Marketing Research

Definition of MR
MR is the systematic and objective
identification, collection, analysis,
dissemination, and use of
information for the purpose of
improving decision making related to
the identification and solution of
problems and opportunities in
marketing.
 Classification of MR
Problem identification research:
market potential, brand image
market share, forecasting etc.
Problem solving research:
segmentation, product, pricing,
promotion, distribution research.
 Growth of MR
Large scale production
Sellers market to buyers market
Emergence of specialists
Increasing use of technology
Rural to urban shift
Coordination between marketing
research team and management
Changing marketing environment
 Scope of MR
Product research
Sales research
Customer research
Pricing research
Advertising research
 When MR is unnecessary
Cost of research is more than
benefits
Past information
Personal information
Difficult to communicate emotions
Short life cycle products
Competitor copying the idea
 The MR process
Defining the problem
Developing an approach to the
problem
Formulating research design
Collecting data
Preparing and analysing data
Preparing and presenting the report
 The MR industry
Research suppliers: internal and
external
External: full service or limited
service
Full service: customised or
syndicated or internet/social media
services
Limited service: field services or
qualititative services or technical and
analytical services or other services.
 Consideration to hire
agency:
Technical expertise
Objectivity
Confidentiality
Economic factors
Timely submission of reports
Expertise in relevant research
Reputation of the agency
 Careers in MR
VP (MR), research director, asst. director, project
manager, statistician, data processing specialist,
senior analyst, analyst, junior analyst, field work
director, operational supervisor.
For getting in MR one should:
Take all marketing subjects; take statistics and QT;
acquire internet, social media and computer
skills; take courses in psychology, sociology and
consumer behavior; acquire effective written and
verbal communication skills; think creatively.
Ethical issues in Marketing research

Ethics means principles that govern a

persons behavior or the conduct of
an activity. Ethics as related to MR
deals with judgement in respect of
two types of activities:
Activities that are undesirable or
inappropriate as they violate ethical
norms.
Activities that need to be
undertaken.
 Unethical examples in MR
The MR agency submit highly
technical proposal which in reality is
not needed.
Elaborate sample plan while proposal
while not adhered to that while
actual research.
Fake data entry in questionnaires
Doctored data to change the
analysis.
Not much effort put in while making
 Ethical issues regards the
constituents of MR
Participants: right to privacy
Sponsor: right to quality research
Market research agency: right to
safety of staff and training.
Public: right to move about freely; no
inconvenience caused due to survey
etc
 Codes of ethics
In order for a code to be effective, it
must satisfy the following conditions:
It should be regulative
It should protect the interest of
profession as well as public interest.
It should be behaviour specific
It should be enforceable.
 MR professional bodies
American marketing association
(AMA)
Direct marketing association(DMA)
Marketing research association(MRA)
Council for American survey research
association(CASRO)
Market research society of
India(MRSI)
 MRSI code of conduct
Responsibilities towards informants
Relations with the general public and
business community.
The mutual responsibilities of clients
and researchers.
Reporting standards
Additional requirements of public
opinion polls
Implementation of the code
 Problem definition
Problem definition involves stating the problem
confronting the managers ( the management
decision problem) and the marketing research
problem that the researcher will address. The
latter should be broken down in a broad statement
and the specific components of the MR problem.
Only when the client and the researcher have
clearly defined and agreed on the MR problem, the
research can be designed and conducted properly.
Thus, it is critical to clearly identify and define the
MR problem.
Problem defining process/ approach

Tasks involved : discussion with decision

makers; interviews with experts; secondary
data analysis; qualitative research.
Environmental context of problem
Problem definition: Management decision
problem to Marketing research problem.
Approach to the problem: analytical models;
research questions and hypotheses;
specification of information needed.
 Tasks involved
Discussions with decision makers: problem audit
is a comprehensive examination of a marketing
problem to understand its origin and nature.
Interviews with industry experts: unstructured
talk; google groups to begin with.
Secondary data analysis
Qualitative research: an unstructured,
exploratory research methodology based on
small samples intended to provide insight and
understanding of the problem setting.
 Environmental context of the
problem
Past information and forecasts
Resources and constraints
Objectives of the decision maker
Buyer behaviour
Legal environment
Economic environment
Marketing and technological skills
 Management decision problem vs
Marketing Research problem
The management decision problem is the
problem confronting the decision maker; it
asks what the decision maker needs to do.
It is action oriented, focuses on
symptoms.
The MR problem entails determining what
information is needed and how can it be
obtained in the most feasible way. It is
information oriented, focuses on the
underlying causes.
 Examples.
Management decision Marketing research problem
problem
Should a new product be To determine consumer
introduced? preferences and purchase
intentions for the proposed new
product.
Should the advertising campaign To determine the effectiveness of
be changed? the current advertising campaign.
Should the price of the brand be To determine the price elasticity
increased? of demand and the impact on
sales and profits of various levels
of price changes.
Should the organisation produce To determine if consumers would
more motor cycles? be loyal buyers of the brand in
the long run.
 Defining the MR problem
Broad statement: the initial
statement of the MR problem that
provides an appropriate perspective
on the problem.
Specific components : the second
part of MR problem definition. They
focus on key aspects of the problem
and provide clear guidelines on how
to proceed.
 Components of Approach to
problem
Analytical framework and models: in general MR
should be based on objective evidence and theory.
Objective evidence is unbiased evidence that is
supported by empirical findings. Theory is a
conceptual framework based on foundational
statements that are assumed to be true and guide
the collection of this data.
An analytical framework is statement of the theory
as applied to the MR problem. An analytical model
consists of set of variables related in a specified
manner to represent all or part of a real system or
process.
Research questions and hypotheses
RQ are refined statements of the specific components
of the problem. They are designed to ask specific
information required to address each problem
component. Eg: how important is food for airline
customers?
A hypotheses is an unproven statement or proposition
about a factor or phenomenon that is of interest to a
researcher. It may be a tentative statement about the
relationships discussed in theory or analytical models.
They reflect the researchers expectation and can be
tested empirically. Eg: H1: food is an important factor
for airline customers. H2: travellers value branded
food.
Specification of information needed

By focusing on each component of

the problem, the research questions
and hypotheses, the researcher can
determine what information should
be obtained. It is helpful to carry out
this exercise for each component of
the problem.
 Research design
A research design is a framework or
blueprint for conducting the
marketing research project. It gives
details of the procedures necessary
for obtaining the information needed
to identify or solve MR problem. It
specifies the detail of implementing
the approach to solve the MR
problem.
 Components of research
design
Analysis of secondary and syndicated data
Qualitative research
Survey and observation research
Experimental research
Measurement and scaling procedures
Design and pretest of a questionnaire
Sampling process
Development of preliminary plan of data
analysis
 Basic research designs
Research design are of two types:
exploratory and conclusive.
Conclusive are of two types:
descriptive and causal.
Descriptive are of two types:
longitudinal and cross sectional.
 Exploratory vs conclusive
Exploratory Research Conclusive

To provide insights and To test specific hypotheses and

understanding. examine relationships.
Information needed is defined Information needed is clearly
only loosely. Research process is defined. Research process is
flexible and unstructured. Sample formal and structured. Sample is
is small and non-representative. large and representative. Data
Data analysis is qualitative. analysis is quantitative.
Findings are tentative Findings are conclusive

Generally followed by an Findings used as input into

exploratory or conclusive decision making.
research.
 Other researches
Descriptive research is a type of conclusive research that
has its major objective as the description of something-
usually market characteristics or functions. They involve
surveys or observations.
Cross sectional designs involve the collection of information
from any given sample of population elements only once.
In longitudinal designs, a fixed sample of population
elements is measured repeatedly on the same variables.
Panel is a sample of respondents who have agreed to
provide information at specified intervals over an extended
period.
Causal research is used to obtain cause and effect
relationships.
 Primary vs secondary data
Primary data Secondary data
Collection purpose For the problem at For other problems
hand
Collection process Very involved Rapid and easy
cost high Relatively low
time long short
 Criteria for evaluating secondary
data
Specifications: methods used to
collect data
Error
Currency: when was it collected
Objective
Nature : content of data
dependability
 Classification of secondary
data
Secondary data: internal and
external
Internal: customer database, data
warehousing and data mining, CRM
and database marketing, social
media.
External: business/ non government,
government, syndicated services,
social media.
 Internal data
Customer database is transfering raw sales information to a
computer. The information is also obtained at warranty cards
and loyalty programs.
Data warehouse is a centralised database that consolidates
company wide data from a variety of operational systems.
Data mining is the technique involving the use of powerful
computers and advanced statistical and other software to
analyse large databases in order to discover hidden patterns
in data.
CRM is a decision support system that is used for managing
the interaction between an organisation and its customers.
Database marketing is the use of CRM database to develop
relationships and highly targeted marketing efforts with
individuals and customer groups.
 External data
Business/non government sources are non
profit organisations, trade and professional
organisations, commercial publishers,
investment broking firms. There are umpteen
journals/ indexes/ sites from where the data
can be accessed.
Government sources: census data and other
data. These data are highly reliable as well as
rich in quality. The US government is the
largest source of secondary data in the world.
 Syndicated data
Syndicated data are a special form of
secondary data that are meant for
use by multiple clients and one can
purchase them on a subscription
basis. Hence, the cost is low
compared to the cost of collecting
primary data. Hence, it is useful to
examine and analyse the syndicate
data before the collection of primary
data.
 Classification of syndicate
data
They are broadly classified in two :
household/customers and institutions.
Household are divided in: surveys ,
purchase/media panels and electronic scanner
services.
Surveys: psychographics ,advertising evaluation
and general.
Purchase/Media panels: purchase panels, media
panels
Electronic scanner: volume tracking data,scanner
panels, scanner panels with cable TV.
 Periodic survey vs panel
survey
Periodic surveys are surveys that
collect data on the same set of
variables at regular intervals, each
time sampling from a new group of
respondents.
Panel surveys are surveys that
measure the same group of
respondents over time, but not
necessary the on the same variable.
They are also called omnibus panels.
 Syndicate services for institutional
data
Syndicate for institutions: audit
services, industry services.
Audit services: retailers, wholesalers
Industry services: direct inquiries,
clipping services, corporate reports.
 MR and social media
Social media is rich source of data for both primary
and secondary data. It is appropriate for
conducting exploratory, descriptive and causal
research. They eliminate cost of building and
maintaining the traditional panels. It is very
convenient as well for the customer. Eg: company
blog, FB page, twitter account can generate rich
secondary data. Information gathered from social
media is used by the syndicate firms to understand
the market, answer the clientsconcerns, connect
to consumers and potential participants, conduct
online research and publish reports.
 Qualitative research
Qualitative research is an
unstructured exploratory research
methodology based on small samples
that provides insights and
understanding of the problem
setting.
In developing an approach,
qualitative research is often used for
generating hypotheses and
identifying variables that should be
 MR data
MR data: primary and secondary
Primary data: qualititative and
quantitative
Quantitative: descriptive and causal
Descriptive: survey data and
observational data
Causal: experimental data
 Quantitative research
A research methodology that seeks
to quantify the data and typically
applies some form of statistical
analysis.
Qualitative research must be
followed by quantitative research.
They are complimentary to each
other.
 Qualitative vs quantitative
research
qualitative quantitative

Objective To gain qualitative To quantify the data

understanding of the and generalise the
underlying reasons results from the
and motivations. sample to the
population of interest.
sample Small ; non Large; representative
representative.
Data collection unstructured structured

Data analysis Non-statistical stattistical

outcome Develop an initial Recommend a final

understanding course of action.
 Classification of qualitative
research
Qualitative research: direct(nondisguised)
and indirect(disguised).
Direct: focus groups and depth interviews
Indirect: projective techniques

Direct approach is where the purpose of the

project is disclosed to the respondent or are
obvious given the nature of the interview.
Indirect approach is where the purpose is
disguised from the respondents.
 Focus group
An interview or a discussion
conducted by a trained moderator
among a small group of respondents
in an unstructured and natural
manner.
The main purpose is to gain insights
by listening to a group of people from
the appropriate target market talk
about issues of interest to the
researcher.
 Characteristics of focus
groups
Group size 8-12

Group composition Homogenous; respondents

prescreened
Physical setting Relaxed, informal atmosphere

Time duration 1-3 hours

recording Use of audio/ video recording

moderator Observational/ interpersonal/ and

communication skills of the
moderator.
 Planning and conducting focus
groups
Specify the objectives of the
qualitative research
State the objectives of the focus
group
Write a screening questionnaire
Develop a moderators outline
Conduct the focus group interview
Prepare the focus group report
Advantages of focus groups
The richness of comments from the
respondents
Group interaction provides insight.
Snowball effect
Spontaneous and candid responses.
 Disadvantages of focus
groups
The researchers can be trapped to
take the findings as conclusive rather
than exploratory.
Difficult to moderate
Quality depends on skills of
moderator; unfortunately skilled
moderators are rare and very
expensive.
Unstructured nature makes coding,
analysis and interpretation very
 Depth interviews
An unstructured, direct, personal interview in
which a single respondent is questioned by a
highly skilled interviewer to uncover
underlying motivations, beliefs, attitudes, and
feelings on a topic.
The interviewer should: a) avoid appearing
superior; b) put the respondent at ease; c)be
detached and objective, yet personable; d)
ask questions in an informative manner; e)not
accept brief yes or no answers; f) probe the
respondent.
 Advantages of depth
interviews
Greater depth of insights.
Attribute response directly to the
respondent
It results in free exchange of
information
Immense applications especially
when probing is needed; sensitive
topics ; consumption experience is
sensory or complicated.
 Disadvantages of depth
interviews
Skilled interviewers are rare and
expensive
Very dependent on the interviewers
skills
Difficult to analyse and interpret data
High cost and lengthy process means
that the number of interviews will be
small and thus it might not be too
representative.
 Projective techniques
An unstructured and indirect form of
questioning that encourages the
respondents to project their
underlying motivations, beliefs,
attitudes or feelings regarding the
issues of concern.
In projective techniques, respondents
are asked to interpret the behavior of
others rather than describe their own
behavior.
 Common projective
techniques
Word association
Sentence completion
Picture response and cartoon test
(ballooning)
Role playing
Third person technique
Advantages of projective techniques

They increase the validity of the

responses by disguising the purpose
especially when the issues are
personal, sensitive or subject to
strong social norms.
They are also helpful when
underlying motivations, beliefs and
attitudes are operating at
subconscious levels.
 Disadvantages of projective
techniques
Highly trained interviewers; skilled
interpreters are needed.
Interpretation bias risk
Open ended and so the analysis is
subjective
The respondents might be unusual in
behavior and thus might not be
representative of the population.
 Other methods: Mystery
shopping
The trained observers pose as
consumers and shop at companys
stores or the competitors to collect
data about the customer employee
interaction and other marketing
variables such as price, display,
layout etc. They might or might not
buy the merchandise.
 Survey method
Survey is a structured questionnaire given
to a population sample and designed to
elicit specific information from
respondents.
Structured data collection: use of a formal
questionnaire that presents questions in a
pre-arranged order.
Fixed alternative questions: questions that
require respondents to choose from a set
of pre determined answers.
 Advantages of survey
Questionnaire is simple to
administer.
Data is reliable as alternatives are
fixed
Data variability is less
Coding, analysis and interpretation is
simple.
 Disadvantages of survey
Respondents might be unable or
unwilling to provide the desired
information.
Respondents might be unwilling to
respond if information is personal.
It might result in loss of validity if the
data is qualitative such as belief and
feelings.
Wording the questions is not easy.
 Classification of survey
methods
Survey: telephone , personal, mail,
electronic.
Telephone: traditional telephone,
computer assisted telephone
interviewing( CATI).
Personal: in home, mall intercept,
computer assisted personal
interviewing (CAPI).
Mail: mail/ fax interview, mail panel
Electronic: e-mail, internet
 CATI
It uses computerised questionnaire
administered over telephone.
The interviewer reads the question from
the screen and directly enters data in the
computer.
The software has built in logic which
enhances data accuracy.
It also has built in branching logic which
skips questions that are not relevant to the
respondent.
 CAPI
The respondent sits in front of the
computer terminal and answers a
questionnaire on the screen. The
interviewer is also present to serve as
host and guide the respondent as needed.
Complex questions can be asked ;
interviewer bias can be reduced; it offers
excellent control of data collection
environment and results in high response
rate.
 Criteria for selection of survey
methods
It is based on the relative merits and
demerits of the various methods;
For perishable foods, mall intercept
method is the best suited; thus we
should select method based on the
situation and the urgency;
There is no method which is
considered always superior to any
other.
 Improving survey response
rates
Prior notification
Incentives: monetary( prepaid or
promised) and non monetary
Follow up
Other facilitators
 Observation methods
Observation is the recording of
peoples behavioral patterns and of
objects and events in a systematic
manner to obtain information about
the phenomenon of interest.
The observer does not question or
communicate with the people being
observed.
Observation: personal and
mechanical
 Personal observation
An observational research method in which
human observers record the phenomenon
being observed as it occurs.
Humanistic inquiry is special form of
observation in which the researcher is
immersed in the system under study.
On-site observation is where the observers are
positioned in supermarkets and presented as
shoppers who need advice from another
shopper in making purchase decisions.
 Mechanical observation
An observational research method in which
mechanical devices, rather than human
observers record the phenomenon being
observed.
Nielsens peoplemeter; pretesting groups
people reader; turnstiles , traffic counters, UPC
scanners; on site cameras, etc.
Specialized equipments designed to monitor
heart beat and breathing rates, skin
temperatures, and other physiological changes
are also mechanical observation tools.
 Relative advantages among
observation methods
Method advantages disadvantages

Personal observation Most flexible. High observation

Highly suitable in bias.
natural settings. High analysis bias.
Mechanical Low observation bias. Can be intrusive.
observation Low to medium Not always suitable in
analysis bias. natural settings.
Relative advantages of observation
methods over survey
They permit measurement of actual
behavior.
There is no reporting bias.
There is less potential for interviewer
bias.
Certain types of data can be
obtained best by observation.
 Relative disadvantages of
observation methods
Very little can be infered about
motives, beliefs, attitudes and
preferences.
There is potential for observer bias.
Most methods are time consuming
and expensive.
It is difficult to observe some form of
behavior.
There is potential for unethical
behavior.
 Concept of Causality( X
causes Y)
Ordinary meaning Scientific meaning
X is the only cause of Y X is only one of a number of
possible causes of Y.
X must always lead to Y. The occurrence of X makes the
occurrence of Y more probable.
It is possible to prove that X is a We can never prove that X is a
cause of Y. cause of Y. at best, we can infer
that X is a cause of Y.
The scientific meaning is more
appropriate to MR. Marketing
effects are caused by multiple
variables, and the relationship
between cause and effect tends
to be probabilistic.
 Conditions for causality
Concomitant variation
Time order of occurrence of variables
Elimination of other causal factors

These conditions are necessary but not

sufficient to demonstrate causality.
 Concomitant variation
A condition for inferring causality,
which requires that a cause, X and an
effect Y, occur together or vary
together as predicted by the
hypotheses under consideration.
Eg: sales(Y) and in-store service(X).
Even we find data on 100 stores and
find these two things are always
together, we can say the hypotheses
is tenable, still we cant prove it.
 Time order of occurrence of
variables
The causing event must occur either
before or simultaneously with the
effect; it cannot occur afterward.
However, it is possible that both the
events to be cause and effects of
each other.
This time order if not followed, is
enough to prove that causality does
not exist and not the other way
round.
 Absence of other possible
factors
The factor or variable being
investigated should be the only
possible causal explanation. In store
service may be the cause of sales if
can be sure that other factors like
pricing, advertising, level of
distribution, product quality,
competition etc are held constant or
controlled.
 Experimentation
Experimentation is a research technique used
in causal research. It is the primary method
for establishing cause and effect relationships
in marketing.
Field experiment is conducted in actual
market conditions, that is in natural
environment.
Laboratory experiment is conducted in
artificial environment in which the researcher
constructs the desired conditions.
 Definitions and concepts
Independent variables : variables that are
manipulated by the researcher and whose effects
are measured and compared. Eg: price levels,
package designs, advertising themes.
Test units: individuals, organisations, or other
entities whose responses to the independent
variables or treatments are being examined. They
may include consumers, stores, or geographic
areas.
Dependent variables: variables that measure the
effect of the independent variables on the test units.
They may include sales, profits and market shares.
 Definitions..contd..
Extraneous variables: variables other than the
independent variables that affect the response of the
test units. These variables can confound the dependent
variable measures in a way that weakens or invalidates
the results of the experiment. Eg: store size, location
and competitive effort.
Experiment Group( EG): The group exposed to the
manipulated independent variable. The results of this
might be compared with another EG or CG.
Control group(CG): The group that is not exposed to the
independent variable manipulation. It provides a point
comparison when examining the effects of the
manipulations on the dependent variables.
 Definitions..contd.
Random assignment: the process of randomly
assigning test units and treatment to the
experimental and control groups. It is one of
the most common techniques used to control
for the effect of extraneous variables on the
dependent variable.
Experiment: the process of manipulating one or
more independent variables and determining
their effect on one or more dependent variables
measured on the test units, while controlling for
the extraneous variables.
 Experimental design
An experimental design is a set of
procedures specifying a) the test
units and how these units are to be
divided into homogeneous
subsamples; b) what independent
variables or treatments are to be
manipulated; c) what dependent
variables are to be measured; d) how
the extraneous variables are to be
controlled.
 Validity in experimentation
Internal validity : it is measure of the accuracy
of an experiment. It measures if the
manipulation of the independent variables, or
treatments actually caused the effect on the
dependent variables.
External validity: whether the cause and effect
relationships found in the experiment can be
generalised.
It is desirable to have an experimental design
having both kinds of validity but in MR, we often
have to trade one type of validity for other.
 Classification of experimental
designs
Pre-experimental True experimental Statistical design
design design
One shot case study Pretest-postest Factorial design
control group
One group pretest- Posttest only control
posttest group
Static group
 Pre-experimental designs
they do not control extraneous factors by
randomisation.
One shot case study: a single group of test units is
exposed to treatment X, and then a single
measurement on the dependent variable is taken.
Test units are not assigned random.
One group pre-test post test study: a group of test
units is measured twice, before and after exposure to
the treatment. Test units are not assigned at random.
Static group: there are 2 groups; the EG and CG.
Measurement on both groups are made only after the
treatment, and test units are not assigned at random.
 True experimental designs
The researcher can randomly assign test units
to experimental groups and also randomly
assign treatments to experimental groups.
Pretest-posttest control group: the EG is
exposed to the treatment and the CG is not.
There is random assignment and pre test and
post test measures are taken on both groups.
Post test only Control group: no pretest measure
is taken. There is random assignment and post
treatment measures are taken on both EG and
CG.
 Statistical designs
Statistical designs allow for the
statistical control and analysis of
extraneous variables. Several basic
experiments are conducted
simultaneously. The advantages of
these are:
a)The effect of more than one
independent variable can be
measured.
b) Specific extraneous variables can be
 Factorial designs
It is used to measure the effects of two or
more independent variables of various
levels and to allow for interactions between
variables. It can be thought of as a table. In
a 2 factor design, each level of one variable
represents a row, and each level of another
variable represents a column.
Interaction: when the simultaneous effect of
two or more variables is different from the
sum of their separate effects.
 An example of factorial
design
Amount of No humour Medium High Humour
brand humour
information
low
medium
high
 Measurement and scaling
Measurement is the assignment of
numbers or other symbols to
characteristics of objects according
to certain set rules. It permits
statistical analysis of the resulting
data.
Scaling is generation of a continuum
upon which measured objects can be
located.
 Primary scales of
measurement
Nominal scale
Ordinal scale
Interval scale
Ratio scale
 Nominal scale
A scale whose numbers serve only as
labels or tags for identifying and
classifying objects, with a strict one
to one correspondence between the
numbers and the objects when used
for identification. The classes are
mutually exclusive and collective
exhaustive. Eg: roll num, gender, t
shirt numbers in sports etc.
 Ordinal scale
A ranking scale in which numbers are
assigned to objects to indicate the
relative extent to which some
characteristic is possessed. Thus, it is
possible to determine whether an
object has more or less of a
characteristic than another object.
Eg: ranking of companies, b-schools
etc.
 Interval scale
A scale in which numbers are used to
rate objects. The numerically equal
distances on the scale represent
equal distances in the characteristic
being measured. The zero is not
defined; the scale point of 6 and 3
does not mean it is double value; eg:
temperature scale; questionnaire
points from 1 to 5.
 Ratio scale
The highest measurement scale
conveying the most information. It
allows the researcher to identify or
classify objects, rank order the
objects, compare intervals or
differences. It is also meaningful to
compute ratios. of scale values. Here
the scale value of 6 and 3, 6 will be
double of 3. The zero is defined on
scale. Eg: ruler, age, weight etc.
Primary scale Basic Common Marketing
characteristic examples examples
s
Nominal scale Numbers Social security Brand numbers,
identify and numbers, sports store types
classify objects etc
Ordinal scale Numbers Quality Preference
indicate the rankings; rankings,
relative Ranking of market
positions of the teams in a positions, social
objects but not tournament; classes.
the magnitude
of differences.
Interval scale Differences Temperature Attitudes,
between the (fahrenheit, opinions, index
objects can be celsius) numbers
compared; zero
point is
arbitrary.
Ratio scale Zero point is Length, weight Age, income,
 Classification of scaling
techniques
Scaling tech: Comparative and non-
comparative scales.
Comparative: Paired comparison;
rank order; constant sum scales
Non comparative: Continuous and
itemized scales.
Itemized : Likert scale; semantic
differential; stapel scale.
 Comparative and non
comparative
Comparative scales : in which there
is a direct comparison of stimulus
objects with one another.
Non comparative scales: in which
each object is scaled independently
of each other in the stimulus set.
 Paired comparison scaling
A comparative scaling in which a
respondent is presented with two
objects at a time and asked to select
one object on the pair according to
some criterion. The data obtained is
ordinal in nature.
 Rank order scaling
A comparative scaling technique in
which respondents are presented
with several objects simultaneously
and asked to order or rank them
according to some criterion. Eg:
ranking of the most admired brands;
etc. the data is ordinal in nature.
 Constant sum scaling
A comparative scaling technique in
which respondents are required to
allocate a constant sum of units such
as points, dollars etc among a set of
stimulus objects with respect to
some criterion. Sometimes, the data
output can be taken as interval data
but in most cases the data is taken
as ordinal in nature.
 Non comparative scales
Scale Basic examples advantages disadvant
characteris age
tics
Continuous Place a mark Reaction to Easy to Scoring is
rating on tv construct cumberso
continuous advertiseme me
line nts
Itemised
rating scales
Likert scale Degree of Measuremen Easy to More time
agreement t of attitudes construct, consuming
from 1 to 5 administer,
understand
Semantic 7 point scale Brand, versatile Difficult to
differential with bipolar product and construct
labels company
images
 Error in scales
Measurement error: the variation in the
information sought by the researcher and the
information generated by the measurement
process employed.
Systematic error: error that affects the
measurement in a constant way; stable factors
that affect the observed score in the same way
each time the measurement is made.
Random error: measurement error that arises
from random changes or differences in
respondents or measurement situations.
 Scale evaluation
Reliability
Validity
 reliability
It refers to the extent to which a
scale produces consistent results if
repeated measurements are made
on the characteristic.
Systematic error do not have any
impact on reliability.
Reliability can be defined as the
extent to which the measure is free
from random error.
 Validity
The extent to which differences in
observed scale scores reflect true
differences among objects on the
characteristic being measured rather
than systematic or random error.
Perfect validity means there is no
measurement error.
All valid are reliable but all reliable
need not be valid!!
 Choosing a scaling
technique
In addition to theoretical considerations
and reliability and validity concept,
there are some practical factors too
which must be kept in mind for selecting
scaling technique for a MR problem:
Level of information desired
Capabilities of respondents
Context
cost/ time and scales used in the past.