.9 0.0 10.6 2
Pradesh
Odisha 37.9 7.6 33.4 3.3 6.7 -0.7 2.5 -0.5 19.5 0.0 9.8 3
Kerala 43.9 8.8 31.3 3.1 9.0 -0.9 6.5 -1.3 9.3 0.0 9.7 4
Bihar 47.5 9.5 29.2 2.9 15.3 -1.5 6.8 -1.4 1.4 0.0 9.5 5
Uttarakhand 35.1 7.0 44.1 4.4 15.5 -1.6 2.6 -0.5 2.7 0.0 9.4 6
Andhra 28.8 5.8 42.1 4.2 11.1 -1.1 2.4 -0.5 15.6 0.0 8.4 7
Pradesh
Jharkhand 46.4 9.3 19.6 2.0 26.6 -2.7 1.4 -0.3 6.0 0.0 8.3 8
Telangana 28.9 5.8 40.6 4.1 13.6 -1.4 2.1 -0.4 14.8 0.0 8.1 9
Assam 23.2 4.6 50.6 5.1 13.1 -1.3 2.1 -0.4 10.9 0.0 8.0 10
Madhya 32.7 6.5 38.5 3.9 19.7 -2.0 3.5 -0.7 5.7 0.0 7.7 11
Pradesh
Delhi 28.2 5.6 41.6 4.2 10.7 -1.1 5.6 -1.1 13.9 0.0 7.6 12
Nagaland 20.4 4.1 54.3 5.4 19.1 -1.9 2.9 -0.6 3.3 0.0 7.0 13
Chhattisgarh 30.1 6.0 33.7 3.4 15.2 -1.5 4.6 -0.9 16.4 0.0 7.0 14
Tamil Nadu 27.4 5.5 40.5 4.1 17.5 -1.8 7.5 -1.5 7.0 0.0 6.3 15
Karnataka 24.6 4.9 43.1 4.3 19.6 -2.0 5.8 -1.2 6.9 0.0 6.1 16
West Bengal 24.1 4.8 36.9 3.7 11.3 -1.1 6.7 -1.3 21.0 0.0 6.0 17
Maharashtra 20.5 4.1 44.3 4.4 16.4 -1.6 5.5 -1.1 13.4 0.0 5.8 18
Gujarat 24.6 4.9 41.7 4.2 18.3 -1.8 7.7 -1.5 7.7 0.0 5.7 19
Punjab 26.2 5.2 29.5 3.0 24.6 -2.5 7.8 -1.6 11.9 0.0 4.2 20
Uttar Pradesh 19.6 3.9 40.7 4.1 20.1 -2.0 11.2 -2.2 8.4 0.0 3.7 21
Rajasthan 13.0 2.6 37.7 3.8 20.8 -2.1 11.9 -2.4 16.5 0.0 1.9 22

Q19
The state rankings for Q19 (How satisfied are you with police performance and their work in your area? (If satisfied
or dissatisfied probe further whether fully or somewhat) are based on summated scores that were arrived at after
weighting each response option. An ‘fully satisfied’ answer was weighted as 0.2, a ‘somewhat satisfied’
answer was weighted as 0.1, a ‘somewhat dissatisfied’ answer was weighted as -0.1, and a ‘fully dissatisfied’
answer was weighted as -0.2. The category of no response (those who did not answer the question) was
weighted as 0 and hence excluded from the ranking analysis. A higher summated score here indicates a more
positive assessment.
State Fully After Somewhat After Somewhat After Fully After No After Score Rank
satisfied W1 satisfied W2 dissatisfied W3 dissatisfied W4 response W5
(%) (%) (%) (%) (%)
Kerala 51.5 10.3 35.4 3.5 2.4 -0.2 1.8 -0.4 8.8 0.0 13.2 1
Himachal 29.5 5.9 61.8 6.2 1.6 -0.2 0.9 -0.2 6.1 0.0 11.7 2
Pradesh
Odisha 36.3 7.3 53.6 5.4 2.0 -0.2 3.6 -0.7 4.4 0.0 11.7 3
Chhattisgarh 42.9 8.6 39.6 4.0 4.2 -0.4 3.2 -0.6 10.1 0.0 11.5 4
Punjab 34.8 7.0 49.8 5.0 3.9 -0.4 2.3 -0.5 9.2 0.0 11.1 5
Gujarat 44.4 8.9 37.9 3.8 10.3 -1.0 4.1 -0.8 3.2 0.0 10.8 6
Haryana 16.3 3.3 72.4 7.2 3.5 -0.4 0.6 -0.1 7.1 0.0 10.0 7
Uttarakhand 31.6 6.3 50.5 5.1 7.8 -0.8 6.8 -1.4 3.3 0.0 9.2 8
Maharashtra 28.7 5.7 51.7 5.2 8.4 -0.8 4.3 -0.9 6.8 0.0 9.2 9
Jharkhand 32.6 6.5 43.0 4.3 21.6 -2.2 2.2 -0.4 0.6 0.0 8.2 10
Assam 22.1 4.4 57.5 5.8 12.4 -1.2 4.1 -0.8 3.9 0.0 8.1 11
Karnataka 18.7 3.7 60.8 6.1 12.1 -1.2 2.5 -0.5 5.9 0.0 8.1 12
Madhya 16.1 3.2 63.8 6.4 12.1 -1.2 2.2 -0.4 5.8 0.0 8.0 13
Pradesh
West Bengal 26.3 5.3 46.7 4.7 16.3 -1.6 2.0 -0.4 8.6 0.0 7.9 14
Telangana 18.6 3.7 58.7 5.9 13.6 -1.4 3.9 -0.8 5.2 0.0 7.5 15
Delhi 21.9 4.4 56.8 5.7 9.2 -0.9 8.6 -1.7 3.6 0.0 7.4 16
Rajasthan 21.5 4.3 54.5 5.5 7.5 -0.8 10.0 -2.0 6.6 0.0 7.0 17
Tamil Nadu 17.1 3.4 57.7 5.8 16.2 -1.6 3.1 -0.6 5.9 0.0 7.0 18
Nagaland 17.1 3.4 53.2 5.3 11.7 -1.2 5.6 -1.1 12.4 0.0 6.5 19
Andhra 27.4 5.5 40.2 4.0 20.4 -2.0 5.8 -1.2 6.2 0.0 6.3 20
Pradesh
Uttar 14.2 2.8 52.5 5.3 11.1 -1.1 12.0 -2.4 10.2 0.0 4.6 21
Pradesh
Bihar 11.2 2.2 52.8 5.3 18.2 -1.8 14.6 -2.9 3.2 0.0 2.8 22

Q22c
The state rankings for Q22c (Please tell me if you agree or disagree with this statement – There is nothing wrong
in the police being violent towards criminals (If agree or disagree probe further whether fully or somewhat) are based
on summated scores that were arrived at after weighting each response option. A ‘fully agree’ answer was
weighted as -0.2, a ‘somewhat agree’ answer was weighted as -0.1, a ‘somewhat disagree’ answer was
weighted as 0.1, and a ‘fully disagree’ answer was weighted as 0.2. The category of no response (those who
did not answer the question) was weighted as 0 and hence excluded from the ranking analysis. A higher
summated score here indicates a more positive assessment.
State Fully After Somewhat After Somewhat After Fully After No response After Score Rank
agree (%) W1 agree (%) W2 disagree (%) W3 disagree (%) W4 (%) W5
Himachal 11.8 -2.4 20.5 -2.1 40.9 4.1 13.2 2.6 13.6 0.0 2.3 1
Pradesh
Odisha 9.4 -1.9 20.6 -2.1 23.2 2.3 19.0 3.8 27.7 0.0 2.2 2
West Bengal 12.0 -2.4 15.0 -1.5 15.0 1.5 21.0 4.2 37.1 0.0 1.8 3
Nagaland 12.2 -2.4 36.4 -3.6 19.3 1.9 28.4 5.7 3.8 0.0 1.5 4
Uttarakhand 24.8 -5.0 19.6 -2.0 11.6 1.2 25.2 5.0 18.8 0.0 -0.7 5
Telangana 13.6 -2.7 29.8 -3.0 19.9 2.0 13.1 2.6 23.7 0.0 -1.1 6
Karnataka 18.5 -3.7 29.6 -3.0 23.5 2.4 14.9 3.0 13.5 0.0 -1.3 7
Chhattisgarh 16.0 -3.2 22.9 -2.3 13.7 1.4 13.9 2.8 33.5 0.0 -1.3 8
Uttar 17.6 -3.5 26.6 -2.7 24.7 2.5 11.7 2.3 19.4 0.0 -1.4 9
Pradesh
Jharkhand 12.8 -2.6 43.6 -4.4 18.2 1.8 18.2 3.6 7.2 0.0 -1.5 10
Madhya 19.7 -3.9 30.0 -3.0 20.8 2.1 12.5 2.5 17.0 0.0 -2.4 11
Pradesh
Maharashtra 19.8 -4.0 36.0 -3.6 21.4 2.1 14.3 2.9 8.6 0.0 -2.6 12
Assam 24.7 -4.9 32.4 -3.2 20.0 2.0 14.6 2.9 8.4 0.0 -3.3 13
Haryana 30.7 -6.1 25.3 -2.5 31.3 3.1 10.6 2.1 2.1 0.0 -3.4 14
Punjab 19.9 -4.0 27.7 -2.8 14.3 1.4 7.8 1.6 30.3 0.0 -3.8 15
Bihar 33.3 -6.7 23.3 -2.3 28.0 2.8 12.0 2.4 3.3 0.0 -3.8 16
Andhra 21.6 -4.3 34.8 -3.5 17.6 1.8 9.1 1.8 17.0 0.0 -4.2 17
Pradesh
Delhi 32.0 -6.4 28.6 -2.9 12.5 1.3 18.7 3.7 8.2 0.0 -4.3 18
Rajasthan 21.6 -4.3 35.7 -3.6 13.0 1.3 7.5 1.5 22.3 0.0 -5.1 19
Kerala 32.0 -6.4 29.2 -2.9 11.6 1.2 15.3 3.1 11.9 0.0 -5.1 20
Gujarat 26.7 -5.3 31.4 -3.1 19.2 1.9 5.8 1.2 16.8 0.0 -5.4 21
Tamil Nadu 31.7 -6.3 29.8 -3.0 14.2 1.4 11.6 2.3 12.6 0.0 -5.6 22

Appendix 5: State-Wise Findings from The Survey on Select
Questions
Table 1:
Q. What kind of police presence would you like to see in your village/area- greater, less or no change? 1.
Greater 2. Less 3. No change, same as before 8. Don’t know
States Respondents who want greater police presence in their locality
Andhra Pradesh 49
Assam 8
Bihar 65
Gujarat 34
Haryana 72
Himachal Pradesh 59
Karnataka 41
Kerala 41
Madhya Pradesh 46
Maharashtra 56
Nagaland 56
Odisha 52
Punjab 30
Rajasthan 41
Tamil Nadu 26
Uttar Pradesh 56
West Bengal 47
Delhi 73
Jharkhand 73
Chhattisgarh 58
Uttarakhand 44
Telangana 35
Overall 47

Table 2:
Q. From time to time, for different purposes, people have some kind of contact with the police. In the last 4-5
years, have you or your family member had any kind of contact with the police?
2. Yes 1. No 8. Don’t know
State Respondents who had some kind of contact with the police in the last 4-5 years
Andhra Pradesh 4
Assam 9
Bihar 34
Gujarat 7
Haryana 9
Himachal Pradesh 3
Karnataka 34
Kerala 21
Madhya Pradesh 15
Maharashtra 18
Nagaland 2
Odisha 10
Punjab 7
Rajasthan 9
Tamil Nadu 9
Uttar Pradesh 33
West Bengal 8
Delhi 16
Jharkhand 21
Chhattisgarh 4
Uttarakhand 8
Telangana 10
Overall 14

Table 3:
Q: In the future, if you have a problem that requires police help, would you go to the police?
2.Yes 1. No 3. Probably 4. Have no other option 8. Don’t Know
States Respondents who said they will be willing to contact the police in future
Andhra Pradesh 59
Assam 82
Bihar 70
Gujarat 75
Haryana 74
Himachal Pradesh 73
Karnataka 69
Kerala 91
Madhya Pradesh 83
Maharashtra 82
Nagaland 63
Odisha 92
Punjab 69
Rajasthan 73
Tamil Nadu 58
Uttar Pradesh 53
West Bengal 80
Delhi 87
Jharkhand 53
Chhattisgarh 83
Uttarakhand 89
Telangana 48
Overall 72

Table 4:
Q: Do you think the police intentionally implicates people under false charges?
2. Yes 1. No 3. Maybe 8. Can’t say
States Respondents who believe that police implicates people under false charges
Andhra Pradesh 18
Assam 14
Bihar 20
Gujarat 15
Haryana 9
Himachal Pradesh 4
Karnataka 29
Kerala 10
Madhya Pradesh 18
Maharashtra 13
Nagaland 5
Odisha 10
Punjab 30
Rajasthan 8
Tamil Nadu 24
Uttar Pradesh 25
West Bengal 12
Delhi 32
Jharkhand 9
Chhattisgarh 10
Uttarakhand 11
Telangana 18
Overall 27

Table 5:
Q: Looking at the present number of Scheduled Castes such as Dalits in the police force, are there adequate
numbers of them in the police?
1. Adequate 2. Less than adequate 3. More than adequate 8. Don’t Know
States Respondents who said there are inadequate number of Scheduled Castes in the police force
Andhra Pradesh 35
Assam 17
Bihar 60
Gujarat 48
Haryana 34
Himachal Pradesh 92
Karnataka 54
Kerala 31
Madhya Pradesh 73
Maharashtra 27
Nagaland 56
Odisha 48
Punjab 41
Rajasthan 67
Tamil Nadu 56
Uttar Pradesh 43
West Bengal 44
Delhi 42
Jharkhand 71
Chhattisgarh 64
Uttarakhand 47
Telangana 65
Overall 51

Table 6:
Q: Looking at the present number of OBCs in the police force, are there adequate number of them in the
police?
States Respondents who said there are inadequate number of OBCs in the police force
Andhra Pradesh 33
Assam 14
Bihar 33
Gujarat 43
Haryana 30
Himachal Pradesh 88
Karnataka 57
Kerala 26
Madhya Pradesh 40
Maharashtra 34
Nagaland 52
Odisha 63
Punjab 50
Rajasthan 51
Tamil Nadu 48
Uttar Pradesh 28
West Bengal 36
Delhi 29
Jharkhand 57
30: Chhattisgarh 53
31: Uttarakhand 32
32: Telangana 23
Overall 41

Table 7:
Q: Looking at the present number of Scheduled Tribes such as Adivasis in the police force, are there adequate
numbers of them in the police?
States Respondents who said there are inadequate number of Scheduled Tribes in the police force
Andhra Pradesh 45
Assam 14
Bihar 78
Gujarat 47
Haryana 76
Himachal Pradesh 79
Karnataka 56
Kerala 39
Madhya Pradesh 70
Maharashtra 53
Nagaland 39
Odisha 54
Punjab 37
Rajasthan 63
Tamil Nadu 63
Uttar Pradesh 46
West Bengal 59
Delhi 68
Jharkhand 35
Chhattisgarh 43
Uttarakhand 62
Telangana 49
Overall 55

Table 8:
Q: Looking at the present number of Muslims in the police force, are there adequate numbers of them in the
police?
States Respondents who said there are inadequate number of Muslims in the police force
Andhra Pradesh 39
Assam 32
Bihar 50
Gujarat 60
Haryana 58
Himachal Pradesh 93
Karnataka 64
Kerala 29
Madhya Pradesh 63
Maharashtra 69
Nagaland 61
Odisha 74
Punjab 62
Rajasthan 67
Tamil Nadu 50
Uttar Pradesh 50
West Bengal 50
Delhi 59
Jharkhand 85
Chhattisgarh 82
Uttarakhand 72
Telangana 23
Overall 56

Table 9:
Q: Looking at the present number of women in the police force, are there adequate numbers of them in the
police?
1 Adequate 2. Less than adequate 3. More than adequate 8. Don’t Know
States Respondents who said there are inadequate number of women in the police force
Andhra Pradesh 44
Assam 64
Bihar 50
Gujarat 71
Haryana 74
Himachal Pradesh 94
Karnataka 62
Kerala 33
Madhya Pradesh 71
Maharashtra 68
Nagaland 61
Odisha 64
Punjab 42
Rajasthan 78
Tamil Nadu 52
Uttar Pradesh 56
West Bengal 61
Delhi 61
Jharkhand 45
Chhattisgarh 80
Uttarakhand 76
Telangana 80
Overall 61

Table 10:
Q: Please tell me whether you agree or disagree with - Police is corrupt- it does not do its job without a bribe?
(Probe further whether ‘fully’ or ‘somewhat’ agree or disagree)
States Respondents who agree that police is corrupt
Andhra Pradesh 26
Assam 22
Bihar 54
Gujarat 23
Haryana 22
Himachal Pradesh 6
Karnataka 28
Kerala 7
Madhya Pradesh 43
Maharashtra 21
Nagaland 6
Odisha 18
Punjab 34
Rajasthan 21
Tamil Nadu 42
Uttar Pradesh 34
West Bengal 18
Delhi 40
Jharkhand 15
Chhattisgarh 20
Uttarakhand 13
Telangana 17
Overall 25

Table 11:
Q: Have you ever seen a police officer violating the law?

2.Yes 1. No 8. Don’t Know
States Respondents who witnessed a police officer violating the law
Andhra Pradesh 11
Assam 18
Bihar 34
Gujarat 13
Haryana 18
Himachal Pradesh 31
Karnataka 14
Kerala 25
Madhya Pradesh 27
Maharashtra 29
Nagaland 17
Odisha 6
Punjab 20
Rajasthan 15
Tamil Nadu 35
Uttar Pradesh 25
West Bengal 9
Delhi 38
Jharkhand 35
Chhattisgarh 13
Uttarakhand 9
Telangana 18
Overall 20

Table 12:
Q: Please tell me how much interference do the political party & politicians have in the functioning of the
police in your locality- a lot, somewhat, not much or not at all?
1. A lot 2. Somewhat 3. Not much 4. Not at all 8. Don’t Know
States Respondents who think that political parties and politicians interfere significantly
(includes both a lot and somewhat) in the functioning of police
Andhra Pradesh 71
Assam 68
Bihar 77
Gujarat 68
Haryana 91
Himachal Pradesh 73
Karnataka 77
Kerala 32
Madhya Pradesh 88
Maharashtra 74
Nagaland 58
Odisha 74
Punjab 88
Rajasthan 74
Tamil Nadu 78
Uttar Pradesh 76
West Bengal 69
Delhi 83
Jharkhand 88
Chhattisgarh 52
Uttarakhand 88
Telangana 83
Overall 74

Table 13:
Q: Being in the police requires physical strength and aggressive behavior which women lack. Please tell me
whether this argument are justified or not?
(Probe further whether very or somewhat justified or unjustified)
States Respondents who said that the above mentioned argument is justified
Andhra Pradesh 62
Assam 60
Bihar 64
Gujarat 56
Haryana 65
Himachal Pradesh 10
Karnataka 74
Kerala 25
Madhya Pradesh 67
Maharashtra 39
Nagaland 36
Odisha 34
Punjab 71
Rajasthan 50
Tamil Nadu 59
Uttar Pradesh 61
West Bengal 56
Delhi 30
Jharkhand 66
Chhattisgarh 24
Uttarakhand 29
Telangana 66
Overall 52

Table 14:
Q: In an area, whenever there is an instance of fight between people from two caste groups, do you think the
police sides with any particular caste group or remains impartial?
1. Sides with a particular religious community 2. Remains impartial 8. Don’t Know
States Respondents who stated that police sides with a particular caste group
Andhra Pradesh 15
Assam 4
Bihar 13
Gujarat 4
Haryana 8
Himachal Pradesh
Karnataka 16
Kerala 0
Madhya Pradesh 5
Maharashtra 7
Nagaland 1
Odisha 1
Punjab 5
Rajasthan 8
Tamil Nadu 14
Uttar Pradesh 13
West Bengal 4
Delhi 10
Jharkhand 1
Chhattisgarh 1
Uttarakhand 4
Telangana 16
Overall 8

Appendix 6a: Police Performance Review Using Official Data
For the calculation of all thematic Indices (crime rate Index, disposal of cases by police and courts Index,
police diversity Index, police infrastructure Index, prison data Index, disposal of cases of crimes against SCs,
STs, women and children Index), Indices have been developed for the five-year average of each variable, and
then the individual variable Indices have been averaged to arrive at the final thematic Index. For instance,
Crime rate Index = arithmetic mean of rate of total cognizable crimes Index, rate of violent crime Index,
rate of total cognizable crimes against women Index, rate of total cognizable crimes against children Index,
rate of total cognizable crimes against SCs Index and the rate of total cognizable crimes against STs Index.
The formula used for the calculation of individual variable Indices is:
State Index = (x-minimum observed in the last five years) / (maximum observed in the last five years-
minimum observed in the last five years)
where ‘x’ is the actual state figure for the variable
In the below section, the explanation for the calculation of the individual variables is given.
I. Crime rate Index

The formula used by NCRB for the calculation of the rate of total cognizable crimes against women, children,
SCs and STs changed in the year 2012 and comparable data was not available for previous years. Therefore,
only data from 2012-2016 has been used for the calculation of the respective Indices. All final indices of all
variables under this theme has been subtracted by 1 so that a higher Index reflects better performance by the
state, i.e., a higher Index indicates lower crime in that state. The variables used in the calculation of crime
rate Index are:
1. Rate of total cognizable crimes under Indian Penal Code (IPC) and the Special and Local Laws (SLL):
The number of total cognizable crimes committed per lakh of population.
Calculation of variable: [(Total incidents of IPC crimes committed in the state+ total incidents of SLL
crimes committed in the state) / Estimated mid-year population of the state] * 100
Source: Crime in India, NCRB, 2007-2016
2. Rate of violent crimes: The number of violent crimes occurring per lakh of population. Crimes included
under the category of “violent crimes” according to NCRB are murder, attempt to commit murder,
culpable homicide not amounting to murder, attempt to commit culpable homicide, dowry deaths,
kidnapping & abduction, dacoity, making preparation & assembly for committing dacoity, robbery, riots,
arson, rape and attempt to commit rape.
Calculation of variable: (Total incidents of violent crimes in the state/ estimated mid-year population of
the state) * 100
3. Rate of total cognizable crimes against women: The number of crimes against women per lakh of female
population. Crimes included under the category of “crimes against women” according to NCRB (2014)
are rape, attempt to commit rape, kidnapping and abduction of women, dowry deaths, assault on women
with intent to outrage her modesty, insult to the modest of women, cruelty by husband or his relatives,
importation of girl from foreign country, abetment of suicide of women, The Dowry Prohibition Act 1961,
the Indecent Representation of Women (Prohibition) Act 1986, the Commission of Sati Prevention Act
1987, the Protection of Women from Domestic Violence Act 2005 and the Immoral Traffic (Prevention)
Act 1956 (women related crimes only)
Calculation of variable: (Total incidents of cognizable crimes against women in the state/ estimated mid-
year female population in the state) * 100
4. Rate of total cognizable crimes against Scheduled Castes: The number of crimes against Scheduled Castes
per lakh of SC population. Crimes included under the category of “crimes against SCs” according to
NCRB (2015) are crimes under Protection of Civil Rights Act 1955, atrocities committed against persons
belonging to SCs by non-SCs, i.e., where SC/ST (Prevention of Atrocities) Act has been applied along
with various sections of IPC, crimes committed against SCs where SC/ST Prevention of Atrocities Act
has not been applied and only IPC sections have been involved, the Employment of Manual Scavengers

and Construction of Dry Latrines (Prevention) Act 1923, other SLL crimes, assault on SC woman with
intent to outrage her modesty and insult to modesty of SC women.
Data on the following states not available/ not computable- Arunachal Pradesh, Mizoram, Nagaland,
Andaman and Nicobar Islands, Lakshadweep
Calculation of variable: (Total incidents of cognizable crimes against SCs in the state / estimated mid-
year SC population in the state) * 100
5. Rate of total cognizable crimes against Scheduled Tribes: The number of crimes against STs per lakh of ST
population. Crimes included under the category of “crimes against STs” according to NCRB (2015) are
crimes under Protection of Civil Rights Act 1955, atrocities committed against persons belonging to STs
by non-STs, i.e., where SC/ST (Prevention of Atrocities) Act has been applied along with various sections
of IPC, crimes committed against STs where SC/ST Prevention of Atrocities Act has not been applied
and only IPC sections have been involved, the Employment of Manual Scavengers and Construction of
Dry Latrines (Prevention) Act 1923, other SLL crimes, assault on ST woman with intent to outrage her
modesty and insult to modesty of ST women.
Data on the following states not available/ not computable- Haryana, Punjab, Chandigarh, Delhi,
Puducherry
Calculation of variable: (Total incidents of cognizable crimes against STs in the state / estimated mid-
year ST population in the state) * 100
6. Rate of total cognizable crimes against children: The number of crimes against children per lakh of
children’s population. Crimes included under the category of “crimes against children” according to NCRB
(2014) are murder, attempt to commit murder, infanticide, rape, unnatural offence, assault on women (girl
child) with intent to outrage her modesty, insult to the modesty of women (girl child), kidnapping and
abduction, foeticide, abetment of suicide of child, exposure and abandonment, procuration of minor
girls, importation of girls from foreign countries (under 18 years of age), buying of minors for prostitution,
selling of minor for prostitution, Prohibition of Child Marriage Act 2006, Transplantation of Human
organs Act 1994 (for persons below 18 years of age), Child Labor (Prohibition and Regulation) Act 1986,
Immoral Traffic (Prevention) Act 1956, Juvenile Justice (care and Protection of Children) Act 2000 and
Protection of Children from Sexual Offences Act 2012.
Calculation of variable: (Total incidents of cognizable crimes against children in the state / estimated
mid-year children’s population in the state) * 100
II. Police diversity Index

Data on SCs, STs. OBCs and women in police taken from BPRD, while data on Muslims in police taken
from NCRB and Census 2011. Data on Muslims in police only available till 2013, therefore, the respective
Index is an average of three years (2011-2013). Data on SCs, STs and OBCs in police available for the year
2015, therefore the respective Indices are averages of four years (2012, 2013, 2014 and 2016). The variables
used in the calculation of police diversity Index are:
1. Percentage of SCs in police in proportion to the reserved percentage for SCs: Actual percentage of SCs
in state police force as a proportion of the reserved percentage of SCs in the state police force. Data not
available for the year 2015.
Data not available for the following states due to no reservation for SCs in the state - Arunachal Pradesh,
Meghalaya, Mizoram, Nagaland, A&N Islands, Lakshadweep.
Calculation of variable: (Percentage of actual SCs in police force/Percentage of reservation of SCs in
police force) * 100
Source: Data on Police Organisations, BPRD, 2008-2017
2. Percentage of STs in police in proportion to the reserved percentage for STs: Actual percentage of STs
in state police force as a proportion of the reserved percentage of STs in the state police force. Data not
available for the year 2015.

Data not available for the following states due to no reservation for STs in the state/data not available-
Haryana, Mizoram, Chandigarh, Puducherry.
Calculation of variable: (Percentage of actual STs in police force/Percentage of reservation of STs in
police force) * 100
3. Percentage of OBCs in police in proportion to the reserved percentage for OBCs: Actual percentage of
OBCs in state police force as a proportion of the reserved percentage of OBCs in the state police force.
Data not available for the year 2015.
Data not available for the following states due to no reservation for OBCs in the state/data not available
- Arunachal Pradesh, Meghalaya, Mizoram, Nagaland, Tripura, Lakshadweep.
Calculation of variable: (Percentage of actual OBCs in police force/Percentage of reservation of OBCs
in police force) * 100
4. Percentage of Muslims in police in proportion to the Muslim population in the state: The percentage of
Muslims in state police as a proportion of percentage of Muslim population in the state. Data taken from
NCRB, and discontinued in 2013, therefore Index only calculated until 2013. Data on this variable not
given in BPRD report.
Data of following state not available- Telangana.
Calculation of variable: (Percentage of Muslims in state police/Percentage of Muslim population in
state) * 100
Source: Crime in India, NCRB, 2006-2013 and Census 2011
5. Percentage of women in police: The percentage of women in state police in proportion to the actual
number of total police personnel in the state.
Calculation of variable: [Actual number of women in state police/actual number of total police personnel
(civil + armed) in the state] * 100
III. Disposal of cases by police and courts Index

The variables used for the calculation of disposal of cases by police and courts Index are:
1. Chargesheeting rate of IPC and SLL cases by the police: The number of IPC and SLL cases in which
chargesheets were filed by the police as a percentage of the total number of cases disposed by the police.
Chargesheeting rate calculation method used by NCRB in each particular year was replicated in this file
for obtaining the rate for IPC+SLL chargesheeting.
Calculation of variable: [(Total number of IPC cases in which chargesheets were filed + Total
number of SLL cases in which chargesheets were filed) / (Total number of cases disposed
off by police under IPC + total number of cases disposed off by police under SLL)] * 100
2. Disposal percentage of IPC and SLL cases by the police: The number of IPC and SLL cases disposed off
by police as a percentage of the total number of cases for investigation by the police.
Calculation of variable: 100- {[(IPC cases pending investigation by police at the end of the year +
SLL cases pending investigation by police at the end of the year) / (total IPC cases for investigation
by the police in the year + total SLL cases for investigation by the police in the year)] * 100}
3. Conviction rate of IPC and SLL cases by the courts: The number of IPC and SLL cases convicted by
court as a percentage of the total number of IPC+SLL cases tried by the court in the year.
Calculation of variable: [(Total IPC cases convicted + total SLL cases convicted) / (total IPC cases in
which trials were completed + total SLL cases in which trials were completed)] * 100

4. Disposal percentage of IPC and SLL cases by the courts: The number of IPC and SLL cases disposed off
by courts as a percentage of the total number of case for trial by the courts.
Calculation of variable: 100- {[(IPC cases pending trial at the end of the year + SLL cases pending trial at the
end of the year) / (total IPC cases for trial during the year + total SLL cases for trial during the year)] * 100}
IV. Police infrastructure Index

The variables used in the calculation of police infrastructure Index are:
1. Percentage utilisation of grants for police modernisation: The grants for police modernisation utilised as
a percentage of the sum of central and state grants for police modernisation
Data of following states not available/ no grants for police modernisation- Assam, Chhattisgarh,
Haryana, Karnataka, Madhya Pradesh, Maharashtra, Manipur, Nagaland, Uttarakhand, A&N Islands,
Meghalaya, Chandigarh, D&N Haveli, Daman &Diu, Delhi, Lakshadweep, Sikkim, Andhra Pradesh,
Himachal Pradesh.
Calculation of variable: [Total grants for police modernisation utilised by the state/ (central + state grants
for police modernisation)] * 100
2. Strength of actual police per lakh of population as a percentage of sanctioned police per lakh of population:
Data of following state not available- Telangana
Calculation of variable: (Actual strength of police per lakh of population in the state / sanctioned strength
of police per lakh of population in the state) * 100
3. Strength of actual police stations as a percentage of sanctioned police stations: The number of actual
police stations in a state as a percentage of the sanctioned number of police stations in the state. Data on
sanctioned number of police stations not available for the year 2007, therefore maximum and minimum
values for Index calculation taken from the years 2008-2016.
Calculation of variable: (Actual strength of police stations in the state/sanctioned strength of police
stations in the state) * 100
4. Police expenditure as a percentage of state budget: The total expenditure on police as a percentage of the
total state budget.
Data of following states not available- Arunachal Pradesh, Bihar, Chhattisgarh, Goa, Jharkhand,
Meghalaya, Nagaland, Punjab, Telangana, UP, WB, D&N Haveli, Daman & Diu, Delhi.
Calculation of variable: (Total police expenditure in state / total budget for state) * 100
5. Training expenditure as a percentage of total police expenditure: The expenditure on police training as a
percentage of the total police expenditure in the state.
Data of following states not available- Arunachal Pradesh, Manipur, Telangana, D&N Haveli,
Lakshadweep, Puducherry.
Calculation of variable: (Total expenditure on police training / total police expenditure in state) * 100
6. Police personnel given in-service training as a percentage of total police strength: The number of police
personnel given in-service training as a percentage of the actual strength of police personnel in the state.
Calculation of of variable: [(Constables + ASI/SIs + DySPs + IPS given in-service training in the state)
/ total number of police personnel in the state (civil + armed) (actual)] * 100

V. Prison data Index
For calculation of the prison data Index, the average of the variables has been subtracted by 1 so that a higher
Index indicates better state performance i.e., the greater the Index, the better the state performs in terms of its
prison data. Percentage of OBC prisoners in proportion to OBC population in the state left out as a variable
because data on OBC population in states not available in Census 2011. Prison data only available until 2015,
therefore for the calculation of the Indices, five-year average of the variables have been taken for the years
2011-2015. The variables used for the calculation of prison data Index are:
1. Percentage non-utilisation of prison budget: The percentage of funds allocated to prisons not utilised by
the state.
Calculation of variable: 100 - [(total annual expenditure on prisons / total annual budget for the prison) *
100]
Source: Prison Statistics India, NCRB, 2006-2015
2. Percentage of SC prisoners in proportion to SC population in the state: States for which data not available-
Arunachal Pradesh, Nagaland, Telangana, A&N Islands, Lakshadweep.
Calculation of percentage of SCs in prisons: [SC prison inmates (sum of convicts, undertrials, detenues
and others)/total prison inmates] * 100
Calculation of percentage of SCs in prisons in proportion to SC population in the state: [(Percentage of
SC inmates in prison / percentage of SC population in state) * 100]
Source: Prison Statistics India, NCRB, 2006-2015 and Census 2011
3. Percentage of ST prisoners in proportion to ST population in the state: States for which data not available-
Haryana, Punjab, Telangana, Chandigarh, Delhi, Puducherry.
Calculation of percentage of STs in prisons: [ST prison inmates (sum of convicts, undertrials, detenues
and others)/total prison inmates] * 100
Calculation of percentage of STs in prisons in proportion to ST population in the state: [(Percentage of
ST inmates in prison / percentage of ST population in state) * 100]
4. Percentage of Muslim prisoners in proportion to Muslim population in the state: State for which data not
available- Telangana.
Calculation of percentage of Muslims in prisons: [Muslim prison inmates (sum of convicts, undertrials,
detenues and others) / total prison inmates] * 100
Calculation of percentage of Muslims in prisons in proportion to Muslim population in the state:
[(Percentage of Muslim inmates in prison / percentage of Muslim population in state) * 100]
5. Number of undertrial prisoners as a percentage of the total strength of prisoners: Undertrial prisoners as
a percentage of the total prison population.
Calculation of variable: (Number of Undertrial prison inmates in the state / total number of prison
inmates in the state) * 100
Source: Prison Statistics India, NCRB, 2006-2015
VI. Disposal of cases of crimes against SCs, STs, women and children Index
The disposal Indices for crimes against SCs, STs, women and children were created separately. Data on all
variables of disposal of crimes against SCs and STs not available for the years 2014 and 2015. Therefore,
Indices for these variables created using three year averages (2012, 2013 and 2016). Data on disposal
percentages of cases of crimes against children by police and courts not available for the years 2014 and
2015. Therefore, Indices for these variables calculated using three-year average for the years 2012, 2013 and
2016. Data on disposal percentages of cases of crimes against women by police and by courts not available
for the years 2010-2015. Therefore, Indices for these variables calculated using only the data for 2016, with

maximum and minimum values taken from the years 2006-2010. The variables used for the calculation of
disposal of cases of crimes against SCs, STs, women and children Indices are:
1. Chargesheeting rate of cases of crimes against SCs, STs, women and children: Number of cases
chargesheeted by the police as a percentage of the total number of cases disposed off by the police (of
crimes against SCs, STs, women and children respectively). Chargesheeting rate of cases of crimes
against SCs and STs not available for the years 2014 and 2015. Therefore, three-year averages used for the
calculation of the respective Indices (2012, 2013 and 2016).
States in which data not available/ not computable in crimes against SCs: Arunachal Pradesh,
Meghalaya, Mizoram, Nagaland, Sikkim, Tripura, A&N Islands, Chandigarh, D&N Haveli, Daman &
Diu, Lakshadweep
States in which data not available/ not computable in crimes against STs: Haryana, J&K, Meghalaya,
Mizoram, Nagaland, Punjab, Sikkim, Chandigarh, Daman & Diu, Delhi, Lakshadweep, Puducherry.
State in which data not available/ not computable in cases of crimes against women: Lakshadweep
State in which data not available/ not computable in cases of crimes against children: Lakshadweep
Calculation of variables: [Number of cases of crimes against SCs, STs, women, children respectively
in which chargesheets were filed / (total number of cases of crimes against SCs, STs, women, children
respectively disposed off by the police)] * 100
Source: Crime in India, 2007-2016, NCRB
2. Disposal percentage by police of cases of crimes against SCs, STs, women and children: Number of
cases of crimes against SCs, STs, women and children disposed off by the police as a percentage of the
total number of cases of crimes against SCs, STs, women and children respectively for investigation by
the police. Data on disposal percentage of cases of crimes against SCs, STs and children by police not
available for the years 2014 and 2015. Therefore, three-year averages taken for the calculation of the
respective Indices of the years 2012, 2013 and 2016. Data on disposal percentage of cases of crimes
against women by police not available for the years 2010-2015. Therefore, Indices for these variables
calculated using only the data for 2016, with maximum and minimum values taken from the years 2006-
2010.
States in which data not available/ not computable in crimes against SCs: Arunachal Pradesh, Meghalaya,
Mizoram, Nagaland, Tripura, A&N Islands, D&N Haveli, Lakshadweep
Mizoram, Nagaland, Punjab, Sikkim, Chandigarh, Daman & Diu.
Calculation of variables: 100 - [(Number of cases of crimes against SCs, STs, women and children
pending investigation at the end of the year / total cases of crimes against SCs, STs, women and children
respectively for investigation by police in the year) * 100]

3. Conviction rate of cases of crimes against SCs, STs, women and children: Number of cases of crimes
against SCs, STs, women and children convicted by the courts as a percentage of the total number of cases
of crimes against SCs, STs, women and children respectively tried by the court in the year. Conviction rate
of cases of crimes against SCs and STs not available for the years 2014 and 2015. Therefore, three-year
averages used for the calculation of the respective Indices (2012, 2013 and 2016).
States in which data not available/ not computable in crimes against SCs: Arunachal Pradesh, Assam,
J&K, Manipur, Meghalaya, Mizoram, Nagaland, Tripura, A&N Islands, Chandigarh, D&N Haveli,
Daman & Diu, Lakshadweep, Puducherry.
States in which data not available/ not computable in crimes against STs: Assam, Haryana, HP, J&K,
Manipur, Meghalaya, Mizoram, Nagaland, Punjab, Sikkim, Chandigarh, D&N Haveli, Daman & Diu,
Lakshadweep, Puducherry.
State in which data not available/ not computable in cases of crimes against women: Lakshadweep
States in which data not available/ not computable in cases of crimes against children: Arunachal
Pradesh, Manipur, D&N Haveli, Daman & Diu, Lakshadweep, Puducherry.
Calculation of variables: [(Number of cases of crimes against SCs, STs, women and children convicted
by court during the year / total cases of crimes against SCs, STs, women and children respectively tried
by the court in the year) * 100]
4. Disposal percentage by court of cases of crimes against SCs, STs, women and children: Number of cases
of crimes against SCs, STs, women and children disposed off by the courts as a percentage of the total
number of cases of crimes against SCs, STs, women and children respectively for trial by the courts in the
year. Data on disposal percentage of cases of crimes against SCs, STs and children by courts not available
for the years 2014 and 2015. Therefore, three-year averages taken for the calculation of the respective
Indices of the years 2012, 2013 and 2016. Data on disposal percentage of cases of crimes against women
by courts not available for the years 2010-2015. Therefore, Indices for these variables calculated using only
the data for 2016, with maximum and minimum values taken from the years 2006-2010.
States in which data not available/ not computable in crimes against SCs: Arunachal Pradesh, Meghalaya,
Mizoram, Nagaland, Tripura, A&N Islands, Daman & Diu, Lakshadweep
Mizoram, Nagaland, Punjab, Sikkim, Chandigarh, Daman & Diu, Lakshadweep, Puducherry.
Calculation of variables: 100 - [(Number of cases of crimes against SCs, STs, women and children pending
trial at the end of the year / total cases of crimes against SCs, STs, women and children respectively for
trial during the year) *100]

Appendix 6b: Police Performance Review Using Objective Data
Crime Index
States Rate of total Rate of Rate of Rate of Rate of Rate of
cognizable violent crimes crimes crimes crimes
crimes crimes against against against SCs against STs
(IPC+SLL) women children
Andhra Pradesh 0.89 0.87 0.66 0.93 0.48 0.58
Arunachal Pradesh 0.92 0.49 0.74 0.86 NA 0.96
Assam 0.88 0.29 0.35 0.91 1 1
Bihar 0.94 0.56 0.87 0.97 0.43 0.89
Chhattisgarh 0.49 0.69 0.77 0.77 0.74 0.8
Goa 0.88 0.73 0.8 0.73 0.28 0.89
Gujarat 0.69 0.87 0.84 0.92 0.58 0.94
Haryana 0.8 0.53 0.63 0.85 0.82 1
Himachal Pradesh 0.91 0.79 0.82 0.88 0.89 0.98
Jammu & Kashmir 0.92 0.58 0.71 0.98 1 1
Jharkhand 0.95 0.69 0.82 0.98 0.71 0.91
Karnataka 0.89 0.65 0.79 0.92 0.68 0.76
Kerala 0.15 0.52 0.7 0.86 0.61 0.28
Madhya Pradesh 0.8 0.67 0.66 0.78 0.5 0.75
Maharashtra 0.86 0.65 0.77 0.85 0.82 0.91
Manipur 0.94 0.57 0.9 0.92 0.99 1
Meghalaya 0.96 0.66 0.89 0.88 1 1
Mizoram 0.91 0.79 0.82 0.77 1 1
Nagaland 0.99 0.91 0.99 0.97 NA 1
Odisha 0.92 0.57 0.62 0.92 0.53 0.77
Punjab 0.92 0.82 0.82 0.9 0.98 1
Rajasthan 0.85 0.78 0.58 0.93 0.21 0.43
Sikkim 0.95 0.71 0.85 0.8 0.65 0.92
Tamil Nadu 0.61 0.81 0.92 0.94 0.83 0.93
Telangana 0.86 0.8 0.57 0.87 0.55 0.61
Tripura 0.96 0.68 0.6 0.9 0.91 0.97
Uttar Pradesh 0.55 0.78 0.83 0.92 0.71 0.96
Uttarakhand 0.43 0.79 0.87 0.93 0.95 0.98
West Bengal 0.92 0.6 0.62 0.91 0.99 0.95
A & N Islands 0.77 0.83 0.81 0.73 NA 0.64
Chandigarh 0.88 0.67 0.7 0.7 0.99 1
D&N Haveli 0.99 0.9 0.96 0.92 1 0.95
Daman & Diu 0.99 0.83 0.9 0.89 0.85 0.97
Delhi UT 0.67 0.01 0.21 0.18 0.97 1
Lakshadweep 0.98 0.9 0.95 0.95 NA 0.99
Puducherry 0.88 0.91 0.96 0.94 0.9 1
AP+ Telangana 0.88 0.85 0.63 0.92 0.52 0.63
TOTAL (ALL 0.77 0.68 0.74 0.9 0.69 0.81
INDIA)
Note: Above Indices have been developed using data from ‘Crime in India’ Report (National Crime Records Bureau) 2007-2016

Police diversity Index
States SCs in Police STs in Police OBCs in Police Muslims in Police Women in Police
Andhra Pradesh 0.07 0.09 0.55 0.69 0.25
Arunachal Pradesh NA 0.08 NA 0.37 0.48
Assam 0.08 0.11 0.52 0.08 0.15
Bihar 0.06 0.14 0.31 0.18 0.38
Chhattisgarh 0.05 0.07 0.3 0.32 0.33
Goa 0.05 0.03 0.18 0.15 0.53
Gujarat 0.09 0.06 0.23 0.31 0.31
Haryana 0.05 0 0.23 0.11 0.55
Himachal Pradesh 0.09 0.15 0.26 0.41 0.91
Jammu & Kashmir 0.04 0.07 0.22 0.56 0.2
Jharkhand 0.08 0.1 0.57 0.21 0.36
Karnataka 0.07 0.13 0.6 0.31 0.4
Kerala 0.08 0.07 0.39 0.21 0.46
Madhya Pradesh 0.06 0.06 0.28 0.36 0.35
Maharashtra 0.08 0.12 0.59 0.17 0.87
Manipur 0.08 0.07 0.11 0.87 0.6
Meghalaya NA 0.1 0.47 0.25 0.25
Mizoram NA NA NA 0.21 0.48
Nagaland NA 0.12 NA 0.3 0.21
Odisha 0.08 0.11 0.53 0.56 0.68
Punjab 0.09 0 0.61 0.13 0.49
Rajasthan 0.07 0.1 0.25 0.09 0.61
Sikkim 0.06 0.14 0.7 0.05 0.58
Tamil Nadu 0.06 0.06 0.29 0.27 1.12
Telangana 0.07 0.17 0.64 NA 0.18
Tripura 0.07 0.1 NA 0.37 0.29
Uttar Pradesh 0.03 0.05 0.22 0.18 0.29
Uttarakhand 0.09 0.18 0.47 0.1 0.63
West Bengal 0.05 0.08 0.1 0.22 0.49
A & N Islands NA 0.09 0.1 0.72 0.94
Chandigarh 0.06 NA 0.34 0.06 1.3
D&N Haveli 0.04 0.84 0.31 0.27 0.93
Daman & Diu 0.53 0.08 0.36 0.24 0.71
Delhi UT 0.08 0.1 0.33 0.11 0.63
Lakshadweep NA 0.08 NA 0 0.61
Puducherry 0.06 NA 0.35 0.2 0.48
AP+ Telangana 0.06 0.11 0.54 0.69 0.22
TOTAL (ALL NA NA NA 0.31 0.5
INDIA)
Note: SCs in police Index, STs in police Index, OBCs in police Index and women in police Index have been developed using data from
‘Data on Police Organisations’ Report (Bureau of Police Research and Development) 2008-2017. Muslims in Police Index calculated
using data from ‘Crime in India’ report 2006-2013 and Census 2011.

Disposal Index
States Chargesheeting Conviction rate Disposal percentage of Disposal percentage of

rate (IPC+SLL) (IPC+SLL) cases by courts (IPC+SLL) cases by police (IPC+SLL)
Andhra Pradesh 0.91 0.41 0.45 0.67
Arunachal Pradesh 0.63 0.42 0.03 0.49
Assam 0.5 0.11 0.19 0.37
Bihar 0.81 0.14 0.09 0.68
Chhattisgarh 0.96 0.84 0.64 0.97
Goa 0.74 0.39 0.24 0.55
Gujarat 0.97 0.54 0.12 0.89
Haryana 0.65 0.49 0.35 0.81
Himachal Pradesh 0.87 0.32 0.13 0.81
Jammu & Kashmir 0.85 0.38 0.22 0.65
Jharkhand 0.7 0.24 0.3 0.56
Karnataka 0.85 0.42 0.32 0.66
Kerala 1 0.9 0.33 0.92
Madhya Pradesh 0.93 0.67 0.37 0.95
Maharashtra 0.82 0.25 0.1 0.61
Manipur 0.18 0.53 0.06 0.04
Meghalaya 0.54 0.55 0.06 0.21
Mizoram 0.93 0.9 0.65 0.74
Nagaland 0.73 0.79 0.48 0.68
Odisha 0.9 0.11 0.11 0.67
Punjab 0.86 0.52 0.31 0.51
Rajasthan 0.78 0.72 0.24 0.94
Sikkim 0.75 0.45 0.53 0.72
Tamil Nadu 0.84 0.81 0.65 0.78
Telangana 0.84 0.37 0.33 0.7
Tripura 0.82 0.22 0.29 0.8
Uttar Pradesh 0.92 0.93 0.5 0.96
Uttarakhand 0.93 0.92 0.81 0.93
West Bengal 0.84 0.15 0.05 0.66
A & N Islands 0.96 0.85 0.12 0.79
Chandigarh 0.74 0.65 0.44 0.65
D&N Haveli 0.76 0.1 0.08 0.52
Daman & Diu 0.7 0.06 0.43 0.5
Delhi UT 0.39 0.52 0.15 0.49
Lakshadweep 0.85 0.6 0.45 0.42
Puducherry 0.9 0.83 0.28 0.76
AP+ Telangana 0.89 0.41 0.41 0.69
TOTAL (ALL 0.89 0.75 0.29 0.81
INDIA)
Note: Above Indices have been developed using data from ‘Crime in India’ Report (National Crime Records Bureau) 2007-2016

Police Infrastructure Index
States Percentage Utilization Strength of Police Police training Police personnel

of grants for police police per lakh station expenditure expenditure given training in
modernization of population Index Index Index service
Andhra Pradesh 0.04 0.53 0.93 0.18 0.04 0.16
Arunachal Pradesh 0.01 0.61 0.77 0.4 0.46 0.06
Assam 0 0.58 0.93 0.25 0.04 0.14
Bihar 0.03 0.4 0.93 0.27 0.01 0.12
Chhattisgarh 0.06 0.52 0.87 0.25 0.06 0.12
Goa 0.01 0.54 0.93 NA 0.01 0.14
Gujarat 0.05 0.36 0.88 0.15 0.04 0.02
Haryana 0.1 0.36 0.93 0.19 0.03 0.47
Himachal Pradesh 0.11 0.6 0.93 0.18 0.03 0.31
Jammu & Kashmir 0.09 0.65 0.93 0.46 0.03 0.14
Jharkhand 0.08 0.45 0.88 0.33 0.03 0.29
Karnataka 0.03 0.41 0.93 0.16 0.04 0.04
Kerala 0.04 0.71 0.92 0.18 0.02 0.08
Madhya Pradesh 0.01 0.58 0.93 0.18 0.06 0.05
Maharashtra 0.02 0.69 0.93 0.25 0.01 0.06
Manipur 0.01 0.49 0.67 0.63 0 0.22
Meghalaya 0 0.57 0.94 0.38 0.02 0.17
Mizoram 0.03 0.56 0.93 0.44 0.07 0.32
Nagaland 0.11 0.8 0.93 0.57 0.05 0.04
Odisha 0.05 0.57 0.91 0.19 0.03 0.07
Punjab 0.02 0.64 0.93 0.44 0.05 0.18
Rajasthan 0.07 0.6 0.93 0.7 0.04 0.09
Sikkim 0.07 0.48 0.93 0.4 0.03 0.13
Tamil Nadu 0.05 0.58 0.92 0.24 0.06 0.45
Telangana 0.02 0.48 0.93 0.07 0.07 0.22
Tripura 0.05 0.61 0.92 0.45 0.07 0.12
Uttar Pradesh 0.03 0.1 0.93 0.27 0.03 0.09
Uttarakhand 0.07 0.66 0.93 0.19 0.02 0.12
West Bengal 0.05 0.35 0.92 0.23 0.01 0.05
A & N Islands 0.04 0.61 0.93 0.44 0.03 0.04
Chandigarh 0.02 0.64 0.93 0.76 0.01 0.89
D&N Haveli 0.02 0.63 1.02 0.17 0.02 0.01
Daman & Diu 0.04 0.48 0.7 0.08 0.04 0.09
Delhi UT 0.04 0.7 0.93 0.86 0.07 0.26
Lakshadweep 0.05 0.52 0.79 0.14 0.01 0
Puducherry 0.08 0.51 0.93 0.19 0.12 0.07
AP+ Telangana 0.04 0.5 0.93 0.14 0.04 0.18
TOTAL (ALL 0.05 0.47 0.92 0.22 0.04 0.15
INDIA)
Note: Above Indices have been developed using data from ‘Data on Police Organisations’ Report (Bureau of Police Research and
Development) 2008-2017.

Prison index
States Percentage SC prisoners in STprisoners in Muslim prisoners Percentage of

non-utlization proportion to SC proportion to ST in proportion to Undertrials
of prison population in the population in the Muslim population Index
budget state state in the state
(100-percent
utilization)
Andhra Pradesh 0.68 0.99 0.85 0.93 0.35
Arunachal Pradesh 0.77 NA 0.94 0.54 0.16
Assam 0.53 0.97 0.87 0.94 0.38
Bihar 0.65 0.99 0.64 0.95 0.16
Chhattisgarh 0.7 0.98 0.91 0.85 0.42
Goa 0.62 0.95 0.94 0.85 0.3
Gujarat 0.68 0.97 0.9 0.85 0.39
Haryana 0.75 0.99 NA 0.91 0.41
Himachal Pradesh 0.8 0.99 0.94 0.87 0.49
Jammu & Kashmir 0.71 0.98 0.94 0.96 0.19
Jharkhand 0.63 0.99 0.91 0.92 0.24
Karnataka 0.7 0.99 0.89 0.93 0.32
Kerala 0.73 0.98 0.85 0.94 0.38
Madhya Pradesh 0.76 0.98 0.92 0.9 0.48
Maharashtra 0.72 0.98 0.87 0.83 0.3
Manipur 0.77 1 0.94 0.88 0.23
Meghalaya 0.55 0.71 0.93 0.82 0.13
Mizoram 0.78 0.63 0.92 0.8 0.45
Nagaland 0.79 NA 0.93 0.56 0.3
Odisha 0.63 0.99 0.92 0.84 0.25
Punjab 0.7 0.99 NA 0.85 0.39
Rajasthan 0.72 0.99 0.88 0.88 0.31
Sikkim 0.67 0.97 0.93 0.86 0.41
Tamil Nadu 0.74 0.98 0.56 0.83 0.43
Telangana 0.69 NA NA 1 0.4
Tripura 0.58 0.99 0.93 0.9 0.58
Uttar Pradesh 0.74 0.99 0.44 0.93 0.31
Uttarakhand 0.72 0.99 0.59 0.85 0.48
West Bengal 0.72 0.99 0.88 0.9 0.3
A & N Islands 0.67 NA 1 0.96 0.72
Chandigarh 0.77 0.99 NA 0.86 0.43
D&N Haveli 0.63 0.93 0.95 0.61 0
Daman & Diu 0.77 0.93 0.77 0.92 0.36
Delhi UT 0.76 0.98 NA 0.9 0.26
Lakshadweep 0.18 NA 0.93 0.95 0.53
Puducherry 0.74 0.98 NA 0.95 0.4
AP+ Telangana 0.68 0.98 0.85 0.91 0.36
TOTAL (ALL 0.7 0.99 0.88 0.92 0.33
INDIA)
Note: Above Indices have been developed using data from ‘Prison Statistics India’ Report (National Crime Records Bureau) 2006-
2015 and Census 2011.

Disposal of cases of crimes against SCs Index
States Chargesheeting rate Disposal percentage Conviction rate Disposal percentage

of cases of crimes by police of cases of of cases of crimes by courts of cases of
against SCs crimes against SCs against SCs crimes against SCs
Andhra Pradesh 0.78 0.52 0.07 0.34

Arunachal Pradesh NA NA NA NA
Assam 0.68 0.47 NA 0.37
Bihar 0.91 0.63 0.15 0.09
Chhattisgarh 0.99 0.84 0.36 0.24
Goa 0.81 0.58 0.03 0.17
Gujarat 0.96 0.9 0.05 0.09
Haryana 0.89 0.91 0.12 0.36
Jammu & Kashmir 0.67 0.56 NA 0.06
Jharkhand 0.57 0.42 0.31 0.22
Karnataka 0.91 0.72 0.04 0.18
Kerala 0.73 0.6 0.08 0.1
Madhya Pradesh 0.99 0.91 0.33 0.21
Maharashtra 0.93 0.69 0.07 0.1
Manipur 0.33 0.58 NA 0.67
Meghalaya NA NA NA NA
Mizoram NA NA NA NA
Nagaland NA NA NA NA
Odisha 0.9 0.57 0.05 0.15
Punjab 0.81 0.6 0.2 0.24
Rajasthan 0.81 0.89 0.44 0.1
Sikkim NA 0.67 0.51 0.8
Tamil Nadu 0.87 0.67 0.13 0.18
Telangana 0.69 0.82 0.07 0.15
Tripura NA NA NA NA
Uttar Pradesh 0.84 0.87 0.54 0.12
Uttarakhand 0.7 0.74 0.53 0.32
West Bengal 0.92 0.51 0.02 0.1
A & N Islands NA NA NA NA
Chandigarh NA 0.5 NA 0.17
D&N Haveli NA NA NA 0.07
Daman & Diu NA 0.17 NA NA
Delhi UT 0.89 0.31 0.12 0.08
Lakshadweep NA NA NA NA
Puducherry 0.92 0.73 NA 0.22
AP+ Telangana 0.79 0.52 0.04 0.32
TOTAL (ALL 0.87 0.72 0.24 0.14
INDIA)
Note1: Above Indices have been developed using data from ‘Crime in India’ Report (National Crime Records Bureau) 2007-2016

Disposal of cases of crimes against STs Index

of cases of crimes by police of cases of of cases of crimes by courts of cases of
against STs crimes against STs against STs crimes against STs
Andhra Pradesh 0.85 0.56 0.03 0.35

Assam 0.67 0.64 NA 0.36
Bihar 0.82 0.71 0.2 0.17
Chhattisgarh 1 0.87 0.34 0.24
Goa 0.92 0.36 0.08 0.18
Gujarat 0.98 0.86 0.04 0.07
Haryana NA NA NA NA
Himachal Pradesh 0.93 0.7 NA 0.14
Jammu & Kashmir NA NA NA NA
Jharkhand 0.69 0.44 0.39 0.19
Karnataka 0.92 0.71 0.02 0.19
Kerala 0.86 0.57 0.1 0.13
Madhya Pradesh 0.99 0.92 0.32 0.22
Maharashtra 0.95 0.7 0.07 0.1
Manipur 0.83 0.35 NA 0.11
Meghalaya NA NA NA NA
Mizoram NA NA NA NA
Nagaland NA NA NA NA
Odisha 0.95 0.64 0.08 0.12
Punjab NA NA NA NA
Rajasthan 0.81 0.88 0.28 0.07
Sikkim NA NA NA NA
Tamil Nadu 0.96 0.67 0.08 0.11
Telangana 0.74 0.49 0.08 0.22
Tripura 0.97 0.65 0 0.3
Uttar Pradesh 0.97 0.91 0.5 0.1
Uttarakhand 0.72 0.72 0 0.44
West Bengal 0.92 0.53 0.03 0.05
A & N Islands 0.61 0.58 0.33 0.05
Chandigarh NA NA NA NA
D&N Haveli 0.5 0.16 NA 0.06
Daman & Diu NA NA NA NA
Delhi UT NA 0.67 0 0.71
Lakshadweep NA 0.67 NA NA
Puducherry NA 0.67 NA NA
AP+ Telangana 0.85 0.56 0.04 0.34
TOTAL (ALL 0.9 0.73 0.2 0.16
INDIA)
Note: Above Indices have been developed using data from ‘Crime in India’ Report (National Crime Records Bureau) 2007-2016.

Disposal of cases of crimes against women Index

for cases of crimes by police of cases for cases of crimes by courts of cases
against women of crimes against against women of crimes against
women women
Andhra Pradesh 0.95 0.67 0.12 0.83
Assam 0.61 0.47 0.08 0.59
Bihar 0.86 0.72 0.17 0.89
Chhattisgarh 0.94 0.87 0.33 1.09
Goa 0.79 0.48 0.15 0.6
Gujarat 0.96 0.86 0.04 1.08
Haryana 0.85 0.88 0.18 1.1
Himachal Pradesh 0.93 0.8 0.15 1
Jammu & Kashmir 0.92 0.68 0.07 0.84
Jharkhand 0.84 0.55 0.29 0.69
Karnataka 0.94 0.64 0.06 0.8
Kerala 0.97 0.72 0.12 0.9
Madhya Pradesh 0.97 0.88 0.32 1.09
Maharashtra 0.94 0.66 0.09 0.83
Manipur 0.35 0.16 0.31 0.2
Meghalaya 0.78 0.38 0.42 0.47
Mizoram 0.98 0.77 0.79 0.96
Nagaland 0.81 0.73 0.75 0.91
Odisha 0.96 0.71 0.08 0.88
Punjab 0.85 0.48 0.3 0.6
Rajasthan 0.89 0.96 0.38 1.2
Sikkim 0.9 0.68 0.46 0.85
Tamil Nadu 0.83 0.53 0.26 0.66
Telangana 0.92 0.66 0.08 0.82
Tripura 0.92 0.82 0.19 1.02
Uttar Pradesh 0.73 0.81 0.53 1.01
Uttarakhand 0.73 0.75 0.55 0.93
West Bengal 0.95 0.75 0.06 0.93
A & N Islands 0.87 0.67 0.16 0.83
Chandigarh 0.87 0.58 0.24 0.72
D&N Haveli 0.8 0.64 0.12 0.8
Daman & Diu 0.86 0.59 0.04 0.74
Delhi UT 0.75 0.4 0.33 0.5
Lakshadweep NA 0.38 NA 0.47
Puducherry 0.93 0.68 0.41 0.85
AP+ Telangana NA 0.66 NA 0.83
TOTAL (ALL 0.88 0.69 0.21 0.87
INDIA)
Note: Above Indices have been developed using data from ‘Crime in India’ Report (National Crime Records Bureau) 2006-2016.

Disposal of cases of crimes against children Index

for cases of crimes by police of cases for cases of crimes by courts of cases
against children of crimes against against children of crimes against
children children
Andhra Pradesh 0.93 0.59 0.1 0.29
Arunachal Pradesh 0.75 0.58 NA 0.02
Assam 0.66 0.46 0.14 0.15
Bihar 0.84 0.64 0.18 0.12
Chhattisgarh 0.87 0.81 0.41 0.23
Goa 0.74 0.41 0.24 0.07
Gujarat 0.81 0.85 0.11 0.06
Haryana 0.82 0.87 0.2 0.26
Jammu & Kashmir 0.81 0.65 0.05 0.25
Jharkhand 0.83 0.64 0.25 0.38
Karnataka 0.83 0.64 0.12 0.19
Kerala 0.96 0.59 0.23 0.08
Madhya Pradesh 0.9 0.83 0.33 0.22
Maharashtra 0.82 0.57 0.16 0.05
Manipur 0.36 0.06 NA 0.02
Meghalaya 0.8 0.18 0.58 0.04
Mizoram 0.98 0.72 0.9 0.23
Nagaland 0.72 0.6 0.81 0.71
Odisha 0.96 0.63 0.13 0.08
Punjab 0.83 0.47 0.37 0.31
Rajasthan 0.8 0.88 0.37 0.14
Sikkim 0.97 0.9 0.55 0.41
Tamil Nadu 0.85 0.54 0.42 0.22
Telangana 0.88 0.62 0.13 0.15
Tripura 0.92 0.69 0.29 0.1
Uttar Pradesh 0.79 0.82 0.55 0.11
Uttarakhand 0.61 0.81 0.63 0.21
West Bengal 0.88 0.4 0.09 0.06
A & N Islands 0.88 0.65 0.24 0.06
Chandigarh 0.78 0.58 0.42 0.37
D&N Haveli 0.68 0.64 NA 0.13
Daman & Diu 0.56 0.62 NA 0.18
Delhi UT 0.44 0.47 0.38 0.14
Lakshadweep NA 0.87 NA 0.67
Puducherry 0.93 0.74 NA 0.07
AP+ Telangana NA 0.62 NA 0.26
TOTAL (ALL 0.81 0.64 0.32 0.14
INDIA)
Note: Above Indices have been developed using data from ‘Crime in India Report’ (National Crime Records Bureau) 2007-2016.

Appendix 7: States’ Compliance with Supreme Court Directives
1. State Security Comission
Term used Lokayukta Judge- Leader of Non- Recommendations
sitting or Opposition political binding
retired members
Assam No Yes No Yes No
Bihar State Police Board No No No No No
Chhattisgarh State Police No No No Yes No
Commission
Gujarat State Security No No No Yes No
Commission
Haryana State Police Board No Yes Yes Yes No
Himachal Pradesh State Police Board Yes Yes Yes Yes Normally binding
Karnataka State Security No Yes Yes No Yes
Commission
Kerala SSC No Yes Yes Yes Yes, but governmentt
can reject/modify with
written reasons specified
Maharashtra SSC No No Yes Yes No
Punjab State Police Board No No-AG No No No
Rajasthan State Police No No Yes Yes No
Commission
Tamil Nadu SSC No No Yes No No
Uttarakhand State Police Board No No Yes Yes No- Board can only
suggest and advise the
government
Note: Above data has been taken from State Police Acts passed post 2006.
2. DGP appointment
2 yr Grounds for removal before end of tenure: any other than disciplinary action,
tenure conviction, otherwise incapacitated
Assam No promotion, public interest, inefficiency, negligence
Bihar Yes promotion; any other administrative reason
Chhattisgarh Yes Superannuation; suspension; on his own request; administrative exigency
Gujarat Yes Only by a written order with reasons mentioned: suspension, physical/mental illness
Haryana No Promotion- on written order specifying reasons

Himachal Pradesh No superannuation
Karnataka Yes Superannuation; appointed to another post with his consent
Kerala Yes Superannuation, dissatisfaction of public, requests in writing, not carrying out duties
sufficiently
Maharashtra Yes Superannuation; gross dereliction of duty
Punjab Yes Superannuation, promotion, transfer by state government for special reasons- to be
recorded in writing
Rajasthan Yes on his own request; administrative exigency- to be recorded in writing
Tamil Nadu Yes Appointment to any other post, with his consent; administrative grounds (to be
recorded in writing)
Uttarakhand Yes Superannuation; promotion; on how own request; gross negligence (to be recorded in
writing)

3. Officers on operational Duties
Min tenure-2 yrs Grounds for removal before end of tenure: any other than
disciplinary action, conviction, otherwise incapacitated, promotion
and retirement
Assam No In public interest
Bihar Yes Transfer; other administrative reasons
Chhattisgarh Yes On his own request, administrative exigency
Gujarat Yes Misuse/abuse of power; on his own request

Haryana No Transfer; inefficiency
Himachal Pradesh Yes Larger public interest- to be recorded in writing

Karnataka No To fill a vacancy; on his own request (in writing); gross misconduct/
negligence
Kerala Yes Superannuation, dissatisfaction of public, requests in writing, not
carrying out duties sufficiently
Maharashtra Yes Administrative exigencies, public interest
Punjab No Filling up a vacancy; Transfer; inefficiency or negligence or non-
performance
Rajasthan Yes Vacancy; on his own request; administrative exigency- recorded in
writing
Tamil Nadu Yes Any other administrative ground (to be recorded in writing)
Uttarakhand SHO has it for only At own request, filling up a vacancy, in public interest (by written
one year. Yes for others order)
4. Separation of investigation and law and order functions of police
Term used state metros/ each Police

level district Station
Assam Yes Yes No
Bihar Yes No No
Chhattisgarh Special Crime Investigation Units: jurisdiction not NA NA NA
mentioned
Gujarat No jurisdiction mentioned; State govt may establish NA NA NA
separate wings by an order under section 7A
Haryana Specialized crime investigation unit No yes Yes
Himachal Pradesh Criminal Investigation Unit+ State Criminal Yes No Yes

Investigation Dept
Karnataka No No Yes
Kerala Jurisdiction not mentioned NA NA NA
Maharashtra No No Yes
Punjab Intelligence Wing, State Crime Investigation Wing Yes Yes No
Rajasthan Crime Investigation Units No yes Yes
Tamil Nadu Law & order wing, Investigation Wing No No Yes
Uttarakhand Special Crime Investigation Units No Yes Yes

5. Police Establishment Board
Term used Decides transfer Bar on state Can give Appeal forum
and posting interference recommendations for greivances
of officers without written regarding transfer related to
below Deputy reasons and posting of transfer and
Superintendent officers above the posting
rank rank of SP
Assam Yes State government No Yes

can reject on
ground of public
interest
Bihar Does not exist No NA No No
Chhattisgarh Yes State government No Yes
can review order
passed by PEB
Gujarat Police Yes No mention No Yes
Establishment
Board
Haryana Police no No No No
Establishment
Committee
Himachal District, Range Yes No mention Yes Yes
Pradesh and State Police
Establishment
Committees
Karnataka Police Yes Yes Yes Yes
Establishment
Board
Kerala Police No Yes No Yes
Establishment
Board
Maharashtra Police Yes No- state No Yes
Establishment government
Board No.1, 2, at directions to be
range level and at binding on board
commissionerate
level
Punjab Police Yes Not Mentioned No No
Establishment
Committees
Rajasthan Police Yes No- transfer only yes Yes
Establishment with the approval
Board of govt
Tamil Nadu PEB- state level, Yes No Yes Yes
disctrict, zonal,
range and city level
Uttarakhand Police Yes No- state govt Yes Yes
Establishment can alter or
Committee amend decisions
for reasons given
in writing

6. Police Complaints Authority
Term used District State Head chosen from Headed by Recommendations

level level a list given by retired judges binding
Chief Justice of
High Court
Assam Yes Yes No Yes No
Bihar District Yes No No Yes No
Accountability
Authority
Chhattisgarh State Police No Yes No Yes No
Accountability
Authority
Gujarat State Police Yes Yes No Yes No
Complaints
Authority
Haryana Police Yes Yes No Yes No
Complaint
Authority
Himachal State, District Yes Yes No No- Yes
Pradesh Police Lokayukta
Complaints
Authority
Karnataka Yes Yes Yes Yes No
Kerala SPCA and Yes Yes No Yes Yes
DPCA
Maharashtra Police Yes Yes Yes Yes May be rejected by
Complaint state government
Authority for reasons given in
writing in exceptional
cases
Punjab State and Yes Yes No No No
District Police
Complaint
Authority
Rajasthan State, District Yes Yes No No No
Accountability
Committee
Tamil Nadu State, District Yes Yes No No No
Police
Complaint
Authority
Uttarakhand State Police No Yes No No No
Complaints
Authority

Appendix 8a: CAG Report on Modernisation of Police
Expenditure and Funds Housing, Police Stations and Shortage of Vehicles
other Infrastructure
Assam Annual Action Plans (AAPs) Requirement of 8395 new 59% of the proposed vehicles
(2009-14) not made on inputs from buildings of which only 125 procured
district units buildings constructed as of
March 2014
Construction of Police Six districts incurred an expenditure
Training Academy at Der of Rs.72.68 crore towards hiring of
gaon incomplete with 18% private vehicles
progress as of June 2014,
despite scheduled date of
completion of July 2014
Centre released only 63% 98% shortage of residential 26 Bullet Proof vehicles valuing
of funds allocated for MPF. buildings in 8 selected Rs.3.51 crore procured under MPF for
State released 89% of the districts operational purposes were allotted for
allocated funds, and the Basic facilities lacking non-operational duties
police could utilise 68% of in 24 test-checked police
the funds realeased. stations- reception room(6
PS), interrogation room (14
PS), weapons store (3 PS),
restrooms (18 PS)
Bihar (2011- Annual Action Plans (AAPs) Only 47% police stations Only 24.6% of the proposed vehicles
16) not made on inputs from have their own buildings procured
district units
Only 55% of the funds from Only 35% of the proposed In the motorcycle section, the
the approved plan received units constructed procurement was only 27%
under the MPF scheme
Because of delay in release of Housing shortage of 76% in There was a delay in procurement
proportionate fund allocation upper subordinates and 83% procedure of 4 anti-landmine vehicle
by state, central fund of the for lower subordinates (ALMV) which resulted in avoidable
amount of Rs.119.82 crore expenditure of Rs1.81 crore towards
lapsed. them
71% funds unspent due to
delay in tender finalization of
tender for procurement
Gujarat Non-utilisation of planned Only 26% utilisation of The availability of vehicles is better
(2009-15) fund of Rs 141.88 crore funds by Gujarat State in Gujarat than most of the states,
during 2013-15 Police Housing Corporation though there were instances of their
Limited (GSPHCL) non-judicious distribution among
various police stations
Slow progress because of
non-availability of land,
pending approval of layout
plans, delay in finalisation of
tenders, etc.
GSPHCL failed to complete
many important works at
Police academy, Karai

Himachal No improvement in the Only 12% of police One light vehicle was available in each
Pradesh financial management of personnel provided housing of the 13 selected PSs whereas two
(2011-16) MPF since the previous audit facility as of May 2016 motor cycles were provided to each
(2004-09) of the eight PSs and one motor cycle
Central government released Satisfaction level of housing each was provided in other five PS
only 53% of the MPF funds facility ranged between 11%
to 16%
Percentage of non-utlization 23 civil works lying In spite of MHA directions (March

of funds ranged between 21% incomplete for 7 years 2000), 37 out of 54 light vehicles
and 87% in each financial purchased during 2011-13 were
year retained at other units instead of the
PS
Madhya 52% of the total expenditure More than two thirds of No motorcycles in six police stations
Pradesh incurred in 2015-16, subordinate police personnel and two outposts of the 39 test
(2011-16) otherwise utilisation could not be provided checked locations
percentages very low government accommodation
because of shortage
75% shortage of good Shortage of 14,107 motorcycles,
houses in selected police 720 heavy vehicles, 636 medium
stations/outposts (PS/OPs) vehicles (jeeps) and still went on to
buy 102 Sedan cars (Rs.5.88 crores) in
violation of MPF guidelines
Toilets and restrooms for In selected districts, the audit found

women personnel lacking in presence of 626 drivers against
PS/Ops availability of 1302 vehicles
Only 57% of the sanctioned Response Time was not monitored

houses/buildings were properly due to lack of flow of data to
completed during 2011-16 PHQ, thus it could not be noted at the
state level
Maharashtra Only 38% funds utilised Only 8% of the planned Despite 70% shortfall in procurement
(2011-16) due to delay in submission buildings constructed of vehicles, only 662 vehicles procured
of AAP to the central through the MPF scheme
government
88% non-utilisation under In selected nine districts for audit,
equipment component there were 5955 vehicles and 5174
drivers against requirement of 8933
drivers. Hawldars and Naiks (57%)
were put into driving duties.
Only 6 of the scheduled 71% unspent balance at the While 10 police stations in Mumbai
60 State Level Empowered end of September 2016 in (out of 140) did not have specific
Committee (SLEC) meetings this component columns for recording the response
held time, the remaining 130 police
stations did not record the response
time in the crime registers

Rajasthan Five year strategic plan not 31% of the required staff 72% of the required vehicles available
(2009-14) sent to central government for quarters available as of
approval March 2014.
Non-unitilisation of funds The department purchased 222
ranging between 36% to 79% vehicles (18%) costing Rs.7.97 crore
during 2009-14 for replacement of old vehicles,
against violation of MPF norms. It
also bought 66 cars during 2009-14 in
violation of AAP despite shortage of
operational vehicles.
Physical shortfall in In 9 districts, quarter not In test checked records of PS over
procurement of equipments available for 78% staff 2 months, in only 4% cases police
ranged between 17% and reached the crime scene on time. In
100% 72% cases police response time could
not be worked out
Only 10 SLEC meetings held
against the stipulated 60
Tamil Nadu No approval was obtained Lack of planning in Around 40% of the test-checked PS
(2011-16) from High Powered identification of locations did not have four wheelers and 47%
Committee of Ministry of for the construction of police PS did not have two-wheelers
Home Affairs (MHA) for stations before making
the deviation in procurement proposals in the AAPs
of items pointed out in the
previous audit report (2006-
10)
AAPs were prepared without 63% of the work completed
adequate planning in as of June 2016
ensuring clear titles of land
for buildings.
Uttar 41% non-utilization of MPF 44% shortfall of police Shortage of 27% police vehicles
Pradesh funds stations- 41% in rural areas
(2011-16) and 51% in urban areas
28% state funds lapsed Shortage of Medium Police Vehicles
(used for patrolling) was as high
as 68% in civil police and 75% in
Provincial Armed Constabulary
(PAC)
No funds for Traffic Police Shortage of 48% residential Despite the district police facing
modernisation. Traffic police quarters and 26% barracks critical shortage of vehicles, the
was able to collect only 23% department purchased 10 bulletproof
of the target amount. Tata Safari and eight General Safari
vehicles for Chief Minister’s security
Police strength less than 50% Only four barracks The government also incurred
of the sanctioned strength for women personnel avoidable expenditure of Rs.3.66
constructed as of March crore by purchasing more expensive
2016. and luxurious vehicles (Mercedes
Model M-Guard) for Chief Minister’s
security instead of Land Cruiser
sanctioned earlier.

Uttarakhand Details of strategic plan 25 police stations and 105 69.7% availability of vehicles
(2011-16) not shared with central outposts operating from
government rented premises or from
Only 2.2% of the projected temporary arrangements
required funds were received
by the state
Delays in release of state Only 10.27% of the required Non-availability of drivers for 47%
share ranging from 6 to 48 staff quarters available operational vehicles
months
West Bengal Department could not In violation of MPF In four test checked districts scrutiny
(2009-14) prepare long term strategic guidelines, attempts were indicated acute shortage of vehicles.
plan as the central made to divert funds of Against the requirement of 1219
government did not intimate Rs 5.72 crores meant for vehicles, only 578 were available in
the outlay for the scheme for construction of quarters for these offices indicating a shortage of
these five years constabulary and inspectors 53 per cent
to build houses for 12 IPS
officers
Only 10% of required Only 36% of the required vehicles
staff quarters available procured
in test-checked districts,
and majority of available
quarters in poor condition
Non-utilization leading to Construction of Regional Three Directorates procured 23 cars
loss of central funds, poor Forensic Science Laboratory at a cost of Rs.1.44 crore during the
monitoring of released funds, (RFSL) at Jalpaiguri could scheme years 2011-13 in violation of
shortfall in components, lack not be completed even after MPF scheme guidelines
of transparency and slow lapse of nine years
A mine protected vehicle (MPV)
progress in construction
(procured in July 2005) deployed
under SP, Bankura met with
an accident due to rash driving
(September 2007). The repairable
vehicle was not repaired and left to rot
in open. A new vehicle was purchased
in 2011 at a cost of Rs.88.7 lakh to
replace it

Appendix 8b: CAG Report on Modernisation of Police
Weaponry Forensic Science Cyber Crime Training Infrastructure
Infrastructure Infrastructure
Assam 26% weapons and 26% vacancy in the 88.4% of ultra high Inadequate training
(2009-14) ammunition declared FSL frequency digital of police personnel.
obsolete by Assam mobile radios (UHF New equipment lying
police headquarters in DMR) valuing Rs.1.79 unused in training
March 2014. crore lay uninstalled for centres.
nearly two years
In the test checked No time frame was Assam Cyber Crime Only 166 training staff
districts, 40.74% fixed for submission Investigation Lab against requirement of
weapons were outdated of the analysis (ACCIL), remained 428 (BPR&D norms)
(.303 Rifles) report. There were non-functional as of in four state police
pendency of samples December 2014 training schools.
for settlement ranging
from 15.35 to 81.63%
during 2009-14
Three battalions No Forensic Mobile Post of the IGP (CID) The state could spend
were holding 2,540 unit and Laboratory was lying vacant since less than half of nearly
arms against the at district level existed September 2011. Rs.33 crore provided by
authorisation of 7,058 as of March 2017 Disposal percentage MHA during 2009-14
as of March 2014. according to State ranged between 1.82% for procurement of
Reason for shortage of FSL list at Directorate to 16.15% training equipment
4,518 arms was not on of Forensic Science
record. Services (DFSS)
website.
For 7 sophisticated Shortage of experts/
equipment along with supervising officers
spare parts costing etc., in the CID
Rs.1.08 crore under coupled with frequent
MPF 2011-12, the FSL engagement of CID
certified that the goods officials in law and
were received in good order duty
condition and showed
them as allotted to
different divisions of
FSL. The divisions,
however, did not
receive the material as
of May 2014
Bihar (2011- Heavy shortage of One Forensic Science Under the MPF Required Training
16) modern weapons, 33% Lab (FSL), Patna, scheme, 53 sets of equipment not found
INSAS rifle, 42% in which is functioning CCTV camera with in any of the 5 selected
Pistol, 36% in Carbine, from a police LCD (worth Rs.3.11 district police offices
52% in AK 47 barrack. Two RFSLs, crore) were purchased and CTS Nathnagar
Muzaffarpur and in 2011-12 and
Bhagalpur (non- provided to different
functional) units. However, during
Availability of modern Four DMFU could not test-check, it was In CTS Nathnagar,
weapons- only 59% in be made operational found that in 8 out of only one firing weapon
10 selected districts and due to delay on the part 11 district units, nine available, making three
4 BMPs (Bihar Military of state govt in forming CCTV camera sets not simulators redundant
Police) forensic governing installed.
body despite funds
being sanctioned. The
four DMFUs are now
operational as per the
DFSS list of State FSLs

.303 rifles still in use Only 21% of the The equipments used
(ratio of 46:54) despite sanctioned strength by Bomb squad such
MHA orders to phase posted at FSLs and as Bomb Suit and
out RFSLs Non Linear Junction
Detector were provided
in August and October
2015 to units. But due
to lack of training
for operation, the
equipments remained
idle
Audit says that gap Ballistic division cases
between availability were pending since
and requirement not August 2013 in absence
minimized despite of Test Fired Shell/
MPF funds at disposal Bullet. Further, 443
exhibits of Viscera
were pending in
Toxicology division
for analysis as the post
of Viscera Cutter was
vacant
Gujarat Overall 60% shortage Five RFSLs in Gujarat Non-functional Non-imparting of
(2009-15) of modern weapons and 10 district mobile POLNET was pointed prescribed rounds of
was pointed out earlier forensic units out in the earlier Audit firing practice was
in 2009 audit report Report for the year pointed out in the
ended March 2009, earlier Audit Report for
however it remained the year ended March
non-functional until 2009.
October 2015.
It has been brought 46.5% Remote
down from 60 to 36% Subscriber Units
and norms of BPR&D (RSUs) not installed
had been fulfilled and remaind 53%
RSUs non-operational
as of October 2015
Shortage of AK-47s to Shortage of staff in Radio Frequency Out of 63 units in the
the tune of 76% “as per eight districts Modulator (RFM) State, a few units had
state norms” cards at Gandhinagar not provided any firing
Shortage of personnel police headquarter practice to the trainees
in various cadres of were not functioning as in a year and some
ATS as of March 2015 of October 2015. units had not provided
the prescribed number
of firing practice in a
year
Himachal Against the Shortage of staff in Only three (POLNET) Rs.1.15 crore of state
Pradesh requirement of 9,074 the Forensic Science V-SATs functioning money not utilised by
(2011-16) numbers of 7.62/5.56 Laboratories (FSLs) and other V-SATs were two training centres
mm SLR Rifles and had been reduced to out of order due to for buying disaster
323 numbers of LMGs, 25% as of May 2016 obsolete technology as management training
there were 5,889 Rifles of April 2016. equipment as of
and 191 LMGs as of December 2016. No
March 2016, resulting disaster response force
in shortfall of 35 and constituted.
41% in 7.62/5.56 mm
SLR Rifles and LMGs
respectively

Department had not 18.6% cases disposed CCTNS system not
disposed off/ replaced off in 2012-16 and being fully utilised in
the old weapons with delay ranging from 2 to 10 (out of 13) test-
modern weaponry as 207 days checked PSs due to
of March 2016 non-availability of
technical/ operational
know-how
Madhya 25% deficiency of One FSL, four RFSL Funds for this Deficiency of in-service
Pradesh modern weapons and 50 District Scene component not training in modern
(2011-16) of Crime Units released under MPF weaponry
(March2016) for during 2013-16.
forensic analysis Available information
Inordinate delays is not on comparable Only basic foundation
noticed in procurement parameters courses in six PTS.
of equipment Trainee pass percentage
related to FSLs, reduced from 84% in
Telecommunication, 2011 to 69% in 2016
Intelligence, CID wing
Machines used in Department incurred 29% vacancy of
Arms Workshop for only 16% expenditure teachers in PTS
repair of weapons were on the sanctioned funds and 32% vacancy in
outdated and were not and met 23% of the academies as of March
fit for repair of modern physical targets 2016
weapons The posts of Director In interview of 150
and Senior Joint police personnel in
Director were vacant. selected districts, only
A vacancy of 46% 26% were satisfied with
in Senior Scientific available resources
Officers, 34% in and 45 % were
Scientific Officers, 48% satisfied with career
in Lab Technicians, progression. However,
67% in Lab Assistant 85% stated that they
and 61 % in Lab are motivated to meet
Attendants. the challenges and 81%
Only 46% funds During 2011-16, were satisfied with the
provided for cases received for reward system
modernisation of CID examination increased
utilized from 19,235 to 24,155,
but disposal of cases
decreased from 26,656
to 23,780. 267 cases
pending for more than
12 months (March
2016)
Maharashtra Shortage of 45% Shortage of technical Only 43.7% funds Only 54.8% of allotted
(2011-16) weapons of the total staff, leading to 18% utilised against the total funds were spent on
state need samples pending released budget training during 2011-
investigation as of 16.
January 2017. One
FSL at Mumbai
56% shortage in test- Seven RFSLs at
checked districts as of Nanded, Nagpur,
September 2016 Pune, Aurangabad,
Nashik Amravati and
Kolhapur

Supply of 41% ordered One evidence Shortfall of 60 to 71%
quantity pending from investigator equipment in armed firing practice
Ordinance Factory for detection of during 2011-16 in test-
Board for a preiod drugs in human body checked districts
ranging 6 months to procured at a cost
4.5 years of Rs.98.91 lakh
remained dysfunctional
(July 2016) since its
installation in August
2011
Rajasthan 44.6% utilization of State FSL at Jaipur, Not available on Only 15.23%
(2009-14) budget on weapons as and five regional comparable parameters equipments available
of March 2014. forensic Science in test-checked police
laboratories (RFSL). stations
Five district mobile
forensic units (DMFU)
75% shortage of Shortage of technical
weapons, but only staff varying from
31% of the required 37% to 53% in various
weapons were ordered labratories
as of March 2014.
Purchased weapons Nine posts sanctioned
were not issued to in October 2006 for
police stations, and DNA lab were not
remained stockpiled at filled up till Nov 2014
Jaipur central store
Training for new Various equipments
weapons procured were worth Rs.55 crore were
not included in training lying out of order due
courses to lack of repairs
Tamil Nadu Not available DNA Unit at Shortage of 50.5% Not available
(2011-16) Madurai established communication
in December 2011 equipments as of
without ensuring March 2016.
availability of requisite
cold storage facilities
and technical staff,
resulted in its sub-
optimal usage for 5
years
Main Lab at Chennai The department
and 9 RFSLs besides sanctioned
33 Mobile Forensic Terrestrial Trunked
Science Units (MFSU), Radio (TETRA)
as per the DFSS list of communication project
State FSLs. (Rs.71.51 crore)
without assessing
its feasibility, only
later to be replaced
by APCO. The state
took four years for
the technology to be
adopted.

Uttar DGP took 17 years Only 44 % of the Only 41% of allotted 80% of funds for
Pradesh to prepare a response districts in the state funds spent procurement of
(2011-16) to the MHA’s 1995 could be equipped training equipment
proposal for revision of with mobile forensic during 2011-16 were
Arming policy of State vans and none of the surrendered due to
Police 500 circles have been delay in finalization of
provided such facility procurements
About 48% of the Four FSLs at Lucknow, About 48% of the Equipment such as
police force in the state Agra, Varanasi and police force was not interactive fire arms
still using .303 bore Ghaziabad while FIVE provided with hand simulators; cybercrime
rifle which had been RFSLs proposed and held sets in violation of lab equipment etc.
declared obsolete more under construction at norms could not be procured
than 20 years ago Allahabad, Ghaziabad, due to delays in tender
Gorakhpur, Jhansi and finalisation
Moradabad
The Central Reserve at Department failed 67.4% sets had
Sitapur lacked proper to utilise 55% of expired and needed
storage buildings, fire the budget allotted replacement, but new
security system and for purchase of lab sets not procured
surveillance system equipment
Arms like 9mm MP5, 31% staff against the Only 38 districts The training capacity
12 Bore Pump Action sanctioned strength in (51%) were found of existing training
Gun (PAG), Stun all the 75 districts and to have functional centres in the State is
Grenade and UBGL 37% in 15 test checked Polnet (a satellite highly inadequate with
were provided first time districts based communication a deficiency of 63 to
to state ATS in 2013-15 system) 65% during 2011-16.
but ammunition for The department did not
these weapons have still establish three Police
not been provided Training Schools (PTS)
as of March 2017
39% CCTVs found
not functional in test-
checked districts
61% shortage on group Shortage of staff Crime and Criminals Police training
B gazetted officers increased from 47 to Tracking Network institutions had huge
in PAC at the end of 67% during 2012-16 System (CCTNS) has shortage of indoor
2015-16 still not been made training instructors
fully operational ranging from 36 to 68%
during 2011-16
Uttarakhand 63.64% of the total Only 40.6% utilisation 80.6% police stations PTC lacked
(2011-16) required weapons of allotted funds covered under CCTNS infrastructure
available and 39.04% but lack of citizens’ like swimming
of required modern portal for online pool, assault/
weapons available registration of FIRs obstacle courses,
and tracking of cases mountaineering class
room/wall, audio-
visual facilities and hill
driving facilities
Despite the gap, only RFSL and DMFUs not PTC did not have a
Rs.1.12 crore was set up even after lapse firing range which is
spent for procurement of more than five years. a fundamental part
of weapons, which According to the DFSS of the training; Only
constituted 2.41% of list of State FSLs, the 5,000 litres of drinking
the total budget allotted state has one RFSL water per day was
under MPF 2011-16 (Uddhamsingh Nagar) being provided against
and two DMFU as of requirement of 36,000
March 2017. litres per day.

37% required bullet FSL Dehradun had Only seven out of The training school
proof jackets and only one scientist 125 connected PS lacked basic amenities
41% required helmets available for the study were registering the like beds, chairs,
available of samples and huge FIRs offline due to hot water, library,
shortage of technical connectivity issues pharmacist
cadres
West Bengal 48% of the required One FSL and two Not available on Existing training
(2009-14) arms procured by the MFSUs comparable parameters system/ infrastructure
state. Kolkata Police Only 9.6% funds in Swami Vivekananda
received only 21% of utilised during 2009-14 State Police Academy
its requirement (SVSPA), Barrackpore
was far short of the
norms prescribed by
the BPR&D
In five districts shortage Shortage of technical Live training was not
of 71% of arms as on manpower in FSL imparted for handling
March 2014 ranged between 38 some of the weapons,
and 88%. At least while scale of actual
88% positions of practice, wherever
lab assistant to 50% done, was substantially
positions of senior lower than BPR&D
scientific assistants are norms
vacant,

Spir2018 PDF

Uploaded by

Copyright:

Available Formats

Spir2018 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spir2018 PDF

Uploaded by

Copyright:

Available Formats

Copyright © Common Cause & CSDS 2018

29 Rajpur Road, Civil Lines, Delhi -110054

Common Cause House, 5, Institutional Area,

Design & Print by GENESIS | 9810033682

Analysis and report writing Research Support

State Coordinators State Supervisors

Chapter 1: Let the Numbers Speak: Police Performance Review 14

Chapter 2: Experience with the Police 33

Chapter 3: Trust in Police 54

Chapter 4: People’s Perception of Discrimination by the Police 70

Chapter 5: Fear of Police and Attitudes towards its Excesses 90

Chapter 6: People’s Perception on Different Aspects of Policing 105

Chapter 7: Analysing CAG Audit Reports 124

Chapter 8: Conclusion 133

Appendix 1: Technical details of study design and sample 141

Appendix 2: Questionnaire 144

Appendix 3: Details of how the indices were constructed 154

Appendix 4: Details of state rankings 160

Appendix 5: State wise findings from the survey 171

Appendix 6a: Police performance review using official data 185

Appendix 6b: Police performance review using objective data 192

Appendix 7: States’ compliance with Supreme Court directives 201

Appendix 8: CAG Report on Modernisation of Police 205

Table no. Title of the tables Page no.

Status of Policing in India Report 2018 | 1

2 | Status of Policing in India Report 2018

Figure no. Title of the figures Page no.

Status of Policing in India Report 2018 | 3

4 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 5

AAP: Annual Action Plan GDP: Gross Domestic Product

AIDS: Acquired Immunodeficiency Syndrome GoI: Government of India

ACCIL: Assam Cyber Crime Investigation GPI: Global Peace Index

AP: Andhra Pradesh HC: Head Constable

ASI: Assistant Sub-Inspector HP: Himachal Pradesh

BPRD: Bureau of Police Research & IGP: Inspector General of Police

CTS: Constable Training School MPM: Modern Prison Manual

CFSL: Central Forensic Science Laboratories MPV: Medium Police Vehicles

CPO: Central Police Organisations NFHS: National Family Health Survey

CrPC: Criminal Procedure Code NPC: National Police Commission

CTS: Constable Training School NSC: National Security Commission

DFSS: Directorate of Forensic Sciences OBCs: Other Backward Castes

6 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 7

8 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 9

Policemen retaliated with batons after a protest by lawyers turned violent.

10 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 11

12 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 13

14 | Status of Policing in India Report 2018

1.1 A momentous non-compliance 7. National Security Commission (NSC) for

Status of Policing in India Report 2018 | 15

16 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 17

Table 1.1: Crime rate Index

18 | Status of Policing in India Report 2018

Status of Policing in India Report 2018 | 19