A STUDY ON PETER PRINCIPLE EFFECT IN SOFTWARE DEVELOPMENT FIRMS OF SRI LANKA

Melanie Samaratunga

(9046)

Master of Business Administration in Management of Technology

Department of Management of Technology

University of Moratuwa Sri Lanka

November 2011

ii

A STUDY ON PETER PRINCIPLE EFFECT IN SOFTWARE DEVELOPMENT FIRMS OF SRI LANKA

Melanie Samaratunga

(9046)

Dissertation submitted in partial fulfillment of the requirements for the degree of Master of Business Administration in Management of Technology

Department of Management of Technology

University of Moratuwa Sri Lanka

November 2011

iv

DECLARATION
I declare that this is my own work and this dissertation does not incorporate without acknowledgement any material previously submitted for a Degree or Diploma in any other University or institute of higher learning and to the best of my knowledge and belief it does not contain any material previously published or written by another person except where the acknowledgement is made in the text. Also, I hereby grant to University of Moratuwa the non-exclusive right to reproduce and distribute my dissertation, in whole or in part in print, electronic or other medium. I retain the right to use this content in whole or part in future works (such as articles or books).

. M. Samaratunga

. Date:

The above candidate has carried out research for the Masters dissertation under my supervision.

. Prof. Vathsala Wickramasinghe

. Date

ABSTRACT
The Peter Principle that individuals in an organization rise to their level of incompetence represents potential problems for all employees. Wondering if the Peter Principle is prevalent in the software development firms of Sri Lanka, the author revisited Dr. Laurence Peters study of 1969, The Peter PrincipleWhy Things Always Go Wrong, which achieved best-seller status and soon became a part of the lexicon of the business world. A survey based study was carried out using respondents from selected software development firms to identify if the effect exists. This study shows that the Peter Principle, the universal phenomenon in which employees, around the world, are said to rise to their level of incompetence, is ingrained in the software development firms of Sri Lanka as well. The behaviors embodied in the Peter Principle still have disrupting effects that occur only too frequently in organizations. As a result, the Peter Principle cannot be ignored. Its effects, however, can be remedied by improving the quality of performance and rewards management practices, recruitment and selection practices and by providing extensive training before employees reach their ultimate level of incompetence.

ii

ACKNOWLEDGMENT
I would like to express my deepest appreciation to my supervisor, Prof. Vathsala Wickramasinghe for her valuable support, ideas, and criticism during the springtime. Her knowledge of the related literature, strong managerial perspective, and willingness to exchange and shape ideas were crucial in the overall development of this study. Your contributions had been of great value to the final result. Additionally, I would like to convey my deep sense of appreciation for the continuous help received from the team members of my organization, Virtusa Corporation. I would like to thank my family members for the contribution towards the success of the research. They have sacrificed in many ways by allowing me to spend more time on the study. Further, the contribution of each and every respondent of the survey who extended their support in the data collection process is very much appreciated. Last but not least I thank all those whose names were, though not mentioned for their help and encouragement in completing this research study.

iii

TABLE OF CONTENTS
DECLARATION....................................................................................................................................I ABSTRACT...........................................................................................................................................II ACKNOWLEDGMENT.....................................................................................................................III LIST OF TABLES.................................................................................................................................V LIST OF FIGURES.............................................................................................................................VI LIST OF ABBREVIATIONS............................................................................................................VII
Table 1.1: Summary of various researches carried out on software project failures................................................3 Table 1.2: Summary of Standish chaos reports........................................................................................................6

1.3 RESEARCH OBJECTIVES.......................................................................................................................8

Figure 3.1: Conceptual Model...............................................................................................................................26 Table 3.1: Hypothesis statements.........................................................................................................................27 Figure 3.2: Conceptual Model with Hypotheses....................................................................................................28 Table 3.2 : Operationalization of the identified variables......................................................................................32

3.6.2 Data collection instrument...................................................................................................37 3.6.3 The method of data collection and analysis.........................................................................38 3.7. SUMMARY......................................................................................................................................38
Table 4.1: Summary table for number of employees in organizations ..............................................................................................................................................................................40 Table 4.2: Summary table for number of employees in different organization structures ..............................................................................................................................................................................40 Table 4.4: Summary table for the gender distribution of the respondents ..............................................................................................................................................................................41 Table 4.5: Summary table for the marital status of the respondents ......................................................................41 Table 4.7: Summary table for the designation level of the respondents.................................................................43 Table 4.8: Summary table for the years of experience in IT of the respondents....................................................43 Table 4.9: Summary table for the years of experience in the current company of the respondents........................44 Table 4.10: Summary table for the years of experience in the current post of the respondents..............................44

iv

LIST OF TABLES
Table 1.1: Summary of various researches carried out on software project failures.......Error: Reference source not found Table 1.2: Summary of Standish chaos reports....................................................Error: Reference source not found Table 3.1: Hypothesis statements.........................................................................Error: Reference source not found Table 3.2 : Operationalization of the identified variables.....................................Error: Reference source not found Table 3.3: Companies selected for the sample.....................................................Error: Reference source not found Table 4.1: Summary table for number of employees in organizations.................Error: Reference source not found Table 4.2: Summary table for number of employees in different organization structures....Error: Reference source not found Table 4.3: Summary table for the age distribution of the respondents..................Error: Reference source not found Table 4.4: Summary table for the gender distribution of the respondents............Error: Reference source not found Table 4.5: Summary table for the marital status of the respondents.....................Error: Reference source not found Table 4.6: Summary table for the education level of the respondents..................Error: Reference source not found Table 4.7: Summary table for the designation level of the respondents...............Error: Reference source not found Table 4.8: Summary table for the years of experience in IT of the respondents...Error: Reference source not found Table 4.9: Summary table for the years of experience in the current company of the respondents..Error: Reference source not found Table 4.10: Summary table for the years of experience in the current post of the respondents........Error: Reference source not found Table 4.11: Reliability analysis for the factors that determine the existence of Peter Principle effect..............Error: Reference source not found Table 4.12: Principle component analysis for the factors that determine the existence of Peter Principle effect .............................................................................................................................Error: Reference source not found Table 4.13: Rotated component matrix for the factors that determine the existence of Peter Principle effect. .Error: Reference source not found Table 4.14: Reliability analysis results for quality of performance and rewards management practices related variables..............................................................................................................Error: Reference source not found Table 4.15: Principle component analysis results for quality of performance and rewards management practices related variables...................................................................................................Error: Reference source not found Table 4.16: Factor loadings for quality of performance and rewards management practices related variables. Error: Reference source not found Table 4.17: Reliability analysis results for quality of recruitment management practices related variables.....Error: Reference source not found Table 4.18: Principle component analysis results for quality of recruitment management practices related variables..............................................................................................................Error: Reference source not found Table 4.19: Factor loadings for quality of recruitment management practices related variables......Error: Reference source not found Table 4.20: Reliability analysis results for quality of human resource development practices related variables .............................................................................................................................Error: Reference source not found Table 4.21: Principle component analysis results for quality of human resource development practices related variables..............................................................................................................Error: Reference source not found Table 4.22: Factor loadings for quality of human resource development practices related variables...............Error: Reference source not found Table 4.23: Descriptive statistics for the factors that explain the existence of Peter Principle effect................Error: Reference source not found Table 4.24: Descriptive statistics for the factors that explain the quality of performance and rewards management practices...............................................................................................................Error: Reference source not found Table 4.25: Descriptive statistics for the factors that explain the quality of selection and recruitment practices .............................................................................................................................Error: Reference source not found

Table 4.26: Descriptive statistics for the factors that explain the quality of human resource development practices .............................................................................................................................Error: Reference source not found Table 4.27: Correlation analysis results...............................................................Error: Reference source not found Table 4.28: Model summary for the existence of Peter Principle related behavior patterns and the contributing factors..................................................................................................................Error: Reference source not found Table 4.29: ANOVA table for existence of Peter Principle related behavior patterns and the contributing factors .............................................................................................................................Error: Reference source not found Table 4.30: Coefficients for existence of Peter Principle related behavior patterns and the contributing factors .............................................................................................................................Error: Reference source not found Table 4.31: Model Summary for the existence of incompetent employees in the higher levels of the hierarchy and the contributing factors........................................................................................Error: Reference source not found Table 4.32: ANOVA table for the existence of incompetent employees in the higher levels of the hierarchy and the contributing factors........................................................................................Error: Reference source not found Table 4.33: Coefficients for the existence of incompetent employees in the higher levels of the hierarchy and the contributing factors..............................................................................................Error: Reference source not found Table 4.34: Group Statistics for structure of the company...................................Error: Reference source not found

LIST OF FIGURES
Figure 3.1: Conceptual Model.............................................................................Error: Reference source not found Figure 3.2: Conceptual Model with Hypotheses..................................................Error: Reference source not found Figure 4.1: Revised Conceptual Model.................................................................................................................47

vi

LIST OF ABBREVIATIONS
IT ICT IFS BPO RPO Information Technology Information Communication Technology Industrial and Financial Systems Business Process Outsourcing Recruitment Process Outsourcing

SPSS Statistical Package for the Social Sciences

vii

CHAPTER 1 - INTRODUCTION

1.1 Introduction Sri Lanka is emerging as a global IT-BPO destination of choice in a number of key focus domain areas. It is ranked among the Top 50 Global Outsourcing destinations by A.T. Kearney according to Sri Lanka Business Portal - Trade Information (2011). In addition, Sri Lanka has emerged as the most preferred ICT/BPO hub in the Asian region and is the destination renowned for Best-Of-Breed in Global Market. Sri Lanka acts as an off-shore development center for several Fortune 500 companies from the USA, Ireland, UK, Australia, etc and joint venture development center companies from Sweden, Norway, USA, Japan etc. Some business entities that have set-up their operations in the island include: HSBC, Industrial & Financial Systems (IFS), Amba Research, RR Donnelley, Quattro, Virtusa, eCollege, Eurocenter, Aepona, Millennium Information Technology and Innodata Isogen etc. At present there are over 300 IT and BPO companies that operate in Sri Lanka, mostly small and medium companies and a few large global players. According to Sri Lanka Business Portal - Trade Information (2011), Sri Lanka offers a rapidly growing, highly adaptable, innovative and loyal workforce. Currently, over 50,000 are employed in the IT and BPO industry in Colombo and the workforce is growing at over 20% year-on-year. The workforce is stable with very low attrition rates ranging from 10-15%. The Software services sector focuses on telecommunication, banking, financial services and insurance and software testing. Earnings from exports of IT-BPO sector have shown a steady upward trend during the past decade, and annual exports of the ICT sector for the last three years recorded as US $ 213 million in 2007, US $ 256 million in 2008 and US $ 271 million in 2009. The industry has set a target of $ 2bn in export revenue from IT-BPO sector by 2012.

Sri Lankan IT industry shows good potential, but in the long term for any industry to succeed it is highly necessary that a competent workforce is available. For this talent should be managed effectively. Talent management is very important for a number of reasons. It is about people and, without people an organization would not exist. Employers should understand where the strengths and weaknesses of their employees lie, in order to ensure they are correctly deployed across the business. Identifying where the promising talent lies, allows a company to plan ahead and nurture staff, so their personal development is serving organizational development benefitting both parties. One of the challenges of talent management in business is, knowing whether a seemingly talented individual will thrive if promoted to a new task or area. Once assigned and moved into a talent pool, these individuals must also be managed properly, in terms of their own expectations and those of the employer. Peter and Hull (1969), in their book, The Peter Principle state that, in a hierarchy, every employee tends to rise to his level of incompetence, which means that in a hierarchy, members are promoted so long as they work competently. Eventually they are promoted to a position at which they are no longer competent i.e. their level of incompetence and there they remain, being unable to earn further promotions. To further elaborate on the Peter Principle theory, it is a concept that in organizations, new employees typically start in the lower ranks, but when they prove to be competent in their current designation, they get promoted to a higher rank, generally management. This process of climbing up the hierarchical ladder can go on indefinitely, until the employee reaches a position where he or she is no longer competent. At that moment, the process typically stops, since the established rules of bureaucracies make it very difficult to demote someone to a lower rank, even if that person would be a much better fit and happier in a lower rank. The net result of this principle is that, most of the higher levels of a bureaucracy will be filled by incompetent people, who got there because they were quite good at doing different and usually, but not always, easier work than the work they are expected to perform at present. For example if you're a proficient and effective software developer, you're

most likely demonstrating high competence in your job right now. As a result of your performance, your valuable contribution results in a promotion to a management position. In this new position, you now do few of the original tasks which gained you acclaim. In fact, little of your current job remains enjoyable, therefore your heart is no longer in your work, and it shows. Given this, promotions stop, and there you stay, until you retire or your company goes under due to mismanagement. According to this principle, work is accomplished by those employees who have not reached their level of incompetence. Thus we can see why organizations still function even as Peter Principled employees accept promotions. Dr. Peter (1969) provides an insightful analysis of why so many positions in so many organizations seem to be populated by employees who seem incompetent. This concept is likely to be ignored by most senior managers, since to admit one's organization is suffering from this bureaucratic dilemma is admission that, people have been improperly promoted. This, in turn, suggests that senior management might have attained their own level of incompetence, and the problem is easily ignored, in case it become suggested that senior management be more closely examined for their incompetence. Once a company forms a culture of incompetence, only the incompetent staff will remain, and the competent ones will eventually get frustrated and leave. As a result the organizations growth will hinder as they have incompetent employees at many levels. 1.2 Research Problem According to the article Facts and Figures Why Technology Projects Fail (2011), a number of studies have been completed that look into the success and failure rates of software projects. These studies indicate that serious problems exist across the industry as a whole. Table 1.1 summarizes some recent reports. Table 1.1: Summary of various researches carried out on software project failures
Source Geneca Type of Survey Interview based study of software projects Date 20102011 Result Interviews with 600 people closely involved in software development projects finds that even at the start of a project many people

expect their projects to fail Fuzzy business objectives, out-ofsync stakeholders, and excessive rework mean that 75% of project participants lack confidence that their projects will succeed. A truly stunning reported 78% that of the respondents

Business is usually or always out KPMG (New Zealand) Survey of 100 Dec 2010 of sync with project requirements KPMG survey of Project Management practices in New Zealand finds some truly startling results: Survey shows an incredible 70% of organizations have suffered at least one project failure in the prior 12 months 50% of respondents also indicated that their project failed to consistently achieve what they set IBM Survey change executives of 1,500 Oct 2008 out to achieve IBM survey in the success / failure rates of change projects finds: Only 40% of projects met schedule, budget and quality goals Best organizations are 10 times more successful than worst organizations Biggest barriers to success listed as people factors: Changing mindsets and attitudes - 58%. Corporate culture - 49%. United States Government Accountability Review of federally funded projects technology 31 2008 Jul Underestimation Lack of senior of complexity management support - 32%. listed as a factor in 35% of projects Study finds 413 of 840 (49%) federally funded IT projects are either poorly planned, poorly performing or both.

businesses across a broad cross-section of industries

management

Office Information Systems Audit and Control Association (ISACA)

400 respondents

May 2008

Key findings 43% of organizations have suffered a recent project failure At a typical enterprise 20% of technology investments are not fully realized Study of government projects reveals $4billion in wasted efforts as a result of failed projects Only 30% of our projects and programs are successful -Joe Harley, programme and systems delivery officer at the Department for Work and Pensions 70% of respondents had been involved in a project they knew would fail right from the start Success rates for Agile projects 72%, success rate for traditional approaches 63% In just a 12 month period 49% of organizations had suffered a recent project failure In the same period only 2% of organizations reported that all of their projects achieved the desired benefits 86% of organizations reported a shortfall of at least 25% of targeted benefits across their portfolio of projects

Guardian Newspaper (UK )

Investigation

into

5 2008

Jan

government waste in the UK since year 2000

Dr Dobbs Journal

586 email

respondents survey

to (Dr

Aug 2007

Dobbs subscriber list)

KPMG - Global IT Project Management Survey

Survey

of

600

2005

organizations globally

Many organizations fail to measure benefits so they are unaware of their true status in terms of benefits realization

The following research findings were listed in the article, Software Project Failure Costs Billions (2008). Standish Chaos Reports: Standish reports define success as projects on budget, of cost, and with expected functionality. Standish findings by year, are shown in table 1.2. Table 1.2: Summary of Standish chaos reports
Succeeded Failed Challenged 1994 16% 31% 53% 1996 27% 40% 33% 1998 26% 28% 46% 2000 28% 23% 49% 2002 34% 15% 51% 2004 29% 18% 53% 2009 32% 24% 44%

Mercer Consulting: When the true costs are added up, as many as 80% of technology projects actually cost more than they return. It is not done intentionally but the costs are always underestimated and the benefits are always overestimated.

Oxford University: Regarding IT project success reported the following figures. Successful: 16% Challenged: 74% Abandoned: 10%

British Computer Society: In 2004, the UK public sector spent an estimated 12.4 bn. on software and the overall amount spent on IT was about 22.6 Billion British Pounds. From those projects the success rate was 16%. National Institute of Standards and Technology (NIST): According to NIST

Software defects cost nearly $60 billion, annually 80% of development costs involve identifying and correcting defects

Tata Consultancy: In 2007 Tata consultancy reported that 62% of organizations experienced IT projects that failed to meet their schedules 49% suffered from budget overruns 47% had higher-than-expected maintenance costs 41% failed to deliver the expected business value and ROI 33% file to perform against expectations

From Bob Lawhorn presentation on software failure (March 2010): Poorly defined applications (miscommunication between business and IT) contribute to a 66% project failure rate, costing U.S. businesses at least $30 billion every year (Forrester Research) 60% 80% of project failures can be attributed directly to poor requirements gathering, analysis, and management (Meta Group) 50% are rolled back out of production (Gartner) 40% of problems are found by end users (Gartner) 25% 40% of all spending on projects is wasted as a result of re-work (Carnegie Mellon) Up to 80% of budgets are consumed fixing self-inflicted problems (Dynamic Markets Limited 2007 Study) Considering all of the research findings above, it is clear that the software project failure rates are high, globally. Most of the problems have occurred due to poor

quality products, lack of proper leadership, unrealistic estimations etc. Hence it is reasonable to assume that these problems might have occurred due to incompetency of the employees i.e. poor quality code could have been a result of software developed by unskilled software developers, inability to meet deadlines would have been a result of estimates created by incompetent project leads etc. As mentioned earlier incompetence in an organizations hierarchy denotes the existence of Peter Principle effect. The risks of over promoting and under developing employees lead to many problems in an organization. One such problem is that, it could lead to the perception that a culture of incompetence is being fostered, resulting in your competent staff becoming frustrated and leaving the organization. Problems as such eventually lead to the downfall of the company. Hence the Peter Principle effect needs to be avoided in organizations for them to grow and succeed. This research was designed to identify if Sri Lankan software development firms are victims of this effect and to develop strategies to avoid it. 1.3 Research Objectives
The research objectives are as follows. Objective 1: To identify if the Peter Principle effect exists in Software development firms of Sri Lanka Objective 2:

To identify and analyze the impact of the determining factors that associate with the existence of Peter Principle effect
Objective 3:

To propose recommendations to help software firms in Sri Lanka to avoid the Peter
Principle effect

1.4 Significance of the Study

A decisive success factor for software producers is the quality of their software. Software systems must meet steadily rising demands regarding stability, performance, usability and maintainability. Economic indicators such as development time and costs need to be considered as well. All these together define high quality software which will define the success of the software development organization. The human resources are the most important assets of any organization. The success or failure of an organization is largely dependent on the caliber of the people working therein. Without positive and creative contributions from people, organizations cannot progress and prosper. In order to achieve the goals of an organization, it should have a highly competent workforce and the right people should be employed in the right jobs. The absence of the above and the presence of incompetent employees in the wrong jobs denote the existence of Peter Principle effect. As mentioned earlier, once a company forms a culture of incompetence, only the incompetent staff will remain, and the competent ones will get frustrated and leave the company making the companys growth hinder as they have incompetent employees at many levels. Hence it is important to avoid this malady to ensure that the software industry in Sri Lanka grows to its maximum potential. This study intended to identify, if the Peter Principle effect exists in software development organizations of Sri Lanka and to provide remedies to avoid it. This will help software development organizations to properly manage talent and identify where the promising talent lies, allowing them to plan ahead and nurture staff, so their personal development is serving organizational development benefiting both parties. 1.5 Scope of the Research Study The main focus area of the study was to identify if the Peter Principle effect exists in software development organizations of Sri Lanka and to recommend a set of solutions to avoid the effect.

The scope of the research will be limited to software development organizations within Sri Lanka. The selected software organizations use either tall or flat organizational structures. The sample of respondents will be selected at random from the selected organizations. The software development firms selected will be small, medium or large scale. The research methodology will be focusing on collecting information from the employees through online questionnaires. The questionnaires will be targeted towards Software Engineers, Business Systems Analysts, Quality Assurance Engineers, Team Leads, Project Managers and Architects who are in the middle level management or the lower levels of the hierarchy so that the sample selected will have fewer employees who have reached their highest level of incompetence.

1.6 Chapter Outline The outline of the report is as follows. Chapter 2, discusses the previous researches carried out on Peter Principle effect by various researchers. Chapter 3, discusses the design of the study, developing the conceptual model based on the literature that was reviewed in chapter 2, development of hypotheses for the purpose of checking the validity of the relationships of among constructs, data collection methods and operationalization of measurements. Chapter 4, discusses the data analysis and hypotheses testing carried out. It includes primary and secondary data analysis, reliability and validity testing of the data set, descriptive data analysis, correlation analysis and regression analysis. Chapter 5, concludes the entire study and discusses the findings of the study and recommends suggestions, guidelines and strategies to avoid the Peter Principle in Sri Lankan IT Firms. The chapter will also highlight limitations of the study and future research opportunities.

10

CHAPTER 2 LITERATURE REVIEW

2.1 Introduction This chapter contains a comprehensive analysis on the researches carried out previously. It is important to understand the viewpoints of the earlier studies to get a thorough understanding on the research area and ensure that the research covers all the required areas in the formation of a comprehensive view of the total scenario. Thus, the literature review would create the foundation of the total study based on the previous literature and the discussions which have taken place on the topic. 2.2 The Peter Principle In a hierarchy every employee tends to rise to his level of incompetence (Peter and Hull 1969, p.25) The above quote was taken from Dr.Laurence J.Peter and Raymond Hulls bestselling book: The Peter Principle Why things always go wrong? This sets the outline for this research. The Peter Principle theory was introduced by Dr. Laurence J. Peter in the year 1969. He was a sociologist, who taught at the University of British Columbia before becoming a professor of education at the University of Southern California (Taylor, 1969). He was an expert in the area of hierarchical incompetence and wrote a couple of books about this controversial topic. His first book, The Peter Principle - Why things always go wrong? introduced the Peter Principle to the world. He claimed that in a hierarchy, every employee tends to rise to his level of incompetence (Peter, 1969, p. 25). Additionally, his view was that one will advance to his highest level of competence and consequently get promoted to a position where he will be incompetent. The book contains many real-world examples and thought-provoking explanations of human behavior, including the fact that Every organization

11

contained a number of persons who could not do their jobs, and that occupational incompetence is everywhere (Peter, 1969, p. 20). It is deciphered in a multifactor framework and is based on a study in which, data in the form of hundreds of case histories were collected through observing overt behavior and avoiding introspection or inferences. Peter and Hull(1969) concluded that an employees process of climbing up the hierarchical ladder in an organization can go on indefinitely until the employee reaches a position where he or she is no longer competent and is, thus, regarded as incompetent. It states that in time, every post tends to be occupied by an employee who is incompetent to carry out their duties and adds that, work is accomplished by those employees who have not yet reached their level of incompetence. Why a research is still needed about the Peter Principle, which was developed over 40 years ago? Schapp and Ogulink(2009), in their recent research paper, shows that the Peter Principle, is still prevalent today and little regarding its presence has changed since 1969. 73% of the participants in their study had said that they have seen a Peter Principle situation happen within the last five years. The ability of an employee is determined not by outsiders, but by his or her superior in the hierarchy. At that point if the employee has reached his level of incompetence, the upward process usually stops since the recognized rules of organizations make it very difficult to demote someone, even if that person would fit in much better in a lower job. The end result is that most of the higher levels of an organization will be filled by inept people. For e.g. managers, who got there because they had previously shown competence in doing a task, different than the new one they are expected to do. Essentially, Peter said that as employees move upward through the chain of command, they do worse, as managers, than they did before having been promoted. And this phenomenon is not limited in scope. According to Peter and Hull(1972, p. 24) Sooner or later, this could happen to every employee in every hierarchy business, industry, trade-unions, politics, government, armed forces, religion, and education.

12

Peter and Hull (1969) points out several symptoms to identify the existence of Peter Principle effect in organizations. These come in the form of behavioral patterns of incompetent employees or techniques used by organizations to deal with employees, who have reached their level of incompetence. Some of the important symptoms are as follows: 2.2.1 Percussive Sublimation According to Peter and Hull (1969, p.37) this is a pseudo-promotion technique where an incompetent employee is promoted to a higher position which brings on no new responsibility, but unclogs the rest of the hierarchy. This is commonly known as kicking a man upstairs. The objective of Percussive Sublimation usually is to deceive the outside world. It camouflages the flaws in the employers promotion policy, supports staff morale, and maintains the hierarchy in lieu of firing the incompetent person which might result in him getting another job with a competitor where, despite his incompetence, his knowledge could be dangerous.

2.2.2 Lateral Arabesque Lateral Arabesque is another pseudo-promotion. Without being raised in rank, sometimes without even a pay raise, the incompetent employee is given a new and longer title and is moved to an office in a remote part of the building, states Peter and Hull (1969, p.39). This is similar to Percussive Sublimation where the main objective is to unclog the hierarchy by removing the incompetent employee, so that the workflow can run smoothly.

2.2.3 Hierarchical Exfoliation Peter and Hull (1969, p.45) states that, in most hierarchies, super-competence is more objectionable than incompetence. Ordinary competence is no cause for dismissal: it is simply a bar for promotion. On the other hand super-competence often leads to dismissal as it disrupts the hierarchy. The super-competents, who seem

13

to know everything and do everything well at all levels in anticipation of always moving up, are more likely to be fired because they disrupt the hierarchy. In Peter's view, these people attract too much attention to themselves, worrying others in the organization so much, that they disrupt the super-competent's climb to the top. Hence super-competent employees often get stuck in their ranks or get dismissed to preserve the hierarchy. This phenomenon is often related with negative selection. Negative selection is a political process that occurs especially in rigid hierarchies, most notably dictatorships. The person on the top of the hierarchy, wishing to remain in power forever, chooses his associates with the prime criterion of incompetence they must not be competent enough to remove him from power. The associates do the same with those below them in the hierarchy, and the hierarchy is progressively filled with more and more incompetent people. If the dictator sees that he is threatened nonetheless, he will remove those that threaten him from their positions and emptied positions in the hierarchy are normally filled with people from those who were less competent than their previous masters. So, over the course of time, the hierarchy becomes less and less effective. As this happens relatively often, once the dictator dies, or is removed by some external influence, what remains is a grossly ineffective hierarchy.

2.2.4 Peters Inversion Peters Inversion also known as the professional automaton, is when there are employees who have little or no capacity for independent judgment but always obey and never decide. These are the kind of employees who show obsessive concern with filling out forms correctly permitting no deviations from established routine. To a person who follows the professional automaton, it is clear that means are more important than ends; the paperwork is more important than the purpose for which it was originally designed. He no longer sees himself as existing to serve the public; he sees the public as the raw material that serves to maintain him, the forms, the rituals,

14

and the hierarchy. Unfortunately for the public, the automaton appears to be competent from the hierarchys point of view. As a result he remains eligible for promotion until by some mischance he is elevated into a position where he absolutely has to make a decision. It is at that point that he reaches his level of incompetence, wrote Peter and Hull (1969, p.41). These people are often managed by incompetent managers who care about sycophancy, courtesy towards bosses, etc. more than one's internal efficiency. 2.2.5 Paternal In Step Paternal In Step is when a family member is promoted several steps above his or her level of incompetence. According to Peter and Hull (1969, 49-50), this could take place in two ways: an existing employee may be dismissed or removed by lateral arabesque or percussive sublimation, to make a place for the in stepper, or a new position with an impressive title is created for the in-stepper . These techniques may cause considerable ill-feeling towards the new appointee. 2.2.6 Promotion by Pull and Push Peter and Hull (1969) suggests two main means by which a person can affect promotion rate, Pull and Push. Pull is an employee's relationship - by blood, marriage or acquaintance with a person above the employee in the hierarchy. Push on the other hand is sometimes manifested by an abnormal interest in study, vocational training and self-improvement. In small hierarchies, Push may have a marginal effect in accelerating promotion; in larger hierarchies the effect is minimal. Pull, is therefore likely to be more effective than push. Peter & Hull (1969, p.63) states Never stand when you can sit; never walk when you can ride, never Push when you can Pull. 2.2.7 Final Placement Syndrome According to Peter and Hull (1969, p.108) when an employee reaches his level of incompetence, he can no longer do any useful work. This was termed as Final Placement Syndrome by Dr.Peter(1969). He further discusses some symptoms to identify employees who have reached this state. It is found out that such employees

15

are overly stressed, mentally disturbed and frequently sick (Peter and Hull, 1969, 109-115). Peter and Hull (1969, 116-127) lists some areas of behavior which identify those who have reached their highest level of incompetence. Abnormal Tabulology: This is an important area of hierarchiology. A competent employee normally keeps on his desk just the books, papers, and apparatus that he needs for his work. After final placement, an employee is likely to adopt some unusual and highly significant arrangement of his desk. Papyromania: This manifestation of final placement causes the employee to clutter his desk with piles of never used papers and books. Consciously or unconsciously, he thus tries to look busy and mask his incompetence by giving the impression that he has too much to do than any human could accomplish. Fileophilia: Here we see a mania for the precise arrangement and classification of papers, usually combined with a morbid fear of losing any document. By keeping himself busy rearranging and re-examining bygone business, the fileophiliac prevents other people-or himself-from realizing that he is accomplishing little or nothing of current importance. Self-pity: One excellent indication of final placement is the telling of chronic hardluck stories. It is always the fault of someone outside and beyond the pitiers control that makes them incompetent. This self-pity is usually combined with a strong tendency to reminisce about "the good old days," when the complainant was working at a lower rank, a level of competence. Cachinatory Inertia: The habit of telling jokes instead of getting on with business. Side-Issue Specialization: a commonplace substitute for competence characterized by the motto: "Look after the molehills and the mountains will look after themselves."

16

Substitution: Once an employee has reached his level of incompetence, he must engage in one or more substitutions to keep sane and happy. Otherwise he would have to face the Sordid Truth, that he is unfit and incompetent to do the job. Buck passing: Passing the buck is another symptom. It is the act of attributing another person or group with responsibility for one's own actions 2.2.8 Peter's Corollary Peter's Corollary states that "in time, every post tends to be occupied by an employee who is incompetent to carry out their duties" and adds that "work is accomplished by those employees who have not yet reached their level of incompetence"(Peter and Hull 1969,p.27). "Managing upward" is the concept of a subordinate finding ways to subtly manage superiors in order to limit the damage that they end up doing. Peter (1985, p. 28) wrote in Why Things Go Wrong or the Peter Principle Revisited, that: I named it The Peter Principle, because it described a generalization or a tendency and not something inevitable. The typical organizational systems encourage individuals to climb to their levels of incompetence. If you are able to do your job efficiently and with ease, you will be told that you need to take up more challenges and you will be moved up. Nevertheless the problem is that when you find something you cant do very well, that is where you stay, inept in the job, frustrating your co-workers, and harming the effectiveness of the organization. However Peter (1972, p. 35) states that he first wrote about the Peter Principle, he assumed it applied to all or at least most professions, but he could not be certain. Although it was impossible for him to study every organization that existed in the world, the ones he investigated conformed to the principle. Hence further research is needed to identify if the Peter Principle is prevalent in the IT industry. As IT is a fairly new industry, minimal research has been done to identify if the Peter Principle exists in its context.

17

2.3 Peter Principle in Software Development Firms A software company is made up of employees who either follow the management track or the technical track. Hence the Peter Principle could affect a software company mainly in two different ways. The first scenario is when a technical person gets promoted to a managerial position. For e.g. programmers make up the majority of the entry level roles in software companies. These people have highly developed technical skills. A successful programmer will usually, after an appropriate time, be offered a promotion. That's what tends to happen in a hierarchy, and the prospect of a career path is what attracts many people to entry-level jobs in these organizations. Once he becomes a Senior Engineer, he'll be dealing with more challenging projects. But he'll still be using the same skills, just at a more advanced level. Then, the time comes for him to be promoted again, and this time he will be made a manager, who is in charge of a team of programmers. This is where things can start to go wrong. While his knowledge of the company, its products and its clients mean that he's well placed to be managing a department, he may not have any of the soft skills needed to handle people, or liaise with other teams and senior management. His technical expertise is no longer useful to him. As a result, his performance in this new role may be poor. If he can't improve his soft skills, he'll never be promoted again. But because people are rarely demoted in a hierarchy, he'll remain at that level his level of "incompetence" doing a bad job. Not only will this make him unhappy, but the organization will suffer too. Taken to its extreme, many of the roles in the upper part of a hierarchical organization may be occupied by people who are not particularly good at their jobs. Cline, Lomow and Girou(1998,p.301) states that when exit interviews are reviewed for technical workers, two troubling facts are noted: 1. Technical workers with the highest appraisal scores tend to leave in the largest numbers. 2. The most common reason cited for leaving a company is I dont like working for bad management.

18

The second scenario is when incompetency occurs in technical positions. For e.g. there may be employees with average technical skills. Once these employees are promoted to a higher level, say from software engineer to senior software engineer, they may lack the necessary competence to cope with the new responsibilities assigned to them. For e.g. a software engineer may not be required to do software designs, but once he becomes a senior software engineer, his new responsibilities may require him to do software designing as well. At this point, if the employee is not good at the task, he may end up designing software that are of very low quality. This may lead to the Software Peter Principle. According to Cline, Lomow and Girou(1998,p.47) ,the Software Peter Principle is in operation when unwise developers improve and generalize the software until they themselves can no longer understand it, then the project slowly dies. The Software Peter Principle can ruin projects. The insidious thing about it is that it's a silent killer. By the time the symptoms are visible, the problem will have spread throughout every line of code in the project. This principle has been derived from the Peter Principle. There may be various other instances where incompetence of employees can occur. But in this research we will focus more on the above mentioned two areas of incompetence.

2.4 Researches Carried on Peter Principle Effect A number of researches has been carried out to identify the causes of Peter Principle Effect and to find a solution to the problem. They are listed down below chronologically, to highlight development of the thinking of various writers about this controversial topic from 1969 to date. Where and by whom, the, know-how of an employee is determined? According to Peter (1969), employee competence is determined not by outsiders, but by the employees superior in the hierarchy. If the superior is still at a level of competence, that person may evaluate subordinates in terms of the performance of useful work i.e. the evaluation of actual output. On the other hand, if the superior has reached a level of incompetence, that person will probably rate subordinates, in terms of institutional

19

values i.e. the superior will see competence as the actions that support the rules, rituals, and forms of the organization as they are, as opposed to how they should be. Peter (1969) concluded from his studies that what appear to be exceptions are not exceptions at all, because even though employees want to be productive, the Peter Principle still applies to all employees in, all hierarchies. Minter (1972) mentioned, as part of the Peter Principle in Training, that individuals who have been in charge of planning and developing training programs have historically had little or no formal training to prepare them for such a position. Thus, they usually lack training in educational principles, psychology of learning, communication and instruction, and in methods of testing and evaluation. As a result, individuals who have to assume responsibility for planning and training, often learn by trial-and-error at the expense of both the trainees and the organization. The author of CPAs Meet the Peter Principle (1988) stated and seemed to support Peter (1969) in that: Everything Dr. Peter predicted in The Peter Principle is coming home to roost in the field where stability is such a virtue, that nobody ever thought it would happen. Employees, who are continually promoted because the next slot is vacant, not necessarily because they are qualified, will eventually be promoted to their levels of incompetence. Koontz and Weihrich (1990, p.236) pointed out that errors in the selection process can lead to actualization of Peter Principle. Odiorne (1991) pointed out, even though not mentioning anything about the Peter Principle per se, that people have more talent and intelligence than we often assume. This researcher also said that employees should be taught the skills and tasks in order to be knowledgeable, because ongoing training can prevent competence from eroding and becoming obsolete. Gately (1996), found that employees can avoid the Peter Principle as long as employees are judged on technical merit and accomplishment, and that promotions go to the technically proficient and verbally expressive employees, while the less technically proficient and verbally expressive wait their turn.

20

Anderson, Dubinsky, and Mehta (1999) wrote that, sales performance is determined by how well the sales manager can motivate, lead, and control sales-force operations. But, whether viewed from the perspective of salespeople, customers, sales managers themselves, or top management, there is concern that sales organizations are not performing as desired. Their findings support the contention that, sales managers may well be marketings, best example of the Peter Principle: They too have arrived at their level of organizational incompetence. Faria (2000) mentioned concern about the role of promotions to manager and its impacts on the firm behavior, assuming an internal labor market structure. Faria further stated that a shortcoming of this process is that, people can be placed in important jobs for which they are ill-qualified. That is, the Peter Principle can be an outcome of this process, where people are promoted out of jobs for which they are overqualified until they reach ones, where the job demands are suited to maximum of individual ability levels. Namely, they are on the edge of their competence, so they cannot achieve anything more than what they had already achieved. Fairburn and Malcolmson (2000) have put forward as a basis of argument, that if a firm provides incentives by promoting those who have performed well in a given job, it may simply transfer them to another job to which they are not well suitedthat is a mild version of the Peter Principle. Lazear (2004, p.159), in his theoretical model, which was a review of Peters work, concluded the following from direct Peter Principle research: Workers who are promoted, receive this treatment because they are observed to have exceeded some standard. Part of the observation is based on lasting ability, but part is based on transitory components that may reflect measurement difficulties, short-term luck, or skills that are job specific. As a result, there is regression to the mean, creating a Peter Principle effect.Workers who are promoted do not appear to be as able, as they were before the promotion. Lazear (2004, p. 159) further deduced the following: Firms take this phenomenon into account in setting up their promotion rule. Under general conditions, when fewer than 50% of the workers are better suited to the high level job, the firm adjusts the

21

promotions standard upward to compensate for the regression to the mean. The amount of the adjustment depends on the tightness of the error distribution. When the pre-promotion error has high dispersion, promotion standards are inflated by more than they are when the error dispersion is low. And finally, Lazear (2004) deduced that all workers who remain at a given level will be incompetent in that they are neither as good as the average worker coming into the job nor are they as good as they were in their previous job relative to their comparison set. Lazear (2004, p. 148) also stated that: The regression-to-the-mean phenomenon implies that movie sequels are lower quality than the original films on which they are based and that excellent restaurant meals are followed by ones that are closer to mediocre King (2004), in the same year, speculated that persons in bureaucratic institutions are promoted until they reach the level of their incompetence and remain there until they quit, retire, or rarely, are fired. Furthermore, King stated that this phenomenon does not occur only in governmental institutions. In many publicly held companies in corporate America, the exercise of less oversight than is exerted in governmental agencies lends itself to layers of bureaucracy and incompetence. To a lesser degree, small businesses are also plagued by this. Fetzer (2006) mentioned that as people climb up the organizational ladder, they reach a level within the organization in which, they cannot perform competently, which leads to a dead-wood supervisor/manager/executive whose position and its duties are too much for this person to handle well. In 2007, Chapman affirmed that: For every job in the world, there is someone who cannot do it. Newman (2008) cited an affable but invasive regional manager (i.e., M. Scott) as the type of person who rises just above his abilities and then plateaus at a level of incompetence. A. Donovan, professor of business ethics at Dartmouths Tuck Business School, posited that: Ninety percent of the population deals with the M. Scotts in their daily lives (Newman, 2008, p. 6).

22

According to Schapp & Ogulink(2009, p.2), in their research, Peter Principle, the phenomenon in which employees around the world, are said to rise to their level of incompetence, is still prevalent today and that little regarding its use has changed since 1969. Seventy-three percent of the participants in their study said that, they have seen a Peter Principle situation happen within the last five years. Pluchino, Rapisarda and Garofalo(2009) stated that, The Peter Principle would realistically act in any organization where the mechanism of promotion, rewards the best members and, where the mechanism at their new level in the hierarchical structure, does not depend on the competence they had at the previous level, usually because the tasks of the levels are very different to each other. They show, by means of agent based simulations, that if the latter two features actually hold in a given model of an organization with a hierarchical structure, then not only is the Peter principle unavoidable, but also it yields in turn a significant reduction of the global efficiency of the organization. Within a game theory-like approach, they explored different promotion strategies and found, that in order to avoid such an effect the best ways for improving the efficiency of a given organization are, either to promote each time an agent at random or to promote randomly the best and the worst members in terms of competence. Thus it appears, at least according to the literature review performed in this study that the Peter Principle is still thriving. It is evident that not much has truly changed in the many years since Dr. Peters study in 1969. 2.5 Summary This chapter introduces the Peter Principle and goes on to explain its various characteristics which helps us to identify whether the effect exists in an organization. It then discusses the possibility of the existence of this effect in software development organizations. A software company is made up of employees who either follow the management track or the technical track. The Peter Principle can be present in two different scenarios. The first scenario is when a technical person gets promoted to a managerial position and the second scenario is when incompetency occurs in the technical positions itself. Both of the above should be avoided in a

23

software company for it to succeed. The chapter concludes with a discussion on studies carried out by previous researchers on the Peter Principle effect.

24

CHAPTER 3 - METHODOLOGY

3.1 Introduction This chapter discusses the methodology adopted for the study in detail to address the research problem identified in Chapter 1. The research study carried out was empirical in nature. An empirical study needs to be supported by theories so that hypotheses can be generated and a basis can be given for interpreting and summarizing the research results. Based on the review of the literature in Chapter 2, this chapter describes the development of the conceptual model and the hypotheses that guide the rest of the study and presents the methodology used in the study, specifically in relation to the research design and the data collection process. 3.2 Conceptual Framework of the Study After formulating the theoretical framework, the researcher has to develop the conceptual framework of the study. A conceptual framework is described as a set of broad ideas and principles taken from relevant fields of enquiry and used to structure a subsequent presentation (Reichel & Ramey, 1987). It has potential usefulness as a tool to scaffold research and, therefore, to assist a researcher to make meaning of subsequent findings. Such a framework should be intended as a starting point for reflection about the research and its context. A theoretical framework or the theory on which the study is based was identified in Chapter 2. The conceptual framework introduced in this section will be the operationalization of the identified theory. The following Figure 3.1 indicates the key concepts associated with the study and their relationships with each other.

25

Company Size

Company Structure

Quality of Performance and Rewards Management Practices

Existence of
Quality of Selection and Recruitment Practices

Peter Prinicple related Behavior Patterns

Quality of Human Resource Development Practices

Figure 3.1: Conceptual Model

According to the information gathered during the literature review, the above relationships were derived from various researches carried out in the past. The theory of Peter Principle and the criteria to identify its existence was introduced by Peter and Hull (1969). The existence of a relationship between performance and rewards management practices and Peter Principle was empirically studied by Gately (1996); Faria (2000); Fairburn and Malcolmson (2000); Lazear (2004); King (2004); Fetzer (2006); Chapman (2007) and Pluchino,Rapisarda and Garofalo(2009). The existence of a relationship between selection and recruitment practices, was described in the works of Koontz and Weihrich (1990). The relationship between human resource development and Peter Principle was empirically studied by Minter (1972) and Odiorne (1991). The relationship between the company structure and Peter Principle was empirically studied by Peter and Hull (1969).

26

3.3 Research Hypotheses The hypothesis would indicate the predetermined relationship between the key variables of the study and this would be based on the literature review ideas and the understanding of the researcher on the considered subject area (Malhotra, 2007). Thus, formulating the hypothesis would ensure that the study remains focused, and on track as the researcher would have to seek to prove whether the hypothesis is correct or incorrect. Formulating the hypothesis indicates that the researcher has a preconceived idea and he would gather data and analyze them with the view of either proving or disproving the hypothesis established. The following Table 3.1 indicates the hypotheses which the study is seeking to prove. Table 3.1: Hypothesis statements Hypothesi s H1
Description There is a significant relationship between the size of the company and the existence of the Peter Principle effect, where larger firms suffer more from this effect. There is a significant relationship between the structure of the company and the existence of the Peter Principle effect, where firms with tall structures suffer more from this effect. The quality of performance and rewards management practices in a company significantly, negatively influence the existence of Peter Principle effect The quality of recruitment practices in a company significantly, negatively influence the existence of Peter Principle effect The quality of human resource development practices in a company significantly, negatively influence the existence of Peter Principle effect

H2

H3 H4 H5

3.4 Research Model with Hypotheses Following is the research model (Figure 3.2) with the introduction of the above mentioned hypotheses. The graphical representation of the proposed framework depicts the pattern and structure of relationships among the set of measured variables. The purpose of the study was to measure relationships among these variables. This research intended to investigate the existence of the Peter Principle

27

effect in Software Development firms of Sri Lanka, and the relationships between the existence of the Peter Principle effect and company size, company structure, quality of performance and rewards management practices, selection and recruitment practices and human resource development practices.

Company Size

H1

Company Structure

H2

Quality of Performance and Rewards Management Practices

H3

Existence of
H4

Quality of Selection and Recruitment Practices

Peter Prinicple related Behavior Patterns

Quality of Human Resource Development Practices

H5

Figure 3.2: Conceptual Model with Hypotheses

In the investigation, the existence of the Peter Principle effect was taken as the dependent variable and, company size, company structure, quality of performance and rewards management practices, quality of selection and recruitment practices and quality of human resource development practices were taken as independent variables. This research used a regression study to establish the existence of relationships between the measured variables. As mentioned earlier the researchers intention was to identify whether any relationships exists between these measured

28

variables. Regression study provides a measure of degree between two or more variables. Therefore, the study was designed as a regression study.

3.5 Operationalization of Variables This section intends to discuss the operationalization of the constructs identified in the research model proposed for this study under section 3.4. These constructs must be operational so as to enable the researcher to measure them. To do so, the abstract notions of the constructs must be reduced into observable behavior or characteristics (Sekaran, 2008). Operational definitions provide meaning to the constructs and a tangible way to measure them. In addition, constructs in the study uses multi items measures and a five point Likert type scale. The constructs were adapted from the literature review carried out. The following sections describe the definitions and item measures of the constructs. 3.5.1 The size of the company As mentioned in Chapter 1, software development firms of Sri Lanka can be categorized into three groups by size; small, medium and large. In general the size of a business has been defined based on the number of employees. This differs from country to country. More than 500, is generally considered to be a large business. According to Campbell, (2007) some software organizations, such as Microsoft, consider a small business as being up to 50 employees. Others consider anything under 100 employees as a small business, and some consider anything under 500 a small business. With respect to Sri Lankan software development companies, the author defined less than 100 employees as small, between 100-300 employees as medium and more than 300 employees as large size. The intention of the author was to identify if there is a significant relationship between the size of a company and the existence of the Peter Principle effect. 3.5.2 The structure of the company

29

Software development firms in Sri Lanka can be categorized as having tall or flat hierarchies in a broad sense. According to Peter and Hull (1969), the Peter Principle effect is more prevalent in tall hierarchies. The intention of the author was to identify if there is a significant relationship between the structure of a company and the existence of the Peter Principle effect. 3.5.3 Quality of performance and rewards management practices Performance and rewards management practices are methods by which the job performance of an employee is evaluated. It is the process of obtaining, analyzing, and recording information about the relative worth of an employee to the organization. It will analyze an employee's recent successes and failures, personal strengths and weaknesses, and suitability for promotion or further training. The intention of the author was to identify if there is a significant relationship between the quality of performance and rewards management practices of a company and the existence of the Peter Principle effect. 3.5.4 Quality of Selection and recruitment practices Selection and recruitment of employees is an important aspect for any company. It cannot be faulted as the success of any firm depends on the quality of human resources or talents in that firm. Therefore it is very important for any company to be very sure of hiring the right staff without compromising anything from the onset. The intention of the author was to identify if there is a significant relationship between the quality of selection and recruitment practices of a company and the existence of the Peter Principle effect. 3.5.5 Human resource development practices Human Resource Development is the integrated use of training, organization, and career development efforts to improve individual, group and organizational effectiveness. It develops the key competencies that enable individuals in organizations to perform current and future jobs through planned learning activities. The intention of the author was to identify if there is a significant relationship

30

between the human resource development practices of a company and the existence of the Peter Principle effect. The following table 3.2 depicts the operationalization of the identified variables.

31

Table 3.2 : Operationalization of the identified variables

Objectives To identify if the Peter Principle effect exists in Software development firms of Sri Lanka Variables Peter Principle related behavior patterns Indicators Percussive Sublimation Sub Indicators Incompetent employees promoted to reduce the harm done by them Isolating incompetent employees to reduce the harm done by them Competent employees are fired to preserve the hierarchy Existing process consistency is more valued than efficient service Employees are placed high up in the hierarchy based on personal relationships Followers of Incompetent superiors get promoted easily Source-from (Peter& Hull,1996) Level of Measurement Likert Question Number 9

Lateral Arabesque

(Peter& Hull,1996)

Likert

Hierarchical Exfoliation Peters Inversion

(Peter& Hull,1996)

Likert

12

(Peter& Hull,1996)

Likert

11

Paternal In Step

(Peter& Hull,1996)

Likert

13

Pull & Promotion

(Peter& Hull,1996)

Likert

14

32

Frustration of incompetent employees Buck-Passing Substitution of work Final Placement Syndrome Work is accomplished by those employees who have not yet reached their level of incompetence The presence of incompetent employees

Frustration due to incompetence Shifting of responsibility or blame to anothers Substitute work to competent employees Employees who have achieved their highest level of incompetence The number of competent employees in the hierarchy and their level of productivity The number of incompetent employees in the hierarchy and their level of productivity The turnover rate of competent employees Competent employees have to push hard for promotions

(Peter& Hull,1996) (Peter& Hull,1996) (Peter& Hull,1996) (Peter& Hull,1996)

Likert Likert Likert Likert

19 16 17 7,8,18

(Peter& Hull,1996)

Likert

21,23

Rate of Incompetence at higher levels

(Peter& Hull,1996)

Likert

20,22

Dissatisfaction of competent employees

Frustration of competent employees Push & Promotion

Author Developed (Peter& Hull,1996)

Likert Likert

10 15,24,25,26

33

To identify and analyze the impact of the determining factors that associate with the existence of Peter Principle effect

Size of the Organization

Company Structure

Performance & Rewards Management Practices

Selection & Recruitment Practices

Size of the organization has an impact or not on the existence of Peter Principle Structure of the organization has an impact or not on the existence of Peter Principle Having proper performance evaluation methods in place Satisfaction of the employees on the performance evaluation method used Clearly defined job responsibilities Rewards based on performance ratings Proper job descriptions for job vacancies Well Structured Interviews that tests all aspects to select the best candidate Competent Interviewers

Author Developed

Likert

Author Developed

Likert

(Drucker,1993)

Likert

27

Author Developed

Likert

28

Author Developed Author Developed (Yate,1997) (Yate,1997)

Likert Likert Likert Likert

29,30,31 32 33 34,36

Author Developed

Likert

35

34

HR Development & Talent Management Practices

Demographics

Age Gender Marital Status Educational Qualifications Designation Experience in IT Industry Experience in current company Experience in current position

Good Training and Development Programs Employee Satisfaction towards training programs Competent Trainers -

(Berger & Berger , 2003) (Berger & Berger , 2003) Author Developed Author Developed Author Developed Author Developed Author Developed Author Developed Author Developed Author Developed Author Developed

Likert Likert

37,38,39 41

Likert Nominal Nominal Categorical Categorical Categorical Nominal Nominal Nominal

40 42 43 44 45 2 1 3 4

35

3.6 Unit of Analysis As the study intended to investigate the effect of Peter Principle in the Software Development firms of Sri Lanka, the study would cover a stratified selected sample from the large, medium and small software development firms in the country. 3.6.1 Target Population According to Sri Lanka Business Portal - Trade Information (2011), over 50,000 are employed in the IT and BPO industry in Colombo. Therefore a stratified sampling technique was used to select the sample. Sampling is the process of selecting a sufficient number of elements from a population to represent the properties or characteristics of that population (Sekaran, 2008, 226-227). A sample consisting of 381 employees was calculated as the ideal sample size for the above population. In determining the sample size, Sekaran (2008, p.294) provided a table that generalized scientific guideline for sample size decisions. According to the table, for a population size of 50,000, the appropriate sample size is 381. This was calculated using a confidence level of 95% and confidence interval of 5. The sample was selected from the companies listed below in Table 3.3. Table 3.3: Companies selected for the sample Company Size (Large : Over 300 Medium:100-300 Small : Less than 100) Large Large Large Medium Medium Medium Medium Medium Medium Medium Small Small Small Year of Establishment 1995 1997 1996 2001 2000 1999 1999 2000 2004 2001 2005 2003 2000

Virtusa IFS MIT Reservations Gateway Mubasher Aepona Creative Solutions Eurocenter Ecollege Aeturnum WSO2 B Sharp Lanka Interblocks

36

Sabre Technologies EDM Systems

Small Small

2001 1995

The selected sample mostly included employees who are in the lower levels of the hierarchy and in the middle level management positions such as technical leads, quality assurance leads, project managers, architects etc. Such employees were selected as there is a high possibility that they have not reached their level of incompetence, thus enabling the researcher to obtain more accurate responses. 3.6.2 Data collection instrument A research survey will be only as good as the questions it asks. Questionnaire design, therefore, is one of the most critical stages in the research process. According to Sekaran (2008), a good questionnaire design should focus upon three areas; the wording of the questions, the principle of measurements, the general appearance of the questionnaire. Taking the above into consideration a structured questionnaire was designed in order to collect the required data for the investigation. The questionnaire was designed only in English language as it was fair to assume that all employees in software development firms will have the general understanding of English language. The questionnaire was made up of two key areas: the core area and the demographics area. Under demographics, all demographic related information was collected. This information included the age, gender, duration of work and other personal information related to the respondent. Even though this information would not have a direct relevance to the research study, it was used to understand the demographic profile of the respondents and was used as cross analysis points where further analysis of the research data was required. The core area consisted of the questions, which had direct relevance to the key information areas the study attempted to cover. That is to identify the existence of Peter Principle effect, and the contribution of the factors listed in section 3.5, to its existence. All core area questions would take the form of Likert scale based questions where the respondents could indicate the levels of agreement using a five

37

point Likert scale. According to Sekaran(2008), the advantage of using the Likert scale is that the respondents would have the freedom to express their views using a range of alternatives and the response would be focused and easy to directly use for analysis purposes.

3.6.3 The method of data collection and analysis The questionnaire was distributed as a web based questionnaire. The researcher was able to collect a total of 396 valid responses from the survey carried out. The questionnaire used can be found under Appendix-A. The responses from the questionnaire were used to study whether the Peter Principle effect exists in Sri Lankan software development firms and to identify the contribution of the factors identified in the literature review to its existence. Proposing recommendations to avoid the Peter Principle, which was the final objective of this study, was based on the entire output of first and second objectives. The raw data set that was obtained from the sampled questionnaires was fed in to the Statistical Package for Social Sciences (SPSS version 15) software and various statistical analysis methods like, correlation analysis and regression analysis were carried out. 3.7. Summary This chapter discusses the methodology adopted for the study. It gives a comprehensive explanation on the conceptual model that was created and the hypotheses that the researcher intends to test. It then discusses the operationalization of the variables identified and the target population selected. It concludes with an explanation on the data collection and analysis method selected to carry out the study.

38

CHAPTER 4 DATA ANALYSIS AND DISCUSSION

4.1 Introduction

This chapter presents the data and a discussion of the findings. The findings presented under this section are derived from the data gathered from the research questionnaire. The study was exclusively limited to employees of the software development firms mentioned in chapter 3. First section of the chapter will highlight the general information and demographics of the studied sample. In the second section of the study reliability and the validity analysis of the identified constructs are discussed. In the next sections of this chapter, descriptive statistics, correlation analysis will be presented followed by the regression analysis among the constructs identified. 4.2 Characteristics of the Sample This section analyzes the general characteristics of the sample in consideration. It is presented in two sections as characteristics of the organizations and the respondents.
4.2.1 Characteristics of the organizations

The characteristics of the organizations in the sample are given below. In question 5 (Refer Appendix A) the respondents were requested to state the size of their respective organizations. The summary of the responses to the said question is as follows. From the total of 396 respondents, 32.8% were from companies had less than 100 employees, 31.6% were from companies that had employees between 100 and 300 and the remaining 35.6% were from companies that had over 300 employees. The Table 4.1 summarizes the data obtained.

39

Table 4.1: Summary table for number of employees in organizations

Cumulative Percent 32.8 64.4 100.0

Less than 100 Between 100 and 300 Greater than 300 Total

Frequency 130 125 141 396

Percent 32.8 31.6 35.6 100.0

Valid Percent 32.8 31.6 35.6 100.0

In question 6 (Refer Appendix A) the respondents were asked about the structure of their company. The summary of the responses to the said question is as follows. From the total of 396 respondents, 50.3% mentioned that their companys hierarchy is flat and the remaining 49.7% said it was tall. The table 4.2 summarizes the data obtained. Table 4.2: Summary table for number of employees in different organization structures
Cumulative Percent 50.3 100.0

Flat Tall Total

Frequency 199 197 396

Percent 50.3 49.7 100.0

Valid Percent 50.3 49.7 100.0

4.2.2

Characteristics of the respondents

This section presents the key demographic features of the respondents who participated in the survey. Age From the responses given to question 42 (Refer Appendix A), the age distribution of the respondents is as follows. From the 396 respondents 10.6% were in the age group below 25, 87.1% were in the age group between 26 and 35. Therefore the majority of the respondents were in their 20s and 30s. 2% were in the group between 36 and 45

40

and the remaining 0.3% were in the above 45 group. The table 4.3 summarizes the data obtained. Table 4.3: Summary table for the age distribution of the respondents
Frequency 42 345 8 1 396 Percent 10.6 87.1 2.0 0.3 100.0 Valid Percent 10.6 31.6 2.0 0.3 100.0 Cumulative Percent 10.6 97.7 99.7 100.0

Below 25 Between 25 and 35 Between 36 and 45 Above 45 Total

Gender Question 43 (Refer Appendix A), inquired about the gender of the respondents. From the 396 respondents 55.8% were male and 44.2% were female. The table 4.4 summarizes the data obtained.

Table 4.4: Summary table for the gender distribution of the respondents
Cumulative Percent 55.8 100.0

Male Female Total

Frequency 221 175 396

Percent 55.8 44.2 100.0

Valid Percent 55.8 44.2 100.0

Marital Status From the responses to the question 43 (Refer Appendix A),, the marital statuses of the respondents are as follows. From the 396 respondents, 77.8% were single and 22.2% were married. The table 4.5 summarizes the data obtained. Table 4.5: Summary table for the marital status of the respondents
Cumulative Percent 77.8 100.0

Single Married Total

Frequency 308 88 396

Percent 77.8 22.2 100.0

Valid Percent 77.8 22.2 100.0

41

Education Level Question 44(Refer Appendix A), inquired about the education level of the respondents. From the 396 respondents, 0.5% had advanced level or below qualifications. These could probably be trainees who have not yet completed their degrees. 87.6% had bachelors degrees or equivalent. The remaining 11.9% had postgraduate qualifications. The table 4.6 summarizes the data obtained. Table 4.6: Summary table for the education level of the respondents
Frequency 2 347 47 396 Percent .5 87.6 11.9 100.0 Valid Percent .5 87.6 11.9 100.0 Cumulative Percent .5 88.1 100.0

Advanced Level or below Bachelor's Degree or equivalent Postgraduate Qualifications Total

Current Designation Question 2 inquired about the current designation of the employees. From the responses given, 12.4% were in the Trainee Level (e.g. associate software engineers, associate QA engineers etc), 42.9% were in the intermediate level (e.g. software engineers, QA engineers, business analysts etc) 31.6% were in the senior level (e.g. senior software engineers, senior business analysts etc), 8.1% were in the lead level (e.g. tech leads, QA consultants etc) and the remaining 5.1% (e.g. project managers, architects etc) were in the managerial level. The designations were split into the above four broad categories as the main intention of the author was to find the level of the employees position in the hierarchy. The Table 4.7 summarizes the data obtained.

42

Table 4.7: Summary table for the designation level of the respondents
Cumulative Percent 12.4 55.3 86.9 94.9 100.0

Trainees/Associate Level Intermediate Level Senior Level Lead Level Managerial Level Total

Frequency 49 170 125 32 20 396

Percent 12.4 42.9 31.6 8.1 5.1 100.0

Valid Percent 12.4 42.9 31.6 8.1 5.1 100.0

Years of experience in IT The total years of experience of the respondent was inquired by question 1. The responses are as follows. 11.9% had below 2 years of experience in total, 65.9% were in the 2-5 years range, 19.7% were in the 5-10 years range and the remaining 2.5% were in the over 10 years range. The table 4.8 summarizes the data obtained. Table 4.8: Summary table for the years of experience in IT of the respondents
Cumulative Percent 11.9 77.8 97.5 100.0

Below 2 years Between 2-5 years Between 5 -10 years Above 10 years Total

Frequency 47 261 78 10 396

Percent 11.9 65.9 19.7 2.5 100.0

Valid Percent 11.9 65.9 19.7 2.5 100.0

Years of experience in the current company Question 3 inquired about the years of experience in the current company. The responses are as follows. 19.7% had below 2 years of experience, 69.9% were in the 2-5 years range, 9.1% were in the 5-10 years range and the remaining 1.3% were in the over 10 years range. The table 4.9 summarizes the data obtained.

43

Table 4.9: Summary table for the years of experience in the current company of the respondents
Frequency 78 277 36 5 396 Percent 19.7 69.9 9.1 1.3 100.0 Valid Percent 19.7 69.9 9.1 1.3 100.0 Cumulative Percent 19.7 89.6 98.7 100.0

Below 2 years Between 2-5 years Between 5 -10 years Above 10 years Total

Years of experience in the current post Question 4 inquired about the years of experience in the current post. The responses are as follows. 61.6% had below 2 years of experience, 36.8% were in the 2-5 years range, 1.3% were in the 5-10 years range and the remaining 0.3% were in the over 10 years range. The table 4.10 summarizes the data obtained. Table 4.10: Summary table for the years of experience in the current post of the respondents
Cumulative Percent 61.6 98.4 99.7 100.0

Below 2 years Between 2-5 years Between 5 -10 years Above 10 years Total

Frequency 244 146 5 1 396

Percent 61.6 36.8 1.3 0.3 100.0

Valid Percent 61.6 36.8 1.3 0.3 100.0

4.3 Goodness-of-Fit Measures Reliability analysis was carried out for all variables in the conceptual framework to test reliability and the consistency of data. The Cronbachs alpha indicates how well the items in a set are positively correlated to one another. The closer the reliability reaches 1.0, the better the reliability and validity. Generally, reliabilities less than 0.6 are considered poor. Those in the range of 0.7 are acceptable and those over 0.8 good (Sekaran 2008, p311). Principle component analysis was used to identify the underlying components that explain the pattern of correlations within a set of observed variables. It is used in data

44

reduction to identify a small number of factors that explain most of the variance observed in a much larger number of variables. The results of the tests carried out and the decisions taken will be discussed in the following sections.

4.3.1 Reliability and validity analysis - the existence of Peter Principle effect

The results of the reliability analysis carried out on the factors that indicate the existence of Peter Principle effect are as follows. Refer Appendix A for the descriptions of the questions. Table 4.11: Reliability analysis for the factors that determine the existence of Peter Principle effect
Concept/Variable Peter Principle Behaviors Questions Retained Q7,Q8,Q9, Q10,Q11,Q12,Q13,Q14, Q15,Q16,Q17,Q18,Q19,Q20,Q22 Questions Rejected Q21,Q23,Q24,Q 25,Q26 Cronbachs Alpha Value 0.96

Q21,Q23,Q24,Q25 and Q26 were very weakly correlated with the rest of the variables. Hence they were removed .The Cronbach alpha value increased to 0.96 after the removal. (Refer appendix B) Principle component analysis was carried out to identify the underlying components (Refer appendix B). The results obtained from the analysis is shown in table 4.12:

45

Table 4.12: Principle component analysis for the factors that determine the existence of Peter Principle effect
Component Initial Eigenvalues % of Cumulative Total Variance % 9.432 62.883 62.883 1.321 0.759 0.589 0.485 0.480 0.390 0.315 0.274 0.247 0.210 0.183 0.136 0.122 0.055 8.805 5.063 3.929 3.235 3.198 2.601 2.100 1.825 1.647 1.399 1.221 0.909 0.816 0.368 71.687 76.750 80.679 83.914 87.113 89.714 91.815 93.640 95.288 96.686 97.907 98.816 99.632 100.000 Extraction Sums of Squared Loadings % of Cumulative Total Variance % 9.432 62.883 62.883 1.321 8.805 71.687 Rotation Sums of Squared Loadings % of Cumulative Total Variance % 6.550 43.668 43.668 4.203 28.019 71.687

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

The analysis showed that there were two components with high factor loadings. Those two components together explained 71.7% of the variance. The factor loadings on each of the components are as follows. Table 4.13: Rotated component matrix for the factors that determine the existence of Peter Principle effect
Component Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22 0.546 0.831 0.835 0.839 0.731 0.900 0.880 1 0.704 0.689 0.581 0.781 0.745 0.855 0.630 0.430 2 0.471 0.885 0.436 0.445

46

Thus the variables were split into two groups based on the factor loadings above. Existence of Peter Principle related behavior patterns Q7,Q9,Q10,Q11,Q12,Q13,Q15,Q16,Q17,Q18 and Q19 loaded heavily on component one. They mainly explained the existence of Peter Principle related behavior patterns. Existence of incompetent employees in the higher levels of the company hierarchy Q8, Q14, Q20, Q22 loaded highly on component two. These variables explained the existence of incompetent employees in the higher levels of software development firms.

Considering the component breakdown above there was a need to adjust the initial conceptual framework accordingly. The revised conceptual framework is shown in Figure 4.1:

Company Size

H1

Company Structure

H2

Quality of Performance and Rewards Management Practices

H3

Existence of Peter Principle related behavior patterns

Quality of Selection and Recruitment Practices

H4

Existence of incompetent Employees in the higher levels of the company hierarchy

Quality of Human Resource Development Practices

H5

47

Figure 4.1 Revised conceptual model

4.3.2 Reliability and validity analysis - the contributing factors for the

existence of Peter Principle effect The results of the reliability analysis carried out on the contributing factors for the existence of Peter Principle effect are as follows. a) Quality of performance and rewards management practices Table 4.14: Reliability analysis results for quality of performance and rewards management practices related variables
Concept/Variable Evaluation and Rewards Management Practices Questions Retained Q27, Q28, Q29 , Q30, Q31, Q32 Questions Rejected Cronbachs Alpha Value 0.91

All the variables in concern were highly correlated to each other. Therefore none of them were rejected. The Cronbach alpha value obtained for the set of variables that explained the quality of performance and rewards management practices in companies was 0.91, which is a very high value and it shows that the variables considered are highly reliable. (Refer appendix B for more details on the analysis) The principle component analysis gave the following results. Table 4.15: Principle component analysis results for quality of performance and rewards management practices related variables
Initial Eigen values Component 1 2 3 4 5 6 Total 4.209 0.680 0.365 0.301 0.273 0.171 % of Variance 70.144 11.331 6.090 5.024 4.557 2.853 Cumulative % 70.144 81.475 87.565 92.589 97.147 100.000 Extraction Sums of Squared Loadings Total 4.209 % of Variance 70.144 Cumulative % 70.144

48

49

The analysis showed that there was one component with high factor loadings. It explained 70.1% of the variance. The factor loadings are as follows: Table 4.16: Factor loadings for quality of performance and rewards management practices related variables
Component 1 Q27 Q28 Q29 Q30 Q31 Q32 0.832 0.829 0.847 0.871 0.819 0.825

All the factors loaded well on the component. Hence none of the factors were rejected. b) Quality of recruitment management practices

Table 4.17: Reliability analysis results for quality of recruitment management practices related variables
Concept/Variable Evaluation and Rewards Management Practices Questions Retained Q33, Q34, Q35 , Q36 Questions Rejected Cronbachs Alpha Value 0.88

All the variables in concern were highly correlated to each other. Therefore none of them were rejected. The Cronbach alpha value obtained for the set of variables that explained the quality of recruitment practices in companies was 0.88, which is a very high value and it shows that the variables considered are highly reliable.

50

The principle component analysis gave the following results. Table 4.18: Principle component analysis results for quality of recruitment management practices related variables
Component 1 2 3 4 Total 2.943 .458 .363 .236 Initial Eigen values % of Variance Cumulative % 73.574 73.574 11.439 85.014 9.085 94.098 5.902 100.000 Extraction Sums of Squared Loadings Total % of Variance Cumulative % 2.943 73.574 73.574

The analysis showed that there was one component with high factor loadings. It explained 73.6% of the variance. The factor loadings are as follows.

Table 4.19: Factor loadings for quality of recruitment management practices related variables

Component 1 Q33 Q34 Q35 Q36 0.820 0.852 0.889 0.868

All the factors loaded well on the component. Hence none of the factors were rejected. c) Quality of human resource development practices Table 4.20: Reliability analysis results for quality of human resource development practices related variables
Concept/Variable Evaluation and Rewards Management Practices Questions Retained Q37, Q38, Q39 , Q40, Q41 Questions Rejected Cronbachs Alpha Value 0.92

All the variables in concern were highly correlated to each other. Therefore none of them were rejected. The Cronbach alpha value obtained for the set of variables that explained the quality of human resource development practices in companies was

51

0.92, which is a very high value and it shows that the variables considered are highly reliable. Principle component analysis results for the above sets of variables are as follows. Table 4.21: Principle component analysis results for quality of human resource development practices related variables
Initial Eigen values Component 1 2 3 4 5 Total 3.839 .468 .313 .241 .138 % of Variance 76.789 9.363 6.256 4.829 2.763 Cumulative % 76.789 86.152 92.407 97.237 100.000 Extraction Sums of Squared Loadings Total 3.839 % of Variance 76.789 Cumulative % 76.789

The analysis showed that there was one component with high factor loadings. It explained 76.8% of the variance. The factor loadings are as follows. Table 4.22: Factor loadings for quality of human resource development practices related variables

Q37 Q38 Q39 Q40 Q41

Component 1 0.887 0.852 0.837 0.885 0.918

All the factors loaded well on the component. Hence none of the factors were rejected.

4.4 Descriptive Statistics

Measure of central tendency and dispersions provide a way to have a feel for the collected data set. Descriptive statistics such as maximum, minimum, means, standard deviations and variance were obtained for the independent variables that described the Peter Principle effect and the factors that contributed to its existence.

52

The values mean and standard deviation values obtained are shown in the below tables. (Refer Appendix B for additional information). It should be noted that all variables were measured on five point Likert scale ranging from 1-Very Less to 5Very High and the total numbers of subjects in the selected sample was 396. The descriptive statistics obtained for the factors that explain the existence of the Peter Principle effect are as follows. Table 4.23: Descriptive statistics for the factors that explain the existence of Peter Principle effect
Question Existence of previously competent employees who have become incompetent after obtaining promotions (Final placement syndrome) Existence of incompetent employees who could not obtain further promotions (Final placement syndrome) Existence of employees who are promoted to minimize the harm done by them in the current position(Percussive sublimation) The rate of competent employees leaving the organization Discouraging innovative ideas and processes to preserve the organizations existing practices(Peters inversion) Highly competent employees being fired to preserve the companys hierarchy (Hierarchical Exfoliation) Promotions given based on personal relationships(Paternal In Step) Followers of superiors getting promoted regardless of their competency level (Pull and Promotion) Competent employees have to push hard to get promoted (Push and Promotion) Existence of employees who shift the blame on to others for their actions(Buck Passing) Existence of incompetent superiors who substitute work to the subordinates under them Existence of employees who ignore their duties and focus more on activities that are not direct responsibilities of their position(Final placement syndrome) Existence of employees who are overly stressed, mentally disturbed and frequently sick even though they do not do a lot of productive work. (Final placement syndrome) The number of incompetent employees in the higher levels of organizations The productivity of the employees in the higher levels of the organization hierarchies Mean 3.17 3.13 2.76 3.55 3.15 2.61 3.16 3.16 3.09 3.30 3.38 3.17 Std. Deviation 1.406 0.983 1.221 1.269 1.396 1.269 1.496 1.350 1.038 1.244 1.284 1.311

2.96 3.07 3.09

1.052 1.000 0.957

53

The mean values obtained above are in the range of 2.61 3.55. This implies that the Peter Principle moderately exists in the software development firms of Sri Lanka. Hierarchal exfoliation and percussive sublimation has a lesser value compared to the other factors but it doesnt imply that those behaviors are nonexistent. The lesser value maybe due to the fact that, it is not ethical to carry out the above activities or maybe because the employees who come under those categories resign prior to the occurrence of the above. The descriptive statistics obtained for the factors that explain the perspective of the employees of the quality of the performance and rewards management practices in their respective organizations are as follows. Table 4.24: Descriptive statistics for the factors that explain the quality of performance and rewards management practices
Question The organization has a standard way of evaluating employee performance Employee satisfaction with the current performance evaluation methods in the organization The organization maintains proper job descriptions for all positions The organization clearly communicates job responsibilities to the employees The work performed by the employees match their job descriptions The organization provides rewards/promotions solely based on performance ratings Mean 3.02 2.75 2.88 3.06 3.03 2.97 Std. Deviation 1.058 0.950 1.051 1.062 1.027 1.030

The mean values obtained above are in the range of 2.75 3.06. This implies that the quality of performance evaluation and rewards management practices in organizations are in a moderate level and improvement is needed for the organizations to reach their maximum potential. The descriptive statistics obtained for the factors that explain the perspective of the employees of the quality of recruitment and selection practices in their respective organizations are as follows.

54

Table 4.25: Descriptive statistics for the factors that explain the quality of selection and recruitment practices
Mean The organization provides comprehensive job descriptions when advertising jobs Selection interviews are well structured to identify the best candidates Members of the selection panel are competent and properly trained The organization gives due attention to skills like teamwork, leadership, attitude etc when selecting a new employee 3.07 3.02 3.01 3.10 Std. Deviation 0.966 1.035 1.010 1.018

The mean values obtained above are in the range of 3.01 3.10. This implies that the quality of selection and recruitment practices in organizations are in a moderate level and improvement is needed for the organizations to reach their maximum potential. The descriptive statistics obtained for the factors that explain the perspective of the employees of the quality of human resource development practices in their respective organizations are as follows. Table 4.26: Descriptive statistics for the factors that explain the quality of human resource development practices
Question The organization provides job related trainings for all employees on technical matters The organization provides training on soft skills The organization has a well defined HR development /talent management process to achieve organizational goals The trainers in the organization are highly competent The employees are satisfied with the training they receive Mean 3.01 2.86 2.72 2.96 2.89 Std. Deviation 1.119 1.090 1.054 1.058 1.016

The mean values obtained above are in the range of 2.72 3.01. This implies that the quality of human resource development practices in organizations is in a moderate level and improvement is needed for the organizations to reach their maximum potential.

55

4.5 Correlation Analysis Correlation analysis was performed on the two dependent and five independent variables identified during the conceptual model. The two dependent variables are The existence of Peter Principle related behavior patterns (AvPPB) The existence of incompetent employees in the higher levels of the company hierarchy (AvIEHL)

The five independent variables are: The size of the company(Q5) The structure of the company(Q6) The quality of performance and rewards management practices in a company(AvPRM) The quality of recruitment practices in a company(AvRP) The quality of human resource development practices in a company(AvHRD)

The correlation analysis results for the above are shown in the below table followed by the interpretation of the results.

56

Table 4.27: Correlation analysis results

AvPRM AvPRM 1 AvRP 0.811** AvHRD 0.794** AvPPB -0.721** AvIEHL -0.495** Q5 0.059 Q6 -0.436** -0.453** -0.332** 0.564** 0.416** 0.077 1 0.066 0.147** -0.019 -0.048 1 0.077 -0.508** -0.381** 0.719** 1 -0.048 0.416** -0.691** -0.587** 1 0.719** -0.019 0.564** 0.739** 1 -0.587** -0.381** 0.147** -0.332** 1 0.739** -0.691** -0.508** 0.066 -0.453** 0.811** 0.794** -0.721** -0.495** 0.059 -0.436** AvRP AvHRD AvPPB AvIEHL Q5 Q6

According to the results shown in table 4.26, the following were identified:

The significant value of the correlation between the size of a company(Q5) and the existence of Peter Principle related behavior patterns(AvPPB) is 0.711, which is greater than 0.05. Hence we can conclude that there is no significant correlation between the two factors in consideration.

The significant value of the correlation between the size of a company(Q5) and the existence of incompetent employees in the higher levels of the company hierarchy(AvIEHL) is 0.344, which is greater than 0.05. Hence we can conclude that there is no significant correlation between the two factors in consideration.

57

The significant value of the correlation between the structure of a company(Q6) and the existence of Peter Principle related behavior patterns(AvPPB) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above-mentioned relationship is 0.564. Hence we can conclude that there is a strong significant positive correlation between the two factors in consideration.

The significant value of the correlation between the structure of a company(Q6) and the existence of incompetent employees in the higher levels of the company hierarchy(AvIEHL) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above-mentioned relationship is 0.416. Hence we can conclude that there is a significant positive correlation between the two factors in consideration.

The significant value of the correlation between the quality of performance and rewards management practices (AvPRM) of a company and the existence of Peter Principle related behavior patterns (AvPPB) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above-mentioned relationship is -0.721. Hence we can conclude that there is a strong significant negative correlation between the two factors in consideration.

The significant value of the correlation between the quality of performance and rewards management practices of a company (AvPRM) and the existence of incompetent employees in the higher levels of the company hierarchy (AvIEHL) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above mentioned relationship is -0.495. Hence we can

58

conclude that there is a significant negative correlation between the two factors in consideration.

The significant value of the correlation between the quality of recruitment practices (AvRP) of a company and the existence of Peter Principle related behavior patterns (AvPPB) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above-mentioned relationship is -0.691. Hence we can conclude that there is a strong significant negative correlation between the two factors in consideration.

The significant value of the correlation between the quality of recruitment practices of a company (AvRP) and the existence of incompetent employees in the higher levels of the company hierarchy (AvIEHL) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for abovementioned relationship is -0.508. Hence we can conclude that there is a strong significant negative correlation between the two factors in consideration.

The significant value of the correlation between the human resource development practices of a company (AvHRD) and the existence of Peter Principle related behavior patterns (AvPPB) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above mentioned relationship is -0.587. Hence we can conclude that there is a strong significant negative correlation between the two factors in consideration.

The significant value of the correlation between the human resource development practices of a company (AvHRD) and the existence of

59

incompetent employees in the higher levels of the company hierarchy (AvIEHL) is 0.001, which is less than 0.05. Further the Pearson correlation coefficient for above-mentioned relationship is -0.381. Hence we can conclude that there is a significant negative relationship between the two factors in consideration.

60

4.6 Multiple Regression Analysis Correlation coefficient r indicates the strength of the relationship between two variables, but it does not give an idea of how much of the variance in the dependent variable will be explained, when several independent variables simultaneously influence it. Multiple regression analysis is used to analyze such situations (Sekaran 2008, p405). Multiple regression analysis was used in this study to analyze the relationships when all independent variables simultaneously influence the dependent variable. It is conducted to examine how well those independent variables predict the dependent variable when taken as a model. As two components were identified as the determining factors of Peter Principle during section 4.3.1, multiple regression analysis was performed for each dependent variable. Multiple regression analysis results for the dependent variable existence of Peter Principle related behavior patterns and the independent variables size of the company , structure of the company, quality of performance and rewards management practices in a company , quality of recruitment practices in a company and quality of human resource development practices in a company , are given below. Table 4.28: Model summary for the existence of Peter Principle related behavior patterns and the contributing factors
Model 1 R 0.784(a) R Square 0.614 Adjusted R Square 0.609 Std. Error of the Estimate 0.65614

a Predictors: (Constant), Q6, Q5, AvHRD, AvRP, AvPRM Q5 - The size of the company Q6 - The structure of the company AvHRD - The quality of performance and rewards management practices in a company AvRP - The quality of recruitment practices in a company AvPRM - The quality of human resource development practices in a company

61

Table 4.29: ANOVA table for existence of Peter Principle related behavior patterns and the contributing factors
Model 1 Sum of Squares 267.197 167.904 df 5 390 Mean Square 53.439 .431 F 124.127 Sig. .000(a)

Regression Residual Total

435.101 395 a Predictors: (Constant), Q6, Q5, AvHRD, AvRP, AvPRM b Dependent Variable: AvPPB Q5 - The size of the company Q6 - The structure of the company AvHRD - The quality of performance and rewards management practices in a company AvRP - The quality of recruitment practices in a company AvPRM - The quality of human resource development practices in a company AvPPB - The existence of Peter Principle related behavior patterns

Table 4.30: Coefficients for existence of Peter Principle related behavior patterns and the contributing factors
Unstandardized Coefficients Model 1 (Constant) AvPRM AvRP AvHRD Q5 Q6 a Dependent Variable: AvPPB AvPPB - The existence of Peter Principle related behavior patterns B 5.163 -.508 -.284 .011 -.002 .586 Std. Error .172 .076 .069 .061 .041 .075 Standardized Coefficients Beta -.417 -.234 .010 -.001 .280

t B 29.973 -6.651 -4.101 .181 -.041 7.781

Sig. Std. Error .000 .000 .000 .856 .967 .000

The model summary table 4.27 above provides the R and R2 values for the relationships between the said dependent and independent variables. R or the correlation coefficient indicates the strength of the relationship. In this case the Rvalue indicates multiple R, which is the correlation of all the independent variables against dependent variable, which is 0.784. This is quite a high value and it confirms there is a strong relationship between the variables. The value of the R2 gives the amount of variance explained by these models. In this case the R2 value is 0.614 which means the independent variables together (the model) explain 61.4% of the variance in the existence of Peter Principle related behavior patterns. Hence we can

62

conclude that the independent variables considered largely define why the Peter Principle effect exists. The Coefficients table 4.29 shows Beta values for all the independent variables when they are regressed jointly against dependent variables. The independent variables, quality of performance and rewards management practices, quality of selection and recruitment practices have negative Beta values. Therefore it shows a negative relationship towards the dependent variables. The results imply that an increase in the above two independent variables will inherently decrease the dependent variable. Even though the independent variable quality of human resource development practices had a significant correlation when correlated individually with the dependent variable, it proves to be insignificant when jointly regressed with the other independent variables. The size of the company proves to be insignificant. Hence there is no relationship between the size of a company and the existence of Peter Principle effect related behaviors patterns. The structure of a company has a significant relationship and the Peter Principle effect related behavior patterns seems to be high in companies that have tall structures. Multiple regression analysis results for the dependent variable existence of incompetent employees in the higher levels of the company hierarchy and the independent variables size of the company, structure of the company, quality of performance and rewards management practices in a company, quality of recruitment practices in a company and quality of human resource development practices in a company, are given below. Table 4.31: Model Summary for the existence of incompetent employees in the higher levels of the hierarchy and the contributing factors
Model 1 R .566(a) R Square .320 Adjusted R Square .311 Std. Error of the Estimate .78509

a Predictors: (Constant), Q6, Q5, AvHRD, AvRP, AvPRM Q5 - The size of the company Q6 - The structure of the company AvHRD - The quality of performance and rewards management practices in a company AvRP - The quality of recruitment practices in a company AvPRM - The quality of human resource development practices in a company

63

Table 4.32: ANOVA table for the existence of incompetent employees in the higher levels of the hierarchy and the contributing factors
Sum of Squares 113.120 240.385 353.506

Model 1

df 5 390 395

Regression Residual Total

Mean Square 22.624 .616

F 36.705

Sig. .000(a)

a Predictors: (Constant), Q6, Q5, AvHRD, AvRP, AvPRM b Dependent Variable: AvIEHL Q5 - The size of the company Q6 - The structure of the company AvHRD - The quality of performance and rewards management practices in a company AvRP - The quality of recruitment practices in a company AvPRM - The quality of human resource development practices in a company AvIEHL - The existence of incompetent employees in the higher levels of the company hierarchy

Table 4.33: Coefficients for the existence of incompetent employees in the higher levels of the hierarchy and the contributing factors
Unstandardized Coefficients Model 1 (Constant) AvPRM AvRP AvHRD Q5 Q6 a Dependent Variable: AvIEHL AvIEHL - The existence of incompetent employees in the higher levels of the company hierarchy B 4.464 -.283 -.304 .109 -.053 .409 Std. Error .206 .091 .083 .073 .049 .090 Standardized Coefficients Beta -.258 -.278 .108 -.046 .217 t B 21.657 -3.098 -3.669 1.488 -1.088 4.543 Sig. Std. Error .000 .002 .000 .138 .277 .000

In this case the R-value, is 0.566. This is quite a high value and it confirms there is a strong relationship between the variables. The R2 value is 0.320 which means the independent variables together explains only 32% of the variance as to why there are incompetent employees in higher levels of the hierarchy. This could be due to the reason that there may be more reasons why incompetent employees exist in the higher levels of the hierarchy than the ones considered here. The Coefficients tables show Beta values for all the independent variables when they are regressed jointly against dependent variables. The independent variables, quality

64

of performance and rewards management practices, quality of selection and recruitment practices have negative Beta values in this case too. Therefore it shows a negative relationship towards the dependent variable. The results imply that an increase in the above two independent variables, will inherently decrease the dependent variable. Even though the independent variable quality of human resource development practices had a significant correlation when correlated individually with the dependent variable, it proves to be insignificant when jointly regressed with the other independent variables. The size of the company proves to be insignificant. Hence there is no relationship between the size of a company and the existence of incompetent employees in the higher levels of the company hierarchy. The structure of a company has a significant relationship and incompetent employees in the higher levels of the company hierarchy seem to be high in companies that have tall structures. The software development companies considered for the study were divided into two main groups considering the structure of the company as Tall or Flat in a broad sense. As the data collected for this aspect is categorical in nature, the Independent samples T-Test was used to test the relationship between the structure of a company and the existence of the Peter Principle effect. The results obtained from the test are as follows. Table 4.34: Group Statistics for structure of the company
Q6 Flat Tall Flat Tall N 199 197 199 197 Mean 2.5313 3.7134 2.7224 3.5076 Std. Deviation .72411 .99216 .81448 .90665 Std. Error Mean .05133 .07069 .05774 .06460

AvPPB AvIEHL

AvPPB - The existence of Peter Principle related behavior patterns AvIEHL - The existence of incompetent employees in the higher levels of the company hierarchy

According to the results obtained, it can be seen that the mean values obtained for existence of the Peter Principle related behavior patterns and incompetent employees

65

in higher levels of the hierarchy is comparatively less for companies with flat hierarchies.
In d p e d e n t S a m le T e t Table 4.35: Independent Sample Test efornstructurepof sthe scompany L e v e n e 's T e s t fo r E q u a lit y o f V a r ia n c e s t- t e s t f o r E q u a lit y o f M e a n s 9 5 % C o n fid e n c e In te r v a l o f th e D iff e r e n c e Mean S td . E r r o r S ig . ( 2 - t a ile dD) iff e r e n c eD iff e r e n c e L o w e r Up p e r .0 0 0 .0 0 0 .0 0 0 .0 0 0 - 1 .1 8 2 1 4 - 1 .1 8 2 1 4 - .7 8 5 2 5 - .7 8 5 2 5 .0 8 7 2 2 - 1 .3 5 3 6 2 - 1 .0 1 0 6 5 .0 8 7 3 6 - 1 .3 5 3 9 4 - 1 .0 1 0 3 3 .0 8 6 5 9 - .9 5 5 4 9 - .6 1 5 0 1 .0 8 6 6 4 - .9 5 5 5 9 - .6 1 4 9 1

F A v P P B E q u a l v a r ia n c e s 3 0 .8 5 8 assu m ed E q u a l v a r ia n c e s no t a ssu me d A v IE H L E q u a l v a r ia n c e s 6 .4 1 6 assu m ed E q u a l v a r ia n c e s no t a ssu me d

S ig . .0 0 0

t - 1 3 .5 5 3

df 394

- 1 3 .5 3 2 3 5 8 . 5 2 2 .0 1 2 -9 .0 6 8 394

-9 .0 6 4 3 8 8 .6 9 1

AvPPB - The existence of Peter Principle related behavior patterns AvIEHL - The existence of incompetent employees in the higher levels of the company hierarchy

According to table above the significant value of the relationships are less than 0.05. Therefore the relationships are significant. Hence we can conclude that there is a significant relationship between the structure of the company and the existence of the Peter Principle effect. The tall organization structures seem to be affected more according to the results obtained above. Hence we can conclude the hypothesis statements generated in Chapter 3 as follows.

There is no significant relationship between size of a company and the existence of Peter Principle effect. Therefore the alternative hypothesis was rejected with a 95% confidence level.

There is a significant relationship between the structure of a company and the existence of Peter Principle effect. Organizations with tall structures suffered more from the Peter Principle effect. Therefore the alternative hypothesis was accepted with a 95% confidence level.

66

There is a significant negative relationship between the quality of performance and rewards management practices in a company and the existence of Peter Principle effect. Therefore the alternative hypothesis was accepted with a 95% confidence level.

There is a significant negative relationship between the quality of selection and recruitment practices in a company and the existence of Peter Principle effect. Therefore the alternative hypothesis was accepted with a 95% confidence level.

There is no significant relationship between human resource development practices in a company and the existence of Peter Principle effect. Therefore the alternative hypothesis was rejected with a 95% confidence level.

4.7 Summary This chapter presented the data gathered through the empirical study conducted using employees working in software development organizations in Sri Lanka. Data collection methods, which were employed for this study, were discussed in detail, supported by the figures of the collected data. The chapter started with a description of the general information and demographics of the respondents. It then went on to check the validity of the data through reliability and validity analysis. Descriptive statistics were provided for the constructs identified. Inferential analysis techniques such as correlation analysis and multiple regression analysis were used to analyze the data set and test the hypotheses. The chapter concluded with the hypothesis test results.

67

CHAPTER 5 CONCLUSIONS AND RECCOMENDATIONS

5.1 Introduction The following sections of the final chapter would discuss conclusions derived from the data analysis while interpreting data analysis results in detail and the next section will provide the recommendations. Then, it would analyze the limitations of this study and highlight the guidelines for the future research. 5.2 Conclusions The purpose of this research was to identify if the Peter Principle effect exists in software development firms of Sri Lanka and to analyze the determining factors that associate with the existence of Peter Principle effect. After identifying the research problem, three research objectives were formed with five hypotheses. A stratified random sample size of 396 was derived from the population which consisted of employees from software development firms of Sri Lanka. A conceptual framework was designed after in depth literature review. The required data was collected with the use of an online questionnaire and the data analysis was carried out in an orderly manner as illustrated in chapter 4. The first objective of the study was to identify if the Peter Principle effect exists in the software development companies of Sri Lanka. With the knowledge gained from the in-depth literature review carried out, the researcher was able to identify the characteristics that indicate that the Peter Principle effect exists in a company. The descriptive statistics obtained for the areas that tested whether these characteristics were present in the software development organizations of Sri Lanka confirmed that the effect was present in a moderate level. Hence, objective one; was successfully achieved by proving that the Peter Principle effect is prevalent in the software development companies of Sri Lanka. The second objective of the study was to identify and analyze the determining factors that associated with the existence of the Peter Principle effect. These factors were

68

identified in the literature review and they were studied thoroughly using the survey data to identify their contribution to the existence of the Peter Principle effect. Five hypotheses were formed to identify if there was a relationship between the size, structure, quality of performance and rewards management practices, quality of selection and recruitment practices and quality of human resource development practices in an organization and the existence of Peter Principle effect. According to the results obtained under the data analysis chapter, there are strong significant relationships between the structure, quality of performance and rewards management practices, selection and recruitment practices and the existence of Peter Principle effect. The size of a company did not have a significant relationship with the existence of the Peter Principle effect. This implies that the Peter Principle effect can exist in any company regardless of its size. Also the quality of human resource development practices of a company did not have a significant relationship with the existence of the Peter Principle effect. This could be due the fact that once an employee reaches his ultimate level of incompetency, it is hard to improve his productivity by providing training as he is not suitable for the job. Hence whatever training that is provided should be provided before an employee reaches his level of incompetency. The third objective was to propose recommendations to help software development companies in Sri Lanka to avoid the Peter Principle effect. This has been achieved by the recommendations provided in the section 5.3 of the dissertation. At the end of the study the researcher was able to achieve all the objectives stated in chapter one. Hence we can conclude that the study was successful. 5.3 Recommendations and Managerial Implications Based on the views provided by the industry professionals, findings gathered from the study and the literature found on the study, this section would propose recommendations that software development firms can adopt to avoid the Peter Principle effect.

69

Opting for flatter hierarchies As mentioned by Peter and Hull (1969), the Peter Principle is most prevalent in organizations that have tall hierarchical structures. This was confirmed further by the results obtained from the analysis as the Peter Principle effect was less prevalent in flatter organizations. Therefore opting for flatter hierarchical structures seems to help software companies avoid the Peter Principle effect to some extent. Higher pay without promotions Employees often accept a promotion, not for the power and prestige, but the increased pay attached to it. If software companies are willing to offer large pay increases for excellent work within the same position, the Peter Principle would be avoided, and the employee could make more money while staying in the position he enjoys and in which he's competent. It is pretty much easy to carry this concept in flatter hierarchies. Clear division of managerial and technical career paths Software development organizations should offer enough opportunities in management as well as technical tracks. The skills needed to succeed and measures of success for each track are very different and sometimes unclear. To succeed in management track, one needs to be good at dealing with ambiguities, taking decisions based on partial data, and be able to deal to managing regular management challenges; measure of success most of the time could be very indirect. It could mostly be through the success of the team members or the success of a project assigned etc and hence can be very subjective and debatable. To succeed in the technical track, one needs to have deep technical and domain expertise, should be good at solving complex technical problems, and be able to provide technical and thought leadership; measure of success is very direct and objective and mostly based on visible results of the individual. Software development firms should identify their employees competencies, guide and groom them to reach their maximum potential

70

in the right track so that their personal development is serving organizational development benefiting both parties.
Demotion and Dismissal

Although not ethical, perhaps a good way to address the Peter Principle in an organization would be to institute a policy of demoting employees to their most appropriate level of work competence. If an employee isn't working out in a higher position, allowing him to go back to whatever position he excelled in would avoid the effects of the principle. This would, however, require the supervisor who made the poor promotion decision to admit he made a mistake, an act not often found in the higher levels of a hierarchy. Werhane, Radin and Bowie (2003, p.69) stated that, in dismissing or demoting employees, the employer is not denying rights to persons; rather, the employer is simply excluding that person's labor from the organization. This is quite a justifiable reason to demote or dismiss an employee who is not performing well. From the employers side, demotion or dismissal will be reducing the harm done by that employee to the organization and from the employees side he will get freed from a job that he is neither enjoys nor good at.
Recruiting employees on contract basis

Another solution to the Peter Principle effect is to recruit employees on contract basis. If the employee doesnt perform well, the employer has the right to not extend his contract period and the employee will have to leave the organization once his contract is over. This will avoid the problems that can arise from demotion and dismissal as such actions are seen as unethical in the eyes of the employees and it could harm the good name of the company. Playing the job role prior to the promotion Allowing the employee to play the next designation role prior to giving the promotion will also help avoid the Peter Principle effect. This way the employer will be able to evaluate whether the employee can do a good job if he was provided the designation. The promotion can be given based on the performance of the employee

71

in the given role. If he performs well he will be eligible for the next promotion, if not he will have to remain in his current position till he performs well enough to carry out the responsibilities of a designation at the next level. Improve selection and recruitment practices It is important to always provide proper job descriptions when advertising to avoid attracting employees who are not suitable for the job. Also companies should ensure that they have a competent, properly trained selection panel to interview and select the best candidates. Incompetent interviewers might end up selecting employees who are far worse than them resulting in an increase in the Peter Principle effect. Also the organization should have a standardize procedure of recruiting employees. According to Carpers (2009, p.50), what seem to give the best results are multiple interviews combined with a startup evaluation period of perhaps six months. Successful performance during the evaluation period is a requirement for joining the organization on a full-time regular basis. Improve performance and rewards management practices Proper performance and rewards management mechanisms should be put into practice. Standardized mechanisms like Management by Objectives (MBO), Balanced Score Cards can be used for proper performance evaluation depending on the objectives defined by the organization. According to Gately (1996) employers can avoid the Peter Principle as long as employees are judged on technical merit and accomplishment, and promotions are given to the technically proficient and verbally expressive employees, while the less technically proficient and verbally expressive wait their turn. Provide proper trainings and improve human resource development practices According to the analysis done, it was identified that the quality of human resource development practices in an organization was insignificant when taken as a model with the other constructs. This could be because when an employee reaches his

72

maximum level of incompetence, no amount of training can fix his state. Nevertheless providing proper training while the person is competent might reduce the chance of him becoming an incompetent employee. Also it will help him identify his strengths and weaknesses and help him identify what he is most talented in doing. Odiorne (1991) stated, that employees should be taught the skills and tasks in order to be knowledgeable because ongoing training can prevent competence from eroding and becoming obsolete. Outsource performance appraisals According to How to get human resources careers article, Performance appraisals may be carried out by the companys human resources department, or it may also be outsourced. Outsourcing and getting the appraisals done by an independent third party will help an organization get unbiased and reliable performance evaluations on their employees. Recruitment process outsourcing Recruitment Process Outsourcing (RPO) is an outsourcing arrangement whereby an external provider takes over part or all of the recruitment functionalities of an organization. The RPO provider assumes responsibility over the hiring process, from job profiling through on boarding, as well as the resources, methodologies and reporting used. With effective implementation RPO can reduce a companys time to hire and associated costs, increase the quality of candidate pool and ensure regulatory compliance. According to Nelson and Gerard (2005), RPO offers companies a proven way to attract the best talent and, ultimately, ensure the highest possible level of customer satisfaction.

73

5.4 Limitations of the Study Every research conducted has limitations of its own nature because of resource limitations, time considerations and many more. This study also has certain limitations of its own. Firstly, the research was limited to employees from the selected organizations. Software organizations can be either product based or project based. This was not considered when selecting the sample. There is a chance that the existence of the Peter Principle effect could differ according to the nature of work carried out. This was not properly captured in the collected sample as it did not have equal number of respondents from project based and product based software companies. This was a limitation. The population of IT professionals in Sri Lanka is high. A proper census could not be done and only a rough estimate was available regarding the number of IT professionals. When a convenience based sample is selected to collect responses, there is a chance that there may be biased results than expected. If a random sample was selected, the results would have been more accurate. Data gathering limitations should also be considered. As mentioned in chapter 3 data was gathered using an online survey. The correctness of gathered data and whether the respondents provided their actual impressions or manipulative responses was a major concern. As mentioned above, there were many limitations in this study and author did his best to make this study successful with those limitations. 5.5 Directions for Future Research This study was primarily concerned with identifying whether the Peter Principle effect existed in software development firms of Sri Lanka. This was successfully achieved and it was identified that the effect exists. If the limitations identified in the previous section can be overcome, the study can be further extended and the results will be more accurate.

74

While studying the factors that determine the existence of the Peter Principle effect only five factors (the size of the company , the structure of the company, the quality of performance and rewards management practices, the quality of recruitment and selection practices and the quality of human resource development practices) were taken into consideration. Further research should be carried out to identify what other factors have an influence on the existence of this effect. Further the researcher only checked whether the there were significant relationships between the factors and the Peter Principle effect in a broad manner. Each of these factors should be broken down into smaller and more specific sections and analyzed. For e.g. we could carry out a thorough search on which kind of performance evaluation mechanisms best reduce the existence of the Peter Principle effect. Research could be carried out to identify various human resource development practices that can help prevent competence from eroding and becoming obsolete. This way we would be able to identify ways to reduce the Peter Principle effect. The psychological factors associated with Peter Principle effect can be studied. That is research should be carried out to identify why employees are eager to accept promotions even when they know they would be happier doing something else in a lower level in the hierarchy. Various organization structures like tall, flat, hybrid etc can be studied to identify the best organization structure that suits software development firms to minimize the Peter Principle effect from occurring. Studies can be carried out on selection and recruitment practices and the skills of an employee that should be tested to select the best talent so that in the long run, the company will have less Peter Principled employees. Problems faced by the HR department when dealing with incompetent employees should be analyzed to identify proper remedies to minimize the harm done to the organization by employees who have reached their level of incompetence.

75

Software development firms will have employees working in many areas. It could be software engineering, quality assurance, business analysis, project management etc. In each of these areas there could be different reasons as to why the Peter Principle effect occurs. Studies should be carried out to identify these reasons and to provide solutions to prevent them. 5.6 Summary This chapter discussed the interpretation of results obtained from the data analysis and provided the recommendations based on the findings throughout this study. Further, it concluded the entire study by exploring pathways for future studies while explaining the limitations of the completed study.

76

REFERENCES
Anderson, R. E., Dubinsky, A. J., & Mehta, R. (1999, January-February). Sales managers: Marketings best example of the Peter Principle. Business Horizons, 42(1), 19-26. Anonymous. (1988, July). CPAs meet the Peter Principle. Management Review, 77(7), 9-10. BUDMAN.Can we turn back the tide of incompetence? The Conference Board http://www.theage.com.au/ed_docs/incompetence.pdf [accessed April 28, 2011]
Berger,Lance A. and Berger,Dorothy R.(2003),The Talent Management Handbook:

Creating Organizational Excellence by Identifying, Developing, and Promoting Your Best People.New York:McGraw-Hill

Better Estimation & Planning Can Help : Project Planning & Estimation. [ONLINE] Available at: http://www.galorath.com/wp/software-project-failure-costs-billionsbetter-estimation-planning-can-help.php. [Accessed 13 November 2011].

Campbell, A.2007. Call it a Small Business, No a Mid-Sized Business, No a Large Corporation.Small Business Trends. http://www.sellingtosmallbusinesses.com/call-it-asmall-business-no-a-mid-sized-business-no-a-large-corporation [accessed October 1,2011] http://www.sellingtosmallbusinesses.com/call-it-a-small-business-no-a-mid-sized-businessno-a-large-corporation/

Capers,Jones, 2009. Software Engineering Best Practices: Lessons from Successful Projects in the Top Companies. 1st Edition. New York: McGraw-Hill Osborne Media. Chapman, S. (2007, April 26). Why nothing fails like success: The Peter Principle in the executive branch. Reason online.http://www.reason.com/news/show/119867.html.

77

Cline, M.P., Lomow, G. and Girou,M. 1998.C++ FAQs.2nd ed.Boston:AddisonWesley CLARK.How the Peter Principle works.How Stuff Works http://money.howstuffworks.com/peter-principle.htm [accessed May 01, 2011]
Drucker,Peter,.F. (1993), The Practice of Management.New Jersey: Collins

Facts and Figures Why Technology Projects Fail. 2011. Facts and Figures Why Technology Projects Fail. Calleam Consulting Ltd.http://calleam.com/WTPF/? page_id=1445. [accessed 13 November 2011].

Faria, J. R. (2000, January). An economic analysis of the Peter and Dilbert Principles. Working Paper No. 101, School of Finance and Economics, University of Technology, Sydney, Australia. [On-line.] Available: ISSN: 1036-7373, 118. Fairburn, J. A., & Malcolmson, J. M. (2000). Performance, promotion, and the Peter Principle. The Review of Economic Studies, 68(1), 45-66. Fetzer, J. (2006, October 3). The curse of having a bad supervisor. Analytical & Bioanalytical Chemistry, 386, 1577-1578. Gately, R. F. (1996, May/June). Selecting managersHow to avoid the Peter Principle. Journal of Management in Engineering, 12(3), 3-5. How to get human resources careers. How to do things . http://www.howtodothings.com/business/human-resources-careers. [accessed 21 November 2011]. Koontz, H. and Weihrich, H. 1990. Essentials of Management.New York, McgrawHill. King, A. (2004, June 14-20). Ask yourself: Incompetent or in the wrong job? Mississippi Business Journal, 26(24), 5-6. Lazear, E. P. (2004, February). The Peter Principle: A theory of decline. Journal of Political Economy, 112(1), 141-163.
Malhotra N.K., Marketing Research: An Applied Orientation, Prentice Hall of India, 2007.

78

Minter, R. L. (1972, July). Mismanagement of training programs. Training and Development Journal, 26(7), 2-5. Nelson,Tim and Gerard, Brett.(2005)Recruitment process outsourcing:Transforming end to end talent acquisition.The Korn/Ferry Institute.
http://www.kornferryinstitute.com/files/pdf1/Futurestep_RPO_Whitepaper_FINAL_1205.pd f.[accessed 21 November 2011].

Newman, R. (2008, March 24). Combating the Peter Principle. U.S. News & World Report, 144(9), 6. North, G. (2004, July 17). Computers vs. the Peter Principle. [On-line]. Available: LewRockwell.com, 1-7. Odiorne, G. S. (1991, May). Competence versus passion. Training & Development Journal, 45(5), 61-64. Peter, L. F. 1985. Why things go wrong or the Peter Principle revisited. New York, NY: William Morrow & Company, Inc. Peter, L. F. 1972. The Peter prescription How to make things go right. New York, NY: Bantam Books. Peter, L. F.,and Hull, R. 1969. The Peter principle Why things always go wrong. New York, NY: William Morrow & Company, Inc. Peter,Laurence.J. and Hull,Raymond (1996), The Peter principle: Why things always go wrong.New York: Buccaneer Books Pluchino,A., Rapisarda, A. and Garofalo,C.2009. The Peter Principle Revisited: A Computational Study.Cornell University Libarary. http://arxiv.org/abs/0907.0455[Accessed October 1,2011]
Reichel, M., & Ramey, M. A. (Eds.). (1987). Conceptual frameworks for bibliographic education: Theory to Practice. Littleton Colorado: Libraries Unlimited Inc.

SchaaP,James I. and Ogulnick Sy.2009.The Peter Principle: Is This Forty-Year-Old Universal Phenomenon in Decline or Growing?Journal of Global Business

79

Managers,vol.5, http://www.jgbm.org/page/1%20James%20Ike%20Schaap%20.pdf [Accessed October 1,2011] Sekaran U. (2008), Research Methods for Business-A Skill Building Approach, 4th ed., Wiley, Delhi. SLDEB . Information Communication Technologies (ICT) - Industry Overview. http://www.srilankabusiness.com/trade_info/srilankaproduct/ict.htm [accessed April 28, 2011] Software Project Failure Costs Billions.. Better Estimation & Planning Can Help : Project Planning & Estimation. 2011. Software Project Failure Costs Billions.. Sri Lanka Business Portal - Trade Information. 2011. Sri Lanka Business Portal Trade Information. http://www.srilankabusiness.com/trade_info/srilankaproduct/ict.htm. [accessed 13 November 2011] Taylor, E. K. 1969, The Peter Principle. Personnel Psychology, 22(4), 535-537. Werhane, P.H, Radin,T.J., and Bowie.N.E (2003), Employment and employee rights.New York :Wiley
Yate,Martin (1997), 4th edition , Hiring the Best: A Manager's Guide to Effective Interviewing.New York :Adams Media Corporation

80

Appendix A Questionnaire
Research Questionnaire A study on worker competence in Software Development Firms of Sri Lanka
Dear Participant, I am a postgraduate student of University of Moratuwa, currently doing a research study which is a requisite to complete the Master of Business Administration in Management of Technology programme. This questionnaire is designed to study on worker competence in software development firms of Sri Lanka. Your responses will be kept strictly confidential and all information provided here will only be used for this research study. Thank you very much for spending your valuable time and I greatly appreciate your help in furthering this research endeavor.

Thanks and Regards, M.Samaratunga

81

1.

Experience in IT industry

Years

2. Which of the following category best describes your current job?

Trainees / Associate Level

( E.g. Associate Software Engineers , Associate QA Engineers , Associate Business Analysts etc)

Intermediate Level (E.g. Software Engineers, Business Analysts, QA Engineers etc) Senior Level (E.g. Senior Software Engineers, Senior QA Engineers, Senior BAs etc) Lead Level (E.g. Technical Team Leads, Business Consultants etc) Managerial Level (E.g. Architects, Project Managers etc) If Other, Please specify: .. 3. Experience in the current organization Years

4. How many years have you been employed in your current position? Years 5. How many employees are there in the company that you work at present? Less than 100 Greater than 300 6. How would you define the structure of your organization in a broad sense? Flat Hierarchical (Tall) Between 100 and 300

From your perspective, in your organization the occurrence of the following scenarios are:

There are previously competent employees who have become incompetent after getting promoted

82

Very High

Very Less

Moderate

High

Less

There are employees who are not successful in obtaining further promotions due to incompetence in their current position Some incompetent employees are promoted to a "higher" position to minimize the harm done by them The rate of competent employees leaving the company is Innovative ideas and processes are discouraged as preserving the organizations existing way of doing work is highly valued than providing efficient service to the customer Highly competent employees are fired to preserve the organizations hierarchy Some employees are placed very high in the organizations hierarchy based on personal relationships Incompetent employees get promoted in the hierarchy if they are followers of the superiors Competent people have to push hard to get promoted There are employees who shift the blame on to others for their own actions There are superiors who take credit by getting work assigned to them, done by competent employees under them There are superiors who ignore their duties and focus more on activities that are not direct responsibilities of their position There are employees who are overly stressed , mentally disturbed and frequently sick even though they do not do a lot of productive work The number of incompetent employees in the higher levels of your organizations hierarchy is The number of competent employees in the lower levels of your organizations hierarchy is The productivity of the employees in the higher levels of your organizations hierarchy is

9 1 0 1 1

1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 2 1 2 2

83

2 3

The productivity of the employees in the lower levels of your organizations hierarchy is disagreeNeither agree nor

Strongly Disagree

With regards to your organization

2 4 2 5 2 6 2 7 2 8 2 9 3 0 3 1 3 2 3 3 3 4 3 5

I am satisfied with my current job with regards to the responsibilities assigned to me I believe I am eligible for further promotions I am happy to take over more responsibilities My organization has a standard method of evaluating employee performance The employees in my organization are happy with the current performance evaluation method My organization maintains proper job descriptions for all job positions My current job responsibilities have been clearly communicated to me The work that I perform matches with my job description In my organization rewards/promotions are solely based on performance ratings My organization provides comprehensive job descriptions when advertising job vacancies In my organization selection interviews are well structured to identify the best candidates The members of the selection panels in my organization are competent and properly are trained

84

Strongly agree

Disagree

Agree

3 6 3 7 3 8 3 9 4 0

My organization gives due attention to skills such as teamwork, leadership, attitude etc when recruiting new employees My organization provides job related training for all employees on technical matters My organization provides training on soft skills (E.g. teamwork, leadership skills etc ) My organization has a well defined HR development /talent management process to achieve organizational goals The trainers in my organization are highly competent

4 1

The employees in my organization are happy with the training they receive Kindly answer the following questions. The information is required solely for statistical purposes. 42. Age: Years

43. Gender: 44. Marital Status Single Married

Male

Female

45. What is the highest academic qualification you have obtained? Advanced Level or below Diploma (E.g. ACS) Bachelors Degree or equivalent professional qualifications (E.g. BCS) Postgraduate Qualifications Thank you for filling the questionnaire

85

Appendix B SPSS Data Analysis Output

This section contains the reliability and validity analysis outputs and the descriptive statistics outputs obtained via the SPSS software .Please refer Appendix A for the question descriptions. Reliability and Validity Analysis a) Factors that denote the existence of Peter Principle effect Scale: Reliability Analysis Round 1
Case Processing Summary N 396 0 % 100.0 .0

Cases

Valid Excluded(a)

Total 396 100.0 a Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha .929

N of Items 20

Item Statistics Mean 3.17 3.13 2.76 3.55 3.15 2.61 3.16 3.16 3.09 3.30 3.38 3.17 2.96 3.07 3.58 3.09 3.55 Std. Deviation 1.406 .983 1.221 1.269 1.396 1.269 1.496 1.350 1.038 1.244 1.284 1.311 1.052 1.000 .763 .957 .750 N 396 396 396 396 396 396 396 396 396 396 396 396 396 396 396 396 396

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22 Q23

86

Q24 Q25 Q26

2.46 2.27 2.23

.903 .467 .465

396 396 396

Item-Total Statistics Corrected Item-Total Correlation .815 .755 .785 .682 .789 .760 .827 .592 .644 .824 .814 .821 .719 .722 .268 .699 .207 -.180 -.154 -.158 Cronbach's Alpha if Item Deleted .921 .923 .922 .924 .921 .922 .920 .926 .925 .921 .921 .921 .923 .924 .931 .924 .932 .938 .934 .934

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22 Q23 Q24 Q25 Q26

Scale Mean if Item Deleted 57.68 57.72 58.09 57.30 57.70 58.24 57.69 57.69 57.76 57.55 57.47 57.68 57.89 57.78 57.27 57.76 57.30 58.39 58.58 58.62

Scale Variance if Item Deleted 180.839 192.109 186.019 188.296 181.969 185.746 178.370 189.790 193.956 184.243 183.667 182.817 191.596 192.609 207.169 194.085 208.602 217.622 215.535 215.577

Scale Statistics Mean 60.85 Variance 213.641 Std. Deviation 14.616 N of Items 20

Scale: Reliability Analysis Round 2

Case Processing Summary N Cases Valid Excluded(a) Total 396 0 396 % 100.0 .0 100.0

a Listwise deletion based on all variables in the procedure. Reliability Statistics

87

Cronbach's Alpha .956

N of Items 15

Item Statistics Mean 3.17 3.13 2.76 3.55 3.15 2.61 3.16 3.16 3.09 3.30 3.38 3.17 2.96 3.07 3.09 Std. Deviation 1.406 .983 1.221 1.269 1.396 1.269 1.496 1.350 1.038 1.244 1.284 1.311 1.052 1.000 .957 N 396 396 396 396 396 396 396 396 396 396 396 396 396 396 396

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22

Item-Total Statistics Corrected Item-Total Correlation .818 .771 .782 .693 .793 .764 .819 .596 .651 .830 .822 .827 .724 .738 .717 Cronbach's Alpha if Item Deleted .951 .953 .952 .954 .952 .953 .952 .957 .955 .951 .951 .951 .954 .953 .954

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22

Scale Mean if Item Deleted 43.60 43.64 44.01 43.21 43.61 44.15 43.61 43.60 43.67 43.46 43.38 43.59 43.81 43.69 43.68

Scale Variance if Item Deleted 177.765 188.641 183.106 184.912 178.851 182.620 175.707 186.610 190.692 181.075 180.424 179.635 188.405 189.125 190.548

Scale Statistics Mean Variance Std. Deviation N of Items

88

46.77

210.418

14.506

15

Principle Component Analysis

Communalities Initial 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 Extraction .718 .910 .665 .536 .717 .664 .800 .510 .483 .788 .783 .794 .617 .904

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22

1.000 .865 Extraction Method: Principal Component Analysis. Total Variance Explained Initial Eigenvalues % of Cumulative Total Variance % 9.432 62.883 62.883 1.321 8.805 71.687 .759 .589 .485 .480 .390 .315 .274 .247 .210 .183 .136 .122 .055 5.063 3.929 3.235 3.198 2.601 2.100 1.825 1.647 1.399 1.221 .909 .816 .368 76.750 80.679 83.914 87.113 89.714 91.815 93.640 95.288 96.686 97.907 98.816 99.632 100.000 Extraction Method: Principal Component Analysis. Component Matrix(a) Extraction Sums of Squared Loadings % of Cumulative Total Variance % 9.432 62.883 62.883 1.321 8.805 71.687 Rotation Sums of Squared Loadings % of Cumulative Total Variance % 6.550 43.668 43.668 4.203 28.019 71.687

Componen t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

89

Component Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22 1 .846 .814 .813 .732 .822 .795 .843 .645 .695 .853 .845 .852 .759 .781 .542 2 .498

.765 .528 Extraction Method: Principal Component Analysis. a 2 components extracted. Rotated Component Matrix(a) Component Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22 .546 .831 .835 .839 .731 .900 1 .704 .689 .581 .781 .745 .855 .630 .430 2 .471 .885 .436 .445

.880 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 3 iterations.

90

Component Transformation Matrix 1 2 .803 .596 -.596 .803 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Component 1 2

b) Factors that explain the quality of performance and rewards management practices in an organization
Case Processing Summary N 396 0 % 100.0 .0

Cases

396 100.0 a Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha .915

Valid Excluded(a ) Total

N of Items 6

Item Statistics Mean 3.02 2.75 2.88 3.06 3.03 2.97 Std. Deviation 1.058 .950 1.051 1.062 1.027 1.030 N 396 396 396 396 396 396

Q27 Q28 Q29 Q30 Q31 Q32

Item-Total Statistics Corrected Item-Total Correlation .751 .750 .772 .806 .736 .743 Cronbach's Alpha if Item Deleted .901 .901 .897 .892 .902 .901

Q27 Q28 Q29 Q30 Q31 Q32

Scale Mean if Item Deleted 14.70 14.96 14.83 14.65 14.68 14.74

Scale Variance if Item Deleted 18.779 19.571 18.659 18.318 19.115 19.039

Scale Statistics

91

Mean 17.71

Variance 26.778

Std. Deviation 5.175

N of Items 6

Principle Component Analysis

Communalities Initial 1.000 1.000 1.000 1.000 1.000 Extraction .692 .688 .718 .759 .670

Q27 Q28 Q29 Q30 Q31 Q32

1.000 .681 Extraction Method: Principal Component Analysis. Total Variance Explained Initial Eigenvalues Component 1 2 3 4 5 6 Total 4.209 .680 .365 .301 .273 .171 % of Variance 70.144 11.331 6.090 5.024 4.557 Cumulative % 70.144 81.475 87.565 92.589 97.147 Extraction Sums of Squared Loadings Total 4.209 % of Variance 70.144 Cumulative % 70.144

2.853 100.000 Extraction Method: Principal Component Analysis.

Component Matrix(a) Component Q27 Q28 Q29 Q30 Q31 Q32 1 .832 .829 .847 .871 .819

.825 Extraction Method: Principal Component Analysis. a 1 components extracted.

92

Rotated Component Matrix(a) a Only one component was extracted. The solution cannot be rotated.

c) Factors that explain the quality of selection and recruitment practices in an organization
Case Processing Summary Cases Valid Excluded(a ) Total N 396 0 % 100.0 .0

396 100.0 a Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha .880

N of Items 4

Item Statistics Mean 3.07 3.02 3.01 3.10 Std. Deviation .966 1.035 1.010 1.018 N 396 396 396 396

Q33 Q34 Q35 Q36

Item-Total Statistics Corrected Item-Total Correlation .688 .732 .788 .754 Cronbach's Alpha if Item Deleted .866 .849 .827 .841

Q33 Q34 Q35 Q36

Scale Mean if Item Deleted 9.13 9.18 9.18 9.10

Scale Variance if Item Deleted 7.393 6.890 6.777 6.877

Scale Statistics Mean 12.19 Variance 11.939 Std. Deviation 3.455 N of Items 4

93

Principle Component Analysis

Communalities Initial 1.000 1.000 1.000 Extraction .673 .727 .790

Q33 Q34 Q35 Q36

1.000 .754 Extraction Method: Principal Component Analysis. Total Variance Explained Initial Eigenvalues Component 1 2 3 4 Total 2.943 .458 .363 .236 % of Variance 73.574 11.439 9.085 Cumulative % 73.574 85.014 94.098 Extraction Sums of Squared Loadings Total 2.943 % of Variance 73.574 Cumulative % 73.574

5.902 100.000 Extraction Method: Principal Component Analysis. Component Matrix(a) Component Q33 Q34 Q35 Q36 1 .820 .852 .889

.868 Extraction Method: Principal Component Analysis. a 1 components extracted. Rotated Component Matrix(a) a only one component was extracted. The solution cannot be rotated.

94

d) Factors that explain the quality of human resource development practices in an organization
Case Processing Summary N 396 0 % 100.0 .0

Cases

396 100.0 a Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha .924

Valid Excluded(a ) Total

N of Items 5

Item Statistics Mean 3.01 2.86 2.72 2.96 2.89 Std. Deviation 1.119 1.090 1.054 1.058 1.016 N 396 396 396 396 396

Q37 Q38 Q39 Q40 Q41

Item-Total Statistics Corrected Item-Total Correlation .815 .771 .750 .811 .862 Cronbach's Alpha if Item Deleted .903 .912 .916 .904 .895

Q37 Q38 Q39 Q40 Q41

Scale Mean if Item Deleted 11.44 11.58 11.72 11.49 11.55

Scale Variance if Item Deleted 13.806 14.299 14.673 14.245 14.208

Scale Statistics Mean 14.44 Variance 21.842 Std. Deviation 4.674 N of Items 5

95

Principle Component Analysis

Communalities Initial 1.000 1.000 1.000 1.000 Extraction .787 .725 .700 .784

Q37 Q38 Q39 Q40 Q41

1.000 .844 Extraction Method: Principal Component Analysis. Total Variance Explained Initial Eigenvalues Component 1 2 3 4 5 Total 3.839 .468 .313 .241 .138 % of Variance 76.789 9.363 6.256 4.829 Cumulative % 76.789 86.152 92.407 97.237 Extraction Sums of Squared Loadings Total 3.839 % of Variance 76.789 Cumulative % 76.789

2.763 100.000 Extraction Method: Principal Component Analysis. Component Matrix(a) Component Q37 Q38 Q39 Q40 Q41 1 .887 .852 .837 .885

.918 Extraction Method: Principal Component Analysis. a 1 components extracted. Rotated Component Matrix(a) a Only one component was extracted. The solution cannot be rotated.

96

Descriptive Statistics
Descriptive statistics for the factors that explain the existence of Peter Principle Effect N 396 396 396 396 396 396 396 396 396 396 396 396 396 396 396 396 Minimum 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Maximum 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 Mean 3.17 3.13 2.76 3.55 3.15 2.61 3.16 3.16 3.09 3.30 3.38 3.17 2.96 3.07 3.09 Std. Deviation 1.406 .983 1.221 1.269 1.396 1.269 1.496 1.350 1.038 1.244 1.284 1.311 1.052 1.000 .957 Variance 1.977 .966 1.490 1.610 1.948 1.610 2.239 1.824 1.077 1.548 1.649 1.719 1.107 1.000 .916

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q22 Valid N (listwise)

Descriptive statistics for the factors that explained the quality of performance and rewards management practices N 396 396 396 396 396 396 396 Minimum 1 1 1 1 1 1 Maximum 5 5 5 5 5 5 Mean 3.02 2.75 2.88 3.06 3.03 2.97 Std. Deviation 1.058 .950 1.051 1.062 1.027 1.030 Variance 1.119 .902 1.105 1.128 1.055 1.060

Q27 Q28 Q29 Q30 Q31 Q32 Valid N (listwise)

Descriptive statistics for the factors that explained the quality of selection and recruitment practices N 396 396 396 396 396 Minimum 1 1 1 1 Maximum 5 5 5 5 Mean 3.07 3.02 3.01 3.10 Std. Deviation .966 1.035 1.010 1.018 Variance .932 1.071 1.020 1.036

Q33 Q34 Q35 Q36 Valid N (listwise)

97

Descriptive statistics for the factors that explained the quality of human resource development practices

Q37 Q38 Q39 Q40 Q41 Valid N (listwise)

N 396 396 396 396 396 396

Minimum 1 1 1 1 1

Maximum 5 5 5 5 5

Mean 3.01 2.86 2.72 2.96 2.89

Std. Deviation 1.119 1.090 1.054 1.058 1.016

Variance 1.253 1.188 1.111 1.120 1.032

Correlations
Corre lations AvPRM Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N AvPRM 1 396 .811** .000 396 .794** .000 396 -.721** .000 396 -.495** .000 396 .059 .240 396 -.436** .000 396 AvRP .811** .000 396 1 396 .739** .000 396 -.691** .000 396 -.508** .000 396 .066 .188 396 -.453** .000 396 AvHRD .794** .000 396 .739** .000 396 1 396 -.587** .000 396 -.381** .000 396 .147** .003 396 -.332** .000 396 AvPPB -.721** .000 396 -.691** .000 396 -.587** .000 396 1 396 .719** .000 396 -.019 .711 396 .564** .000 396 AvIEHL -.495** .000 396 -.508** .000 396 -.381** .000 396 .719** .000 396 1 396 -.048 .344 396 .416** .000 396 Q5 .059 .240 396 .066 .188 396 .147** .003 396 -.019 .711 396 -.048 .344 396 1 396 .077 .128 396 Q6 -.436** .000 396 -.453** .000 396 -.332** .000 396 .564** .000 396 .416** .000 396 .077 .128 396 1 396

AvRP

AvHRD

AvPPB

AvIEHL

Q5

Q6

**. Correlation is significant at the 0.01 level (2-tailed).

98