mimd lexicon[1]

Lexicon of Management Information for Marketing Decisions

Economy is the energy of all people in motion with one serving the other without

superiority and subordination.

Markets are people and markets live in the minds of People.

By Anone

Source : N. K Malhotra Reproduced by : Premchandrahas Sastry 1

Contents Page No

Definition 3

Classification of Marketing Research 4

Details of Problem Solving Research 5

Marketing Research Process 6

Role of Marketing Research in Marketing Decision Making 8

Decision to conduct marketing Research 10

Defining a marketing research problem and developing an

approach

12

Environment context of the problem 14

Management Decision Problem & Marketing Research

Problem

16

Defining the Marketing Research Problem 17

Components of the Approach 20

Research Design 23

Comparison of Research Designs 25

Potential Sources of Error 29

Flow chart of Errors 30

Marketing Research Proposal 31

Example of Data from Syndicated Firms 33

Exploratory Research Design 34

Advertising Evaluation 36

Media Panels 37

Types of Marketing Research & Degree of Problem Definition 38

Data Types in Marketing Research Process 39

Criteria for Evaluating Secondary Data 40

Overview of Syndicated Services 41

Primary Research Process 43

Focus Group Vs Depth Interviews 45

Projective Techniques 46

Descriptive Research Design 49

Causal Research Design 52

Classification of Experimental Designs 59

Sampling 64

Sampling Process & Numericals 76

Measurement & Scaling 81


Value of information

Creativity and Common Sense Command a premium in the Marketing Research

1) Definition

AMA ( American Marketing Association ) defines

Marketing Research is the function which links consumers, customers, public to the marketer

through the information used to identify and define marketing opportunities and problems ,

generate , refine and evaluate marketing actions; monitor marketing performance and improve

understanding of marketing as a process.

2)Marketing Research specifies the information required to address these issues ,designs the

method for collecting information, manages and implements the data collection process, analyzes

the results and communicates the findings and their implications.

3)Marketing Research is the systematic and objective identification ,collection, analysis,

dissemination and use of information for the purpose of decision making related to identification

and solution of problems and opportunities in marketing.


Classification of Marketing Research

Organizations engage in marketing research for 2 reasons

1) to identify problems

2) to solve marketing problems

so we have problem identification research

problem solving research

Problem identification research : identify problems that are not on the surface ,but likely to arise

in the future.

Examples : market potential, market share, brand or company image,

Market characteristics, sales analysis, short range forecasting,

Long range forecasting and business trends research.

A survey of companies conducting market research –

97% for market potential, market share & market characteristics research

90% problem identification research. This type of research provides information about the

marketing environment and helps diagnose a problem.

Eg : declining market potential indicates that the firm is likely to have a problem achieving its

growth targets.


Once problem is identified – problem solving research is undertaken to arrive at a solution.

Details of Problem solving research

Segmentation Research

Determine basis of segmentation

Establish market potential and responsiveness for various segments

Select target markets and create lifestyle profiles, demography, media and product image

characteristics.

Product Research

Test concept

Optimal product design

Package tests

Product modification

Brand positioning and repositioning

Test marketing

Control store tests.

Pricing Research

Importance of price in brand selection

Pricing policies

Product line pricing

Price elasticity of demand

Response to price changes

Promotional Research

Optimal promotional budget

Sales promotion relationship

Optimal promotional mix

Copy decisions

Media decisions

Creative advertising testing

Claim substantiation


Evaluation of advertising effectiveness

Distribution Research

Type of distribution ,Attitudes of channel members, Intensity of wholesale and retail coverage

Channel margins , Location of retail and wholesale outlets

The marketing Research Process

Consists of 6 steps

1) Problem definition

Researcher should take into account

The purpose of the study

Relevant background information

The information needed & how it will be used in decision making.

This would further involve discussion with the decision makers, interviews with

Industry experts , analysis of secondary data and perhaps some qualitative research such as

focus groups.

Once the problem has been precisely defined the research can be designed and conducted

properly.

2) Development of approach to the problem

Includes formulating of an objective or theoretical framework, analytical models, research

questions, hypothesis and identifying the information needed. This process is guided by

discussions with management and industry experts, analysis of secondary data, qualitative

research, pragmatic considerations

3) Research Design Formulation

A research design is a framework or blueprint for conducting the marketing research project . It

details the procedures necessary for obtaining the required information and its purpose is to

design a study that will test the hypotheses of interest , determine possible answers to the

research questions and provide the information needed for decision making.


Conducting exploratory research ,

Precisely defining the variables

Designing the appropriate scales to measure them are also part of the research design

The issue of how the data should be obtained from the respondents ( by conducting a survey or an

experiment) must be addressed. It is also necessary to design a questionnaire and sampling plan

to select respondents for the study.

Formulating the research design involves the following steps:

1) definition of the information needed

2) secondary data analysis

3) qualitative research

4) methods of collecting quantitative data( survey, observation, and experimentation)

5) measurement and scaling procedures

6) questionnaire design

7) Sampling process and sample size

8) Plan of data analysis

4) Field work or Data Collection

Data collection involves a field force or staff that operates either in the field , as in the case of

personal interviewing inhome , mall intercepting or computer assisted personal interviewing )

from an office by telephone or computer assisted telephone interviewing ,Through mail

traditional mail and mail panel surveys with pre recruited households) or electronically( email

or

internet) . Proper selection, training, supervision, and evaluation of the field force help minimize

data collection errors.

5) Data preparation and analysis

Data preparation includes the editing , coding , transcription and verification of data. Each

observation ,questionnaire is inspected or edited and if necessary corrected .

6) Report Preparation and Presentation

The entire project should be documented in a written report that addresses the specific research

questions identified ; describes the approach, the research design, data collection and data

analysis procedures adopted and presents the results and the major findings. The findings should


be presented in a comprehensible format so that management can readily use them in the decision

making process.

In addition an oral presentation should be made to management using tables, figures, and graphs

to enhance clarity and impact

The role of marketing research in marketing decision making.

Can be understood with the help of a basic marketing paradigm

The emphasis in marketing is on the identification and satisfaction of customer needs. In order to

determine customer needs and to implement marketing strategies and programs aimed at

satisfying those needs, marketing managers need information. They need information about

customers , competitors , and other forces in the marketplace.

As firms have become national and international , more need for market research and

information on distant markets has increased.

Source : N. K Malhotra Reproduced by : Premchandrahas Sastry

Marketing Research

Customer GroupsConsumersEmployeesShareholdersSuppliers

Controllable marketing variablesProductPricingPromotionDistribution

Uncontrollable environmental factorsEconomyTechnologyCompetionLaws & RegulationsSocial & cultural factorsPolitical factors

Assessing Information needs

Providing Information

MarketingDecisionMaking

Marketing Managers

Market SegmentationTarget Market SelectionMarketing ProgramsPerformance Control

8

As consumers have become more affluent and sophisticated, marketing managers need better

information on how they will respond to products and other marketing offerings.

As competition has become more intense , managers need information on the effectiveness of their

marketing tools.

As environment changes more rapidly , marketing managers need more timely information.

The task of marketing research is to assess the information needs and provide management with

relevant , accurate , reliable , valid , current and actionable information.

Today’s competitive marketing environment and the ever increasing costs attributed to poor

decision making require marketing research to provide sound information. Sound decisions are

not based on gut feeling, intuition or even pure judgment.

Marketing Research & Competitive Intelligence

Competitive intelligence is a crucial part of the emerging knowledge economy.

By analyzing rivals moves CI allows companies to anticipate market developments rather than

merely react to them.

Although marketing research plays a central role in the collection, analysis and dissemination of

CI information . CI has evolved into a discipline of its own. The society of competitive

Intelligence professionals ( SCIP) consists of members conducting CI for large and small

Companies providing management with early warning of changes in the competitive landscape.


Decision to conduct Market Research

Decision to conduct market research is not automatic.

This decision is guided by a number of considerations

1) Cost versus the benefits

2) The resources available to conduct the research

3) The resources available to implement the research findings

4) Management attitude towards research

Marketing research should be conducted when the expected value of information it generates

exceeds the costs of conducting the marketing research project. Formal procedures are available

for quantifying the expected value as well as the costs of a marketing research project.

Reasons for not conducting research

1) it is better not to conduct a research than undertake one in which the integrity of research is

compromised because of lack of resources.

2) a firm a lack resources to implement the recommendations arising from the findings of

marketing research.

3) If management does not have positive attitude towards research , then is is likely the project

will gather dust after the project is conducted.


Marketing Research Industry Org chart of VNU

External Suppliers : Outside firms hired to supply marketing research data


Full Service Suppliers : Offer entire range of marketing research services from problem

definition , approach development, questionnaire design, sampling data collection, data analysis

and interpretation , to report preparation and presentation.

Syndicated Services : collect information of known commercial value that they provide to

multiple clients on a subscription basis. Surveys , panels , scanners , and audits are the main

means by which these data are collected.

Chapter 2 :

Defining a market research problem & Developing an Approach

Of all the tasks in a marketing research project, none is more vital to the ultimate fulfillment of a

client’s needs than a proper definition of the research problem .

In adequate problem definition is a leading cause of failure of marketing research projects.


i) Discussion with Decision Makers : extremely important to interact with the

Key decision makers

The DM needs to understand the capabilities & limitations of the Research

Research provides information relevant to management decisions ,but it cannot provide solutions

Because solutions require managerial judgement.

The researcher also needs to understand the nature of the decision that manager’s face & what

they hope to learn from the research

The researcher should treat the underlying causes not merely address the symptoms.

ii) Interview with industry Experts : individuals have knowledge of the firms can help formulate

the marketing research problem , for products of technical nature

iii) Secondary Data Analysis : Secondary data is economical and quick source of the

background information .

Analysis of the secondary data is essential in the marketing problem research

iv) Qualitative Research : If all the above are not enough to define the research

problem then Qualitative research is undertaken to understand the problem.

It is unstructured , exploratory in nature , based on small samples & may use techniques such as

Focus Groups , word associations (asking respondents to indicate their first response to word

Stimulus), pilot surveys , case studies etc. Research is not conducted in a formal way but it can

provide valuable insights into the problem.


Environment context of the problem

To understand the background of a marketing research problem , the researcher must understand the

client’s firm and industry.

Factors that have impact on the definition of the marketing research problem.

Eg :

past information : forecasts & trends with respect to sales , market share, profitability ,technology,

population , demographics, & lifestyle can help researcher understand the underlying marketing

research problem.

forecasts pertaining to the industry and the firm , resources and constraints of the firm, objectives of

the decision maker , buyer behavior, legal environment, economic environment, and marketing ,

technological skills of the firm .

Resources & Constraints : To formulate and scope a marketing Research problem it is necessary to

take into consideration both money and research skills and other constraints such as cost & time.

Objectives : Decisions are made to accomplish objectives.

The formulation of the management decision problem must be based on clear understanding of two

types of objectives

1) the organizational objectives ( the goals of the organization )

2) the personal objectives of the decision maker.

The researcher needs skills to extract the objectives of the management.

Buyer Behavior : It is the central component of the environmental context. In most marketing

decisions , the problem can ultimately be traced to predicting the response of buyers to specific

actions by the marketer. An understanding of the underlying buyer behavior can provide valuable

insights into the problem.


Factors ( Buyer Behavior)

The number and geographical location of the buyers and non buyers

Demographic and psychological characteristics

Product consumption habits and the consumption of related problem categories

Media consumption behavior and response to promotions

Price Sensitivity

Retail outlets patronized

Buyer preferences

Legal Environment :

Public policies, laws, government agencies and pressure groups that influence and regulate various

organizations and individuals in society. Important areas of laws include patents , trademarks,

royalties, trade agreements, taxes and tariffs.

Federal laws have an impact on each element of the marketing mix

Laws have been passed to regulate specific industries

The legal environment can have an important bearing on the definition of the marketing research

problem.

Economic Environment :

Comprises of purchasing power, gross income , disposable income, discretionary income , prices ,

savings , credit availability and general economic conditions.

The general state of economy ( rapid growth , slow growth , recession or stagflation) influences the

willingness of consumers and businesses to take on credit and spend on big –ticket items. Thus the

economic environment can have important implications for marketing research problems.

Marketing & Technological Skills :


Many of the factors to be considered in the environmental context of the problem can be researched

via the internet.

After gaining and adequate understanding of the environmental context of the problem , the

researcher can define the management decision problem and the marketing research problem.

Management Decision Problem & Marketing Research Problem

The management decision problem asks what the DM needs to do , whereas the marketing research

problem asks what information is needed and how it can best be obtained.

The management decision problem is action oriented

The marketing research problem is information oriented

The management decision problem focuses on symptoms , the marketing research problem focuses

on underlying causes.


Defining The Marketing Research Problem

The general rule to be followed in defining the marketing research problem is that the

definition should

1) allow the researcher to obtain all the information needed to address the

management decision problem

2) guide the researcher in proceeding with the project .

Researchers make two common errors in problem definition :

1) Research problem is defined too broadly ( 1st type of Error)

A broad definition does not provide clear guidelines for the subsequent steps involved in the project.

eg: i) develop a marketing strategy for the brand

ii) improve the competitive position of the firm

iii) improve the company’s image

All the above defined research problems are not specific enough to suggest an approach to the

problem or a research design.

2) The marketing research problem is defined too narrowly.(2nd Type of Error)

This can prevent the researcher in addressing some important components of the

Management Decision problem

Eg : i) decrease the price of the firms brand to match the competitors price cut

ii) maintain price but increase advertising heavily

The likelihood of committing either type of error in problem definition can be reduced by

stating the marketing research problem in broad , general terms and identifying its specific

components.

Example : In a project conducted for a major consumer products firm, the management problem was

how to respond to a price cut initiated by a competitor.


The alternative courses of action initially identified by the firm’s research staff were

1) decrease the price of the firm’s brand to match the competitor’s price cut

2) maintain price but increase advertising heavily

3) decrease the price somewhat, without matching the competitor’s price ,and

moderately increasing advertising.

None of the above alternatives seemed promising

When outside marketing research experts were brought in , the problem was redefined as improving

the market share and profitability of the product line.

Qualitative research indicated that in blind tests consumers could not differentiate products offered

under different brand names.

Further more consumers relied on price as an indicator of product quality .

These findings led to a creative alternative: increase the price of the existing brand and introduce

two new brands – one priced to match the competitor and the other priced to undercut it. This

strategy was implemented leading to an increase in market share and profitability.

The likelihood of committing either type of error in problem definition can be reduced by stating the

marketing research problem in broad, general terms and identifying its specific components.

The broad statement provides perspective on the problem and acts as a safeguard against

committing the second type of error .

The specific components focus key aspects of the problem and provide clear guidelines on how to

proceed further , thereby reducing the likelihood of the first type of error.


Examples of problem research for Sears :

i) what criteria do households use when selecting department stores ?

ii) How do households evaluate sears and competing stores in terms of the

Choice criteria identified in question 1 ?

iii) Which stores are patronized when shopping for specific product categories ?

iv) What is the market share of Sears and its competitors for specific product categories ?

v) What is the demographic and psychographic profile of the customers of Sears ?

Does It differ from the profile of customers of competing stores ?

vi) Can store patronage and preference be explained in terms of store evaluation and

customer characteristics ?


Management Research Problem

Broad Statement

Specific components

Component 1 Component 2 Component 3

19

Components of the approach

While developing the approach, we must not lose sight of the goal – the outputs . The outputs of the

research development process should include the following components : objective/ theoretical

framework , analytical models, research questions, hypothesis and specification of information

needed.

Objective/Theoretical framework :

It is gathered by compiling relevant findings from secondary sources . The researcher can rely on

the theory; to determine which variables should be investigated.

Further more theoretical considerations provide information on how the variables should be

operational zed and measured as well as how the research design and sample should be selected.

After all theories have been formulated after several experimentations and observations .

However applying theory to the research problem requires lot’s of creativity.

A theory may not specify adequately how its abstract constructs ( variables) can be embodied in a

real world phenomenon. Moreover theories are incomplete.

They deal with only a subset of variables that exist in the real world.

Hence the researcher must also identify and examine other non theoretical variables.


Analytical Model : a set of variables & their interrelationships designed to represent in whole or in

part some real system or process .

In verbal models : the variables and their relationships are stated in prose form. Such models may

be mere restatements of the main tenets of a theory.

Graphical models :are visual. They are used to isolate variables and to suggest directions of

relationships but are not designed to provide numerical results. They are logical preliminary steps to

developing mathematical models.

Model building : the consumer experiences the stores and evaluates the stores in terms of factors

comprising the choice criteria. If preference is strong enough , the consumer will patronize the store.

Verbal model : A consumer first becomes aware of a department store . That person then gains an

understanding of the store by evaluating the store in terms of the factors comprising the choice

criteria. Based on the evaluation , the consumer forms a degree of preference for the store. If the

preference is strong enough, the consumer will patronize the store.

Graphical model :

Mathematical Model :

n y = a0 + ∑ ai xi i=1 where y = degree of preference

a0 xi=model parameters to be estimated statistically xi=store patronage factors that constitute the choice criteria.


Patronage

Preference

Understanding : Evaluation

Awareness

21

The phenomenon of store patronage stated verbally is represented for clarity through a

figure(graphical model)and is put in equation form (mathematical model) for ease of statistical

estimation and testing. Graphical models are particularly helpful in conceptualizing an approach to

the problem.

In Harley Davidson example , the underlying theory was that brand loyalty is the result of positive

beliefs, attitude ,affect and experience with the brand. This theory may be represented by the

following graphical model.

Research Questions :

Research questions (RQs )are refined statements of the specific components of the problem. Although

the components of the problem define the problem in specific terms, further detail may be needed to

develop an approach.

RQs ask what specific information is required with respect to the problem components.

The formulation of the research questions should be guided not only by the problem definition but

also by the theoretical framework and the analytical model adopted.

Hypothesis : A hypothesis is an unproven statement or proposition about a factor or phenomenon

that is of interest to the researcher.


Attitudes

Purchase

Experimentation/ Evaluation

Repeat/ Purchase

Loyalty

Affect

22

Beliefs

Example :

H1 : customers who are store loyal are less knowledgeable about the shopping environment

H2 : Store loyal customers are more risk averse than are non loyal customers.

Research Design : It is a framework or blueprint for conducting the marketing research project. It

details the procedures necessary for obtaining the information needed to structure or solve marketing

research problems.

A good research design will ensure that the marketing research project is conducted effectively and

efficiently.


Components of the Marketing Research

Problem

Research Questions

Hypothesis

Research Design

Exploratory Research Design

Conclusive Research Design

Descriptive Research

Causal Research

Cross sectional Design

Longitudinal Design

Single Cross-sectionalDesign

Multiple Cross- sectionalDesign

23

Objective /theoretical framework

Analytical Model

Exploratory Research : To provide insights into and an understanding of the problem confronting

the researcher. It is used when you must define the problem more precisely, identify relevant courses

of action, or gain additional insights before an approach can be developed.

Insights gained from exploratory research might be verified or quantified by conclusive research as

in the opening example.

Conclusive research : It is typically more formal and structured than exploratory research. It is

based on large , representative samples and the data obtained are subjected to quantitative analysis.

The findings from this research are considered to be conclusive in nature that they are used as input

into managerial decision making. ( however from the philosophy of science nothing can be proven

and nothing is conclusive )

Exploratory Conclusive

Objective To provide insights and understanding To test specific hypothesis and examine relationships

Characteristics Information needed is only defined loosely. Research process is flexible and unstructured. Sample is small and non representative. Analysis of primary data is qualitative

Information needed is clearly defined. Research process is formal and structured. Sample is large and representative . Data analysis is quantitative.

Findings/Results: Tentative Conclusive

Outcome: Generally followed by further exploratory or conclusive research

Findings used as input into decision making

Comparison of Basic Research Designs

Exploratory Descriptive Causal

Objective Discover ideas & Insights Describe market characteristics or functions

Determine cause and effect relationships

Characteristic Flexible , versatile, Often the front end of total research design

Marked by prior formulation of specific hypothesis

Manipulation of one or more independent variables. Control of other meditating variables

Methods Expert surveys, Pilot surveys, Secondary data(analyzed

Secondary data( analyzed quantitatively)Surveys, Panels,

Experiments


qualitatively)Qualitative research

Observational and other data

Eg for Descriptive Research : Describe the characteristics of relevant groups , such as consumers,

sales people , organizations , market areas. “ profile of heavy users of prestigious department stores

like Saks Fifth Avenue , Neiman Marcus “ , determining perception of product characteristics, degree

to which marketing variables are associated, “ to what extent is shopping at department stores

related to eating out ?

Cross sectional Designs : Most frequently used descriptive design in marketing research . Involve

the collection of information from any given sample of population elements only once. Can be single

cross sectional or multiple cross sectional.

Single Cross Sectional Design : Only one sample of respondents is drawn from the target

population and information is obtained from the sample only once. Also called as sample survey

research design.

Multiple cross sectional Design : There are two or more samples of respondents and information

from each sample is obtained only once.

Often information from different samples is obtained at different times over long intervals. Multiple

cross sectional designs allow comparisons at the aggregate level but not at the individual respondent

level. Because a different sample is taken each time a survey is conducted , there is no way to

compare the measures on an individual respondent across surveys . One type of multiple cross-

sectional design of special interest is cohort analysis.

Cohort Analysis : Consists of series of surveys conducted at appropriate time intervals , where the

cohort serves as the basic unit of analysis.

A Cohort is a group of respondents who experience the same event within the same time interval.

The term cohort analysis refers to any study in which there are measures of some characteristics of

one or more cohorts at two or more points in time.

Examples : Used to predict changes in voter opinions during a political campaign.

Cohort is a group of persons with common statistical characteristic

For example the age cohort of people between 8 and 19 years old was selected and their soft drink

consumption was examined every 10 years for 30 years.


In other words , every 10 years a different sample of respondents was drawn from the population of

those who were then between 8 and 19 years old. This sample was drawn independently of any

previous sample drawn in this study from the population of 8 to 19 years old.

Obviously people who were selected once were unlikely to be included again in the same

age cohort(8 to 19 years old ) as these people would be much older at the time of subsequent

sampling. This study showed that the cohort had increased consumption of soft drinks over time.

Longitudinal Designs : In longitudinal designs a fixed sample of population elements is measured

repeatedly on the same variables.

A longitudinal design differs from a cross sectional design in that the sample or samples remain the

same over time.

Longitudinal data enables researchers to examine changes in the behavior of individual units and to

link behavioral changes to marketing variables , such as changes in advertising , packaging, pricing

and distribution.

In other words the same people are studied over time and the same variables are measured

The term panel is used interchangeably with the term longitudinal design. A panel consists of a

sample of respondents , generally households that have agreed to provide information at specified

intervals over an extended period .

Syndicated firms maintain panels and panel members are compensated for their participation with

gifts , coupons ,information or cash.

Longitudinal Design Cross Sectional Design

Gives snapshot of the variables of interest at a single point in time

Provides a series of pictures that gives an in depth view of the situation and the changes that take place over time

How did the American people rate the performance of George Bush immediately after the war in Afghanistan

How did the American people change their view of bush’s performance during the war in Afghanistan.

Advantages & Disadvantages of Longitudinal and Cross Sectional Designs

Evaluation Criteria Longitudinal Design Cross Sectional Design

Detecting change - +

Large amount of data

collection

- +

Accuracy - +


Representative

Sampling

+ -

Response Bias + -

+ : indicates relative advantage over the other

- : indicates relative disadvantage

Causal Research :

To obtain evidence of cause and effect (causal) relationships . Marketing managers continually make

decisions based on assumed causal relationships. These assumptions may not be justifiable and the

validity of the causal relationships should be examined via formal research.

Eg : The common assumption that a decrease in price will lead to increased sales and market share

does not hold in certain competitive environments.

Causal research are appropriate for the following purposes

1) to understand which variables are the cause(independent variables) and which variables are

the effect ( dependent variables) of a phenomenon

2) To determine the nature of the relationship between the causal variables and the effect to be

predicted.

Like descriptive research causal research requires a planned and structured design.

Although descriptive research can determine the degree of association between variables, it is not

appropriate for examining causal relationships. Such an examination requires a controlled

environment. A relatively controlled environment is one in which the other variables that may effect


the dependent variable are controlled or checked as much as possible. The effect of this manipulation

on one or more dependent variables is then measured to infer causality.

The main method of causal research is experimentation

Potential Sources of ERROR

Several potential sources of error can affect a research design. A good research design attempts to

control the various sources of error.

The total error is the variation between the true mean value in the population of the variable of

interest and the observed mean value obtained in the marketing research project.

Eg : avg annual income of target population from census records is $75,871

Marketing research project estimates it as $67,157

Total error is composed of random sampling error and non sampling error

Random sampling Error : Occurs because the particular sample selected is an imperfect

representation of the population of interest. It is the variation between the true mean of the

population and the true mean for the original sample.


Non Sampling Error: Can be attributed to sources other than sampling and they may be random or

non random . They result from a variety of reasons , including errors in problem definition,

approach, scales , questionnaire design, interviewing methods, data preparation and analysis.

Example : researcher designs a poor questionnaire which contains several questions that lead the

respondents to give biased answers. Non sampling errors consist of non response errors and

response errors.

Flow Chart of Errors


Total Error

Random sampling error

Non Sampling Error

Response Error

Non Response Error

Researcher Error Interviewer Error Respondent Errors

29

*Surrogate Information Error*Measurement Error*Population Definition Error*Sampling Frame Error*Data Analysis Error

* Respondent Selection Error* Questionnaire Error* Recording Error* Cheating Error

*Inability Error*Unwillingness Error

Non Response Error : When some of the respondents included in the sample do not respond

Response Error : When respondents give inaccurate answers or their answers are misrecorded

or misanalyzed.

Surrogate information error : Variation between the information needed for the marketing research

problem and the information sought by the researcher

Measurement Error : Variation between the information sought and the information generated by

the measurement process employed by the researcher. Eg : While seeking to measure consumer

preferences, the researcher employs a scale that measures perceptions than the preferences.

Population Definition Error : Variation between the actual population relevant to the problem at

hand and the population as defined by the researcher.

Sampling frame error : Variation between the population defined by the researcher and the

population as implied by the sampling frame(list) used.

Eg : the telephone directory used to generate a list of telephone numbers does not accurately

represent the population of potential consumers because of unlisted, disconnected and new numbers

in service.

Data Analysis Error : Errors that occur while raw data from questionnaires are being transformed

into research findings. Eg : in appropriate statistical procedure.

Respondent Selection Error : When interviewers select respondents other than those specified by

the sampling design or in a manner inconsistent with the sampling design.

Questioning Error : Errors made in asking questions of the respondents or in not probing when

more information is needed.

Recording Error : Errors in hearing, interpreting and recording the answers given by the

respondents

Cheating Error : when the interviewer fabricates answers to part or all of the interview.

Inability Error : Respondents inability to provide accurate answers, due to unfamiliarity, fatigue,

boredom, faulty recall , question format, question content, other factors.


Unwillingness Error : Respondents unwillingness to provide accurate information. Respondents may

intentionally misreport their answers because of a desire to provide socially acceptable answers ,

avoid embarrassment or please the interviewer. For example a respondent intentionally misreports

reading the Time magazine to impress the interviewer.

Non Sampling Error is likely to be more problematic than sampling Error.

After the research design , appropriately controlling the total error has been specified , the budgeting

& Scheduling decisions can be made.

Budgeting and scheduling the project : Ensures that the marketing research project is completed

with in the available resources – financial , time , personnel .

One can use CPM( Critical Path Method) or PERT( Program Evaluation & Review Technique)

, GERT ( graphical evaluation & review technique)

Marketing Research Proposal :

After the research design is formulated , budgeting & scheduling is accomplished , a written research

proposal is made.

The proposal contains the essence of the project and serves as a contract between the researcher and

management. Covers all the phases of the marketing research process .

Describes the Research Problem , The approach , the research design , and how the data will be

analyzed collected and reported. Gives cost estimate and time schedule for completing the project.

The elements of the research Proposal :

1) Executive Summary : The proposal should begin with a summary of the major points from

each of the other sections, presenting an overview of the entire proposal.

2) Background : The background to the problem , including the environmental context

should be discussed.

3) Problem Definition / Objectives of the Research : Normally a statement of the problem ,

including the specific components, should be presented. If the problem statement has not been

developed , the objectives of the marketing research project should be clearly specified.


4) Approach to the Problem : A review of the relevant academic and trade literature should be

presented, along with some kind of an analytical model. If research questions and hypothesis

have been identified , then these should be included in the proposal.

5) Research Design : The research design adopted, whether exploratory, descriptive or causal

should be specified. Information should be provided on the following components :

i) kind of information to be obtained

ii) method of administering the questionnaire(mail, telephone, personal, electronic

interviews)

iii) Scaling techniques

iv) Nature of the questionnaire (type of questions asked, length. Average interviewing

time

v) Sampling plan and sample size

6) Field Work & Data Collection : The proposal should discuss how the data will be collected

and who will collect it . If the field work is to be subcontracted to another supplier and this

should be stated. Control mechanisms to ensure the quality of data collected should be

described.

7) Data Analysis : The kind of data analysis that will be conducted ( simple cross tabulations,

Univariate analysis , multivariate analysis ) and how the results will be interpreted should be

described.

8) Reporting : The proposal should specify whether intermediate reports will be presented and

at what stages , what will be the form of the final report, and whether a formal presentation of

the results will be made.

9) Cost & Time : The cost of the project and a time schedule , broken down by phases should be

presented . A CPM or PERT chart might be included. In large projects a payment schedule is

also worked out in advance.

10) Appendices : Any statistical or other information that is of interest only to a few people

should be contained in appendices.


Example of Data Available from Syndicated Firms

1) Demographic Data

Identification(name,address,telephone)

Sex

Marital status

Names of family members

Age

Income

Occupation

Number of children present

Home ownership

Length of residence

Number and make of cars owned

2) Psychographic Lifestyle Data

Interest in golf


Interest in snow skiing, book reading , running, bicycling, pets, fishing, electronics

Cable television.

Exploratory Research Design : Secondary Data

Published External Secondary Data :


Secondary Data

Internal External

Ready to useRequires further

ProcessingPublished Materials

ComputerizedData Bases

Syndicated Services

34

Syndicated Sources of Secondary Data :

Syndicated sources are companies that collect and sell common pools of data of know commercial

value designed to serve information needs shared by a number of clients. These data are not collected

for the purpose of marketing research problems specific to individual clients , but the data and

reports supplied to client companies can be personalized to fit particular needs .

For example reports could be organized on the basis of the clients sales territories or product lines.

Using Syndicated services is frequently less expensive than collecting primary data.


PublishedSecondary Data

General BusinessSources

Government Sources

Guides

Directories

Census Data

Other Government Publications

Indexes

Statistical Data

35

Advertising Evaluation :

Aim : to assess the effectiveness of advertising using print and broadcast media

Tools : MIRS ( Magazine Impact Research Service ) Gallup & Robinson

Process : ads are tested using an at home in magazine context among widely dispersed samples, the

system offers standardized measures with flexible design options.

Test ads may appear in the magazine or are inserted as tip-ins.


Unit of Measurement

Households/ Consumers

Institutions

Surveys Panels Retailers WholesalersElectronic ScannerServices

Purchase MediaPsychographi

c & Lifestyles

General

AdvertisingEvaluation

Volume Tracking

Data

Scanner Panels

Scanner Panels

With Cable TV

Industrial Firms

DirectInquiries

ClippingServices

Corporate Reports

36

It provides strong, validated measures of recall, persuasion, and ad reaction with responsive

scheduling.

TV Advertising : Method 1) Recruited audience method ( respondents are recruited and brought to

a central viewing facility , such as theater or mobile viewing laboratory , the respondents view the

commercials and provide data regarding knowledge, attitudes, and preferences related to the product

being advertised and the commercial itself)

Method 2) In home viewing method ( consumers evaluate commercials at home in

their normal viewing environment). New commercials can be pre tested at the network level or in

local markets distributed via VCR cassettes. A survey of viewers is then conducted to assess the

effectiveness of the commercials)

Surveys : Properly analyzed survey data can be manipulated in many ways so that the researcher

can look at intergroup differences , examine the effects of independent variables such as age or

income or even predict future behavior.

Psychographics & Lifestyles : Psychographics refer to the psychological profiles of individuals and

to psychologically based measures of lifestyle. Lifestyles refer to the distinctive modes of living of a

society or some of its segments. The measures are referred to as activities , interests & opinions or

simply AIO’s.

Media Panels : Electronic devices automatically record viewing behavior, thus supplementing a

diary or an online panel.

Company : Nielsen Media National Ratings service.

System : Electronic measurement system called the Nielsen People Meter.

The meters are placed in a sample of 5000 households (13,000) persons in the US randomly selected

and recruited by Nielsen Media so as to be representative of the population.


The Meter measures two things . 1) what program or channel is being tuned in

2) who is watching

The information is also disaggregated by 10 demographic and socioeconomic characteristics such as

household income , education of head of house, occupation of head of house, household size , age of

children, age of woman and geographic location. This is useful for firms in selecting specific TV

programs on which to air their commercials.

Purchase Panels : Respondents record their purchases of a variety of different products , as in the

NPD Panel. Eg : forecasting sales , estimating market shares, assessing brand loyalty , brand

switching behavior, establishing profiles of specific user groups, measuring promotional

effectiveness, and conducting controlled store tests.

Advantages : Panels provide longitudinal data ( data can be obtained from same respondents

repeatedly)

Disadvantages : Lack of representative ness, maturation and response biases.

Volume tracking Data : info on purchases by brand, size, price, flavor or formulation based on

sales data collected from the check out scanner tapes.

Electronic Scanner Services : Scanner data reflect some of the latest technological developments in

the marketing research industry. Scanner data are collected by passing merchandise over a laser

scanner, which optically reads the bar coded description( the universal product code ) printed on the

product. Being used by Reliance Fresh.

Advanced use of scanning : Scanner panels with cable TV. Eg : Tata Sky

Types of Marketing Research and Degree of Problem Definition

Exploratory Research

(Unaware of the problem )

Descriptive Research

(Aware of the Problem)

Causal Research

(Problem Clearly Defined)

Possible Situation

Our’s sales are decreasing “What kind of people are “Will buyers purchase more


and we don’t know why buying our product? Who

buys our competitor’s

product?”

of our products in a new

package”

Would people be interested

in our new product idea?

What features do buyer

prefer in our product?

“which of the two

advertising campaigns is

more effective

Exploratory Research Design : Qualitative Research

Primary Data : Qualitative Versus Quantitative Research

Qualitative Research : Provides insights and understanding of the problem setting .

Quantitative Research : Seeks to quantify the data and typically applies some form of statistical

analysis.

Whenever a new marketing research problem is being addressed ,quantitative research must be

preceded by appropriate qualitative research.

Data Types in Marketing Research Process :


Marketing Research Data

Secondary Data Primary Data

Qualitative DataQuantitative

Data

Research Procedures

Descriptive

Causal

Survey

Observational & Other Data

Experimental Data

DirectNondisguised

InDirectDisguised

Focus Group

Depth Interviews

Projective Techniques

Association Technique

Completion Technique

ConstructionTechnique

ExpressiveTechnique

Criteria for Evaluating Secondary Data

Criteria for Evaluating Secondary Data Criteria Issues RemarksSpecifications/ Methodology Data Collecting methods Data should be reliable,

valid and generalizable to the problem at hand

Response RateQuality of data Sampling Technique


Sample Size Questionnaire Design Field work Data Analysis

Error / Accuracy Examine errors in : Assess accuracy by

comparing data from different sources

Approach, research design Sampling, data collection data analysis, reporting Currency Time lag between collection

and publication , frequency of updates

Census data are periodically updated by syndicated firms

Objective

Why were the data collected

The objective will determine the relevance of the data

Nature Definition of key variables

Reconfigure the data to increase their usefulness if possible

units of measurement categories used Relationships examined

DependabilityExpertise, credibility,reputation Data should be obtained

from an original rather than an acquired source trustworthiness of data

Overview of Syndicated Services

Type Characteristics Advantages Disadvantages Uses

Surveys

Surveys conducted at regular intervals

Most flexible ways of obtaining data; information on underlying motives

Interviewers errors, Respondent errors Market segmentation

advertising theme selection

advertising effectiveness


Purchase Panels

Households provide specific information regularly over an extended period of time

Record purchase behaviors can be linked to the demographic / psychographic characteristics

Lack of representativeness; response bias; maturation Forecasting Sales

respondents asked to record specific behaviors as they occur

market share & trends

establishing consumer profilesbrand loyalty & switchingevaluating test marketsadvertisingdistribution

Media Panels

Electronic devices automatically recording behavior, supplemented by a diary

Same as purchase panel

Same as purchase panel

Establishing advertising rates

selecting media program or air time

establishing viewer profiles

Scanner Volume Tracking Data

Household purchases recorded through electronic scanners in supermarkets

Data reflect actual purchases; timely data, less expensive

Data may not be representative Price tracking

errors in recording purchases modeling

difficult to link purchases to elemets of marketing mix other than price

effectiveness of in store modeling

Scanner Panels with Cable TV

Scanner panels of households that subscribe to cable TV

Data reflect actual purchases;

Data may not be representative

Promotional mix analysis

sample controlQuality of data is limited copy testing

ability to link panel datato household characteristics new product testing

positioning


Audit Services

Verification of product movement by examining physical records of performing inventory analysis

Relatively precise information at the retail and wholesale levels

Coverage may be incomplete; matching of data on competitive activity may be difficult

Measurement of consumer sales and market share

competitive activity

analyzing distribution patternstracking of new products

Industrial Product Syndicated Services

Data banks on industrial establishments created through direct inquiries of companies

Important source of information in industrial firms

Data are lacking in terms of content quality and quality

Determining market potential by geographic area

clipping services

particularly useful in initial phases of projects

defining sales territories

corporate reports allocating advertising budget

Primary Research Process


Variations in Focus Groups

Two way focus group : this allows one target group to listen to and learn from a related group. In

one application physicians viewed a focus group of arthritis patients discussing the treatment they

desired . A focus group of these physicians was then held to determine their reactions.

Dual moderator group : This is a focus group interview conducted by two moderators. One

moderator is responsible for the smooth flow of the session, and the other ensures that specific issues

are discussed.

Dueling moderator Group : Two moderator groups taking opposite positions on the issues to be

discussed. This allows the researcher to explore both sides of controversial issues.

Respondent moderator group: In this type of focus group, the moderator asks selected participants

to play the role of moderator temporarily to improve group dynamics.

Client participant groups :Client personnel are identified and made part of the discussion group.

Their primary role is to offer clarifications that will make the group process more effective.


Surveys Personal interview (intercepts) Mail In-house, self-administered Telephone, fax, e-mail, Web

Quantitative Data

Primary Research

Experiments

Mechanical observation

Simulation

Qualitative Data

Case studies

Human observation

Individual depthinterviews

Focus groups

44

Mini Groups : These groups consist of a moderator and only 4 or 5 respondents. They are used when

the issues of interest require more extensive probing than is possible in the standard group of 8 to 12.

Telession Groups :Focus group sessions by phone using the conference call technique.

Online focus groups are an emerging form of focus groups .

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Depth Interviews : Like focus groups it is an unstructured and direct way of obtaining information,

but unlike focus groups depth interviews are conducted on a one to one basis. The respondent is

probed by a highly skilled interviewer to uncover underlying motivations , beliefs , attitudes and

feelings on a topic.

Duration of depth interview : 30 mins to more than 1 hour.

Techniques in Depth interviews :

1) Laddering Technique : The line of questioning proceeds from product characteristics to user

characteristics. This allows the researcher to tap into the consumer’s network of meanings. It

Provides a way to probe into the consumers deep underlying psychological and emotional

reasons that affect their purchasing decision.

Eg : when researcher wants to find out why a person buys a product apart from quality and

low price . to find out the in depth underlying motivators a laddering technique should be

used. ( color , taste, brand , price size) The underlying motivators are found by climbing the

ladder to the real reasons for purchasing the products.

2) Hidden issue Questioning : The focus is on not socially shared values but rather on personal

“sore spots” not on general lifestyles but on deeply felt personal concerns.

3) Symbolic Analysis : Attempts to analyze the symbolic meaning of objects by comparing them

With their opposites.To learn what something is , the researcher attempts to learn what it is

not. The logical opposites of a product that are investigated are : non usage of the product,

attributes of an imaginary “ non product” and opposite types of products. The three

techniques are illustrated with the following examples.

Focus Group Versus Depth Interviews


Characteristic Focus Group

Depth Interviews

Group Synergy and dynamics + -Peer Pressure/ group influence - +Client involvement + -Generation of innovative ideas + -In depth probing of individuals - +Uncovering hidden motives - +Discussion of sensitive topics - +Interviewing respondents who are competitors - +Interviewing respondents who are professionals - +Scheduling of respondents - +Amount of information + -Bias in moderation and interpretation + -Cost per respondent + -A " +" indicates a relative advantage over the other procedure , - indicates relative disadvantage

Application of Depth Interviews :

The use of depth interview is for exploratory research to gain insights and understanding .

1)Probing of respondent( automobile purchase)

2)Discussion of confidential, sensitive or embarrassing topics (personal finances,loose dentures)

3)Situations where strong social norms exist and the respondent may be easily swayed by group

response( attitude of college students towards sports)

4)Detailed understanding of complicated behavior (department store shopping)

5)Interviews with professional people( industrial marketing research)

6)Interviews with competitors who are unlikely to reveal the information in a group setting

( travel agents perception of airline package travel programs)

7)Situations where the product consumption experience is sensory in nature , affecting mood states

and emotions( perfumes , bath soap)


Projective Techniques : Both Focus groups and depth interviews are direct approaches in which the

true purpose of the research is disclosed to the respondents .

Projective techniques attempt to disguise the purpose of the research.

A projective technique is an unstructured , indirect form of questioning that encourages respondents

to project their underlying motivations, beliefs, attitudes or feelings regarding the issues of concern.

The respondents are asked to interpret the behavior of others rather than describe their own

behavior. In interpreting the behavior of others , respondents indirectly project their own

motivations, beliefs , attitudes or feelings into the situation. Thus respondents attitudes are uncovered

by analyzing their responses to scenarios that are deliberately unstructured , vague and ambiguous.

The more ambiguous the situation , the more respondents project their emotions, needs, motives ,

attitudes and values as demonstrated by work in clinical psychology on which projective techniques

are based. In psychology these techniques are classified as association, completion , construction

and expressive.

Association Techniques : An individual is presented with a stimulus and asked to respond with the

first thing that comes to mind.

Word association : In word association respondents are presented with a list of words one at a time

and asked to respond to each with the first word that comes to mind.

The underlying assumption of this technique is that association allows respondents to reveal their

inner feelings about the topic of interest.

Responses are analyzed by calculating

1) the frequency with which any word is given as a response

2) the amount of time that elapses before a response is given

3) the number of respondents who do not respond at all to test a word within a reasonable period of

time. those who do not respond at all are judged to have an emotional involvement so high that it

blocks a response. It is often possible to classify the associations as favorable , unfavorable or

neutral. An individuals pattern of responses and the details of the response are used to determine

the person’s underlying attitudes or feelings on the topic of interest .

Completion Techniques : The respondent is asked to complete an incomplete stimulus situation.

Common completion techniques in marketing research are sentence completion and story

completion.

Sentence Completion : It is similar to word association . Respondents are given incomplete sentence

and asked to complete them. Generally the first word that comes to their mind.


Story Completion : Respondents are given part of a story- enough to direct attention to a particular

topic but not to hint at the ending. They are required to to give the conclusion in their own words.

The respondents completion of this story will reveal their underlying feelings and emotions .

Construction Techniques :

Are closely related to completion techniques. Require the respondent to construct a response in the

form of a story , dialogue or description. The researcher provides less initial structure to the

respondent than in a completion technique. The two main construction techniques are

1) Picture response : The roots can be traced to the (TAT : Thematic Apperception Test) The

respondent is exposed to pictures of persons or objects that are clearly depicted and others that

relatively vague. The respondent is asked to tell stories about these pictures . The interpretation

gives indications of the individual’s personality. The individual may be characterized as

impulsive, creative , unimaginative and so on. The name thematic is used because of the themes

are elicited based on the subject’s perceptual interpretation( apperception) of pictures.

2) Cartoons : Cartoon characters are shown in a specific situation related to the problem. The

respondents are asked to indicate what one cartoon character might say in response to the

comments of another character. The response indicates the respondents feelings, beliefs and

attitudes toward the situation.

Expressive Techniques : Respondents are presented with a verbal or visual situation and asked to

relate the feelings and attitudes of other people to the situation. The respondents express not their

own feelings or attitudes but those of others. The two main expressive techniques are role playing

and third – person technique.

Role Playing : In role playing respondents are asked to play the role or assume the behavior of

someone else. The researcher assumes that the respondents will project their own feelings into the

role. These can then be uncovered by analyzing the responses.

Third –Person Technique: The respondent is presented with a verbal or visual situation and asked to

relate the beliefs and attitudes of a third person rather than directly expressing personal beliefs and

attitudes. Again the researcher will assume that the respondent is revealing personal beliefs and

attitudes while describing the reactions of a third party.


Application of Projective Techniques :

Word association is commonly used to test brand names and to measure attitudes about particular

products , brands packages or advertisements.

The usefulness of Projective Techniques is enhanced when the following guidelines are observed.

1) Projected technique should be used because the required information cannot be accurately

obtained by direct methods.

2) For exploratory research to gain initial insights and understanding

3) Given their complexity , projective techniques cannot be used naively.

A comparison of Focus Group , Depth Interviews & Projective TechniquesCriteria Focus group Depth Interviews Projective TechniquesDegree of Structure Relatively high Relatively medium Relatively lowProbing of individual respondents Low High Medium

moderator biasRelatively medium Relatively high Low to high

interpretation bias Relatively Low Relatively medium Relatively highuncovering subconscious information Low Medium to high Highdiscovering innovative information High Medium Low to highobtaining sensitive information Low Medium Gighinvolve unusual behavior /questioning No To limited extent Yesoverall usefulness Highly useful Useful Somewhat useful


Descriptive Research Design :

The key to good descriptive research is knowing exactly what you want to measure and selecting

a survey method in which every respondent is willing to co operate and capable of giving you

complete information efficiently.

Survey & Observation

Computer Assisted Telephone Interviewing : ( CATI) : uses a computerized questionnaire

administered to respondents over the telephone.

Observation Methods :

Recording the behavioral patterns of people , objects and events in a systematic manner to obtain

information about a phenomenon of interest. The observer does not question or communicate

with the people being observed. Observational methods may be structured or unstructured , direct

or indirect.

Structured : The researcher specifies in detail what is to be observed and how the measurements

are to be recorded.

Unstructured : The observer monitors all aspects of the phenomenon that seem relevant to the

problem at hand for eg : observing children playing with toys at hand.


Survey Methods

Telephone Interviewin

g

PersonalInterviewin

g

Mail Interviewin

g

Electronic Interviewin

g

TraditionalComputerAssisted

In- HomeMall

InterceptMail

Computer Assisted

MailPanel

E-mail internet

ObservationMethods

PersonalObservation

MechanicalObservation

AuditContentAnalysis

Trace Analysis

50

Disguised : The respondents are unaware of that they are being observed.

Undisguised: the respondents are aware that they are under observation .

Mechanical Observation : use of cameras. UPC system( Universal Product code). The barcode

recording of the purchases in reliance etc.

Audit : The researcher collects data by examining physical records or performing inventory

analysis.

Content Analysis : The phenomenon to be observed is communication rather than behavior or

physical objects. It is defined as objective ,systematic, and quantitative description of the manifest

content of a communication. It includes observation as well as analysis. The unit of analysis may

be words ( different words or types of words in the message), characters(individuals or objects),

themes (propositions), space and time measures(length , duration of the message) or

topics(subject of the message).

Eg : observing and analyzing the content or message of advertisements, newspapers articles,

television and radio programs and the like.

Trace Analysis : An observation method that can be inexpensive if used creatively is trace

analysis. In trace analysis , data collection is based on physical traces or evidence of past

behavior. These traces may be left intentionally or unintentionally by the respondents.

Eg : Store charge card slips are traces shoppers leave behind , which can be analyzed to examine

their store credit usage behavior.

Applications :

1) The selective erosion of tiles in a museum indexed by the replacement rate was used to

determine the relative popularity of exhibits.

2) The number of different fingerprints on a page was used to guage the readership of various

advertisements in a magazine

3) The position of the radio dials in cars brought in for service was used to estimate share of

listening audience of various radio stations. Advertisers used the estimates to decide on which

stations to advertise.

4) The age and condition of cars in a parking lot were used to assess the affluence of customers.

5) The magazines people donated to charity were used to determine people’s favorite magazines.

6) Internet visitors leave traces that can be analyzed to examine browsing and usage behavior

by using cookies.


Criteria Personal Observation

Mechanical Observation

Audit Content Analysis Trace Analysis

Degree of structure Low Low to high High High MediumDegree of disguise Medium Low to high Low High HighAbility to observe in natural setting High Low to high High Medium LowObservation bias High Low Low Medium Medium

Analysis Bias HighLow to Medium Low Low Medium

General Remarks

Most Flexible

Can be intrusive Expensive

Limited to communications

Method of last resort

Comparison of Observation & Survey methods : Only 1 % of marketing research projects rely solely

on observational methods to obtain primary data. This implies that observational methods have some

major disadvantages as compared to survey methods.

Relative advantages of observational methods :

1) Permit measurement of actual behavior rather than reports of intended or preferred behavior

2) No reporting bias and potential bias caused by the interviewer and the interviewing process is

eliminated or reduced.

Certain types of data can be collected only by observation.

i) behavior patterns that respondent is unaware of or is unable to communicate.

Eg : information on baby’s toys is best obtained by observing babies at play since they are unable

to express themselves.

Disadvantages of observational methods :

1)Reasons for the observed behavior may not be determined because little is known about the

underlying motives, beliefs , attitudes , preferences .

Eg : people buying a brand of cereal may not like it themselves.

2)Selective perception : ( bias in the researchers perception) can bias the data.

Observation has the potential to provide valuable information when properly used.

Other Methods : ( for Descriptive research design)

Mystery Shopping : Trained observers pose as customers and shop at company or competitor owned

stores to collect data about customer – employee interaction and other marketing variables such as

prices, displays, layout . The mystery shoppers question the store employees , mentally take note of

the answers and observe the variables of interest.


Causal Research Design :

The relationship between cause and effect tends to be probabilistic. We can never prove

causality(i.e demonstrate it conclusively ); we can only infer a cause and effect relationship.

Conditions for causality :

Ordinary meaning Scientific meaningX is the only cause of Y x is only one of a number

of possible causes of Y.

X must always lead to Y (X is deterministic cause of Y)

The occurrence of X makes the occurrence of Y more probable.(X is a probabilistic cause of Y)

It is possible to prove that X is a cause of Y

We can never prove that X is a cause of Y . At best we can infer that X is a cause of Y

1)Concomitant variation : is the extent to which a cause X and an effect Y occur together or vary

together in the way predicted by the hypothesis under consideration. Evidence pertaining to

concomitant variation can be obtained in a qualitative or quantitative manner.

Eg : in the qualitative case the management believes that sales are highly dependant upon the quality

of in store service. This hypothesis could be examined by assessing concomitant variation.

Causal factor X : Instore service

Effect Factor Y : Sales

2)Time order occurrence of variables : The causing event must occur either before or

simultaneously with the effect; it cannot occur afterwards.

By definition : an effect cannot be produced by an event that occurs after the effect has taken place.

It is possible for each event in a relationship to be both a cause and an effect of the other event.

In other words a variable can be both a cause and effect in the same causal relationship.

3) Absence of other possible causal factors : The absence of other possible causal factors means that

the factor on variable being investigated should be only possible causal explanation.

Eg: instore service may be a cause of sales if we can be sure that changes in all the other factors

affecting sales such as pricing , advertising level of distribution, product quality , competition and so

on were held constant or otherwise controlled.


Definitions and concepts :

Independent variables : They are variables or alternatives that are manipulated ( i.e levels of these

variables are changed by the researcher) and whose effects are measured and compared. These

variables are known as treatments , may include price levels, package designs, and advertising

themes.

Test units : are individuals, organizations, or other entities whose response to the independent

variables or treatments is being examined , may include consumers, stores , geographic means.

Dependent variables : Are the variables that measure the effect of the independent variables on the

test units. These variables may include sales, profits and market shares.

Extraneous Variables : are all variables other than the independent variables that affect the

response of 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, store location, and competitive effort.

Experiment : An experiment is formed when the researcher manipulates one or more independent

variables and measures their effect on one or more dependent variables, while controlling for the

effect of extraneous variables.

Experimental Design : is a set of procedures specifying

1) test units and how these units are to be divided into homogenous subsamples

2) what independent variables or treatments are to be manipulated

3) what dependent variables are to be measured

4) how the extraneous variables are to be controlled

Definition of Symbols :

X = the exposure of a group to an independent variable , treatment, or event, the effects of which

are to be determined

O= the process of observation or measurement of the dependent variable on the test units or group

of units

R=the random assignment of test units or groups to separate treatments.

Conventions :

movement from left to right indicates movement through time

horizontal alignment of symbols implies that all those symbols refer to a specific treatment

group

vertical alignment of symbols implies that those symbols refer to activities or events that

occur simultaneously.


Eg :

X O1 O2

Means that a given group of test units was exposed to the treatment variable (X) and the response

was measured at two different points in time O1 & O1.

R X1 O1

R X2 O2

Means that two groups of test units were randomly assigned to two different treatment groups at

the same time and the dependent variable was measured in the two groups simultaneously.

Validity in Experimentation :

When conducting an experiment , a researcher has two goals.

1) draw valid conclusions about the effects of independent variables on the study group

2) make valid generalizations to a larger population of interest.

The first goal concerns internal validity and the second goal concerns external validity.

Internal Validity : Whether the manipulation of the independent variables or treatments actually

caused the observed effects on the dependent variables . if the observed effects are influenced or

confounded by extraneous variables , it is difficult to draw valid inferences about the causal

relationship between the independent and dependent variables.

External Validity :refers to whether the cause and effect relationships found in the experiment can be

generalized. Can the results be generalized beyond the experimental situation and if so to what

populations , settings, times, independent variables and dependent variables can the results be

projected.

Threats to external validity arise when the specific set of experimental conditions does not

realistically take into account the interactions of other relevant variables in the real world.

To control for extraneous variables , a researcher may conduct an experiment in an artificial

environment. This enhances internal validity, but it may limit the generalizability of the results

thereby reducing external validity.

Extraneous Variables :

History


Refers to specific events that are external to the experiment but occur at the same time as the

experiment. These events may effect the dependent variable.

Eg : O1 X1 O2

O1 , O2 : measure of sales of a department store chain in a specific region and X1 represents a new

promotional campaign. Sales before and after the campaign.

The difference O2 – O1 is the treatment effect is the treatment effect. If there was no difference

between O2 and O1 then can we conclude that the promotional campaign was ineffective?

The promotional campaign is not the only possible explanation of the difference between O2 and O1 .

The campaign might be effective what if the general economic conditions declined during the

experiment.

The greater the time interval between observations, the greater the possibility that history will

confound an experiment of this type.

Maturation : Maturation (MA) is similar to history except that it refers to changes in the test units

themselves. The changes are caused by impact of the independent variables or treatments occur with

the passage of time.

In an experiment involving people , maturation takes place as people become older , more

experienced , tired , bored or uninterested . Tracking and market studies that span several months

are vulnerable to maturation , because it is difficult to know how respondents are changing over

time.

Testing Effects : Are caused by the process of experimentation. Typically these are the effects on the

experiment of taking a measure on the dependent variable before and after the presentation of the

treatment.

Two kinds of testing effects .

1) Main testing 2) Interactive testing effect (IT)

The main testing effect (MT) occurs when a prior observation affects a latter observation.

Eg : Experiment to measure the effect of advertising on attitudes toward a certain brand. The

respondents are given a pretreatment questionnaire measuring background information and attitude

toward the brand. They are then exposed to the test commercial embedded in an appropriate

program. After viewing the commercial the respondents again answer a questionnaire measuring

among other things , attitude towards the brand . Suppose there is no difference between the Pre and

Post treatment attitudes. Can we conclude that the commercial was in effective.


Alternative explanation is that the respondent tried to maintain consistency between their pre and

post treatment variables. As a result of the main testing effect , post treatment attitudes were

influenced more by pretreatment attitudes than by the treatment itself.

The main testing effect may also be reactive , causing the respondents to change their attitudes

simply because these attitudes have been measured.

In the interactive testing effect( IT) a prior measurement affects the test unit’s response to the

independent variable. Continuing with our advertising experiment when people are asked to indicate

their attitudes toward a brand, they become aware of that brand. They are sensitized to that brand

and become more likely to pay attention to the test commercial than people who were not included in

the experiment.

The measured effects are then not generalizable to the population : therefore the interactive testing

effects influence the experiment’s external validity.

Instrumentation : Refers to the changes in the measuring instrument , in the observers or in the

scores themselves. Sometimes measuring instruments are modified during the course of an

experiment. In the advertising experiment this could lead to variations in the responses obtained.

Eg : experiment in which dollar sales are being measured before and after exposure to an in-store

display (treatment). If there is a non experimental price change between O1 and O2 , this results in a

change in instrumentation because dollar sales will be measured using different unit prices.

In this case the treatment effect ( O2 – O1) could be attributed to a change in instrumentation.

Statistical Regression : effects occur when test units with extreme scores move closer to the average

score during the course of the experiment. In the advertising experiment suppose that some

respondents had either a very favorable or very unfavorable attitudes. On post treatment

measurement , their attitudes might have moved toward the average because people’s attitudes

change continuously . People with extreme attitudes have more room for change, so variation is more

likely. This has a confounding effect on the experimental results, because the observed effect( change

in attitude) may be attributable to statistical regression rather than to the treatment( test

commercial).

Selection Bias : Refer to the improper assignment of test units to treatment conditions. This bias

occurs when selection or assignment of test units results in treatment groups that differ on the


dependent variable before the exposure to treatment condition. If test units self select their own

groups or are assigned to groups on the basis of the researchers judgment , selection bias is possible.

Eg: merchandising experiment in which two different merchandising displays (old and new) are

assigned to different department stores . the stores vary on the key characteristic called store size.

Store size is likely to affect sales regardless of which merchandising display was assigned to a store.

Mortality : Refers to the loss of test units while the experiment is in progress. This happens for many

reasons , such as test units refusing to continue in the experiment. Mortality confounds results

because it is difficult to determine if the lost test units would respond in the same manner to the

treatment as those that remain .

Eg : merchandising display experiment. Suppose during the course of the experiment , three stores in

the new display treatment condition drop out. The researcher could not determine whether the

average sales for the new display stores would have been higher or lower if these three stores had

continued in the experiment.


Controlling Extraneous Variables : They pose a serious threat to the internal and external validity

of an experiment. Unless they are controlled for they affect the dependant variable and thus confound

the results. They can be controlled by 4 ways .

1) Randomization : Involves randomly assigning test units to experimental groups by using

random numbers . Treatment conditions are also randomly assigned to experimental groups.

But randomization may not be useful for small sample sizes.

2) Matching : Involves comparing test units on a set of key background variables before

assigning them to the treatment conditions. In the merchandising display experiment stores

could be matched on the basis of annual sales, size or location. The one store from each

matched pair would be assigned to each experimental group. Matching has two drawbacks .

First test units can be matched on only a few characteristics , so the test units may be similar

on the variables selected but unequal on others. Second if the matched characteristics are

irrelevant to the dependent variable , then the matching effort has been futile.

3) Statistical Control: Involves measuring the extraneous variables and adjusting for their

effects through statistical analysis .

Eg : relationship between purchase of fashion clothing and education , controlling for the

effect of income. Advanced statistical procedures such as analysis of covariance (ANCOVA)

the effects of the extraneous variables on the dependent variable are removed by an

adjustment of the dependent variable’s mean value within each treatment condition.

4) Design Control : Involves the use of experiments designed to control specific extraneous

Variables .


Classification of Experimental Designs

ExperimentsSymbolic

Representation

DescriptionAdvantag

esDisadvantages Case study example

1

One shot case study

X O1

Known as after only design . A single group of test units is exposed to a treatment X and then a single measurement on the dependent variable is taken ( O1). There is no random selection of test units. The test units are selected arbitarily by the researcher or by self selection procedures.

It is more appropriate for exploratory than for conclusive research

The danger of drawing valid conclusions . It does not provide the basis of comparing the level of O1 to what would happen when X was Absent.The level of O1 might be effected by many extraneous variables including history, maturation, selection and mortality.

One shot case study for measuring the effectiveness of a test commercial for a department store Sears is as follows. A telephone interview is conducted for respondents who have watched a test commercial(X) on sears , the previous night. The dependent variables (Os) are unaided and aided recall. First unaided recall is measured by asking the respondents whether they recall seeing a commercial for a department store, if recall is done further information about the same is solicited regarding the commercial content . If the respondents do not recall the test commercial aided recall is done by the question" do you recall the seeing a commercial on Sears last night" . The results of aided and unaided recall are compared to norm scores to develop an index for interpreting the scores.


Experimental

Designs

Pre Experimen

tal

True Experimen

tal

QuasiExperimen

talStatistical

One Shot case Study

One Group Pretest - Posttest

Static Group

Pretest-PosttestControl Group

Post test only

Control Group

Solomon Four - Group

Time Series

Multiple Time Series

Randomized Blocks

Latin Square

Factorial

60

Experiments

Symbolic Representati

onDescription Disadvantages Case study example

2

one Group Pretest- Post Test Design

O1 X O2

A group of test units is measured twice. There is no control group. First a pre treatment measure is taken (O1), then the group is exposed to the treatment (X). Finally a post treatment measure is taken ( O2). History , maturation, testing , instrumentation,selection, mortality and regression could possibly be present.

The treatment effect is computed as O2 - O1, but the validity of this conclusion is questionable because extreneous variables are largely uncontrolled

Example of sears again : Respondents are recruited to central theatre locations in different test cities. At the central location, respondents are first administered a personal interview to measure , among other things attitudes towards the store , Sears (O1). Then they watch a TV Program , the respondents are again administered a personal interview to measure attitudes towards the store Sears(O2). The effectiveness of the test commercial is measured as O2- O1.


Experiments

Symbolic Representati

onDescription Disadvantages Case study example

3Static Group Design

EG: X O1 CG: O2

Two group experimental design. One group called the experimental group EG, is exposed to the treatment and the other called the control group (CG) is not. Measurements on both groups are made only after the treatment, and test units are not assigned at random. The control group is sometimes defined as the group that receives the current level of marketing activity , rather than a group that receives no treatment at all. The control group is defined this way because it is difficult to reduce current marketing activities , such as advertising and personal selling to zero.

The measured difference O1-O2 can be attributed to at least two extraneous variables(seletion and mortality) . Also because the test units are not randomly assigned the two groups(EG and CG ) may differ before the treatment and selection bias may be present. There also can be mortality effects because more test units may withdraw from the experimental group than from the control group. This would happen if the treatment was unpleasant.

Eg for Sear's store : Two groups of respondents would be recruited

on the basis of convenience . Only the experimental group would be exposed to the TV program containing the test ( Sears) commercial. Then

attitudes toward Sear's store of both the experimental and control

group respondents would be measured. The effectiveness of the test commercial would be

measured as O1- O2

Experiments

4

True Experimental Designs

The distinguishing feature of the true experimental desings as compared to pre experimental designs is randomization. In true experimental designs , the researcher randomly assigns test units to experimental groups and treatments to experimental groups. True experimental designs include the pretest - post test control group design, the post test only control group design


ExperimentsSymbolic

Representation

5

Post Test Only Control Group Design

EG : R O1 X O2 CG : R O3 O4

Test units are randomly assigned to either the experimental or the control group and a pretreatment measure is taken on each group. Only the experimental group is exposed to the treatment but the posttest measures are taken on both groups. The treatment effect (TE ) is measured as ( O2 - O1 ) - (O4 - O3). This design controls for most extraneous variables . Selection bias is eliminated by randomization. The other extraneous effects are controlled as follows. O2-O1 = TE + H + MA + MT + IT+ I + SR+ MO O4- O3= H + MA+MT+I + SR+ MO = EV(extraneous variables ). The experimental result is obtained by ( O2 - O1) - ( O4- O3) = TE + IT

Experiments

6

Quasi Experimental Designs

The researcher can control when measurements are taken and on whom they are taken. Secon the researcher lacks control over the scheduling of the treatments and also is unable to expose test units to the treatments randomly. Quasi experimental designs are useful because they can be used in cases when true experimentation cannot and because they are quicker and less expensive. Full experimental control is lacking and the researcher must take into account the specific variables that are not controlled. Popular forms of quasi experimental designs are time series and multimple time series designs.

ExperimentsSymbolic

RepresentationDescription Disadvantages Case study example

7

Time Series Designs

O1 O2 O3 O4 O5 O6 O7

O8 O9 O10

Involves a series of periodic measurements on the dependent variable for a group of test units. The treatment is then administered by the researcher or occurs naturally. After the treatment periodic measurements are continued to determine the treatment effect.

a major disadvantage is the failure to control history. Another limitation is that the experiment may be effected by the interactive testing effect, because multiple mesurements are being made on the test units.

The effectiveness of a test commercial ( X ) may be examined by broadcasting the commercial a predetermined number of times and examining the data from a preexisting test panel. Although the marketer can control the scheduling of the test commercial, it is uncertain when or whether the panel members are exposed to it. The panel members purchases before during, and after the campaign are examined to determine whether the test commercial has a short -term effect , a long term effect or no effect.


Experiments Symbolic Representation Description Case study example

8

Multiple Time Series Designs

EG : O1 O2 O3 O4 O5 X O6 O7 O8 O9 O10 CG : O 11 O12 O13 O14 O15 O 16 O17 O18 O19 O20

Similar to time series design except that another group of test units is added to serve as a control group . If a control group is selected carefully this design can be an improvement over the simple time series experiment.

The test commercial would be shown in only a few of the test cities. Panel members in these cities would comprise the experimental group. Panel members in cities where the commercial was not shown would constitute the control group.

Statistical DesignsSno Description Example

1Randomized block design

It is useful when there is only one major external variable such as sales ,store size, or income of the respondent that might influence the dependent variable. The test units are blocked or grouped on the basis of the external variable. The researcher must be able to identify and measure the blocking variable. Randomized block designs are more useful than completely random designs, their main limitation is that the researcher can control only one external variable. when more than one variable must be controlled the researcher must use latin square design

2Latin square Design

allows the researcher to statistically control two non interacting external variables as well as to manipulate the independent variable. Each external or blocking variable is divided into equal number of blocks or levels. The independent variable is also divided into the same number of levels.

3 Factorial Design

used to measure the effects of two or more independent variables at various levels. Unlike the randomized block design and the latin square design , factorial designs allow for interactions between variables. An interation is said to take place when the simultaneous effect of two or more variables is different from the sum of their separate effects. eg : an individual's favorite drink might be coffee and favorite temperature level might be cold but this individual might not prefer cold coffee leading to an interaction.

For more indepth information of the above topics you can refer to N. K Malhotra


Sampling - Design & Procedures

The formulation of Research design is the third step of marketing research process. At this stage the information needed to address the marketing research problem has been identified and the nature of the research design ( exploratory , descriptive or causal ) has been determined.

Further Scaling & measurement process has been specified and the questionnaire designed .

The next step is to design a suitable sampling procedure.

Basic questions for sampling design .

1) should a sample be taken2) if so what process should be followed3) what kind of sample should be taken4) how large should it be5) what can be done to control and adjust for non response error

Sampling process

Target Population : Collection of elements or objects that possess the information sought by the researcher and about which inferences are to be made. The target population must be defined precisely.

Element : The object about which or from which the information is desired , the element is usually a respondent .

Sampling Unit : It is an element or a unit containing the element , that is available for selection at some stage of the sampling process .

Flow chart of Sampling Process .

Example for Target Population : The target population for the department store project was defined as follows.

Elements : Male or Female head of the household responsible for most of the shopping at department stores


Define the target population

Determine the sampling frame

Select a sampling technique

Determine the sample size

Execute the sampling process

65

Sampling units : HouseholdsHere Sampling unit and population element are different. Example for Defining Target Population ?

Consider a marketing research project assessing consumer response to a new brand of men’s cologne. Who should be included in the target population ? Men 17 or older ? should females be included , because some women buy colognes for their husbands ? these and similar questions must be resolved before the target population can be approximately defined ?

Sampling Frame : It is a representation of the elements of the target population . It consists of a list or a set of directions for identifying the target population .

Examples : Telephone bookA city directoryMap

Sampling Technique : It involves several decisions of a broader nature . The researcher must decide whether to use a Bayesian or traditional sampling approach, to sample with or without replacement , and to use non probability or probability sampling.

Bayesian Approach : The elements are selected sequentially. After each element is added to the sample, the data are collected, sample statistics computed and sampling costs determined.

Sampling with Replacement : An element is selected from the sampling frame and appropriate data are obtained . Then the element is placed back in the sampling frame. As a result , it is possible for an element to be included in the sample more than once.

Sampling without Replacement : Once an element is selected for inclusion in the sample , it is removed from the sampling frame and therefore cannot be selected again.

The most important decision about the choice of sampling technique is whether to use probability or non probability sampling.

Sample Size : Refers to the number of elements to be included in the study . Determining the sample size is complex and involves several qualitative and quantitative considerations.

Conditions for precision or accurate information .1) Larger Samples2) The standard deviation of the mean is inversely proportional to the square root of the sample

size.

The nature of research also has an impact on the sample size .

1) Exploratory Research Design : Sample size is typically small.2) Conclusive or descriptive research design : Sample size is large


Sample Sizes used in Marketing Research .

Type of Study Minimum Size Typical RangeProblem identification research ( eg : market potential )

500 1000 to 2000

Problem Solving Research(eg : pricing )

200 300 to 500

Product Tests 200 300 to 500Test – marketing Studies 200 300 to 500TV/radio/print advertising( per commercial or ad test )

150 200 to 300

Test-market audits 10 stores 10 – 20 storesFocus group 2 groups 6 to 15 groups

Finally the sample size should be guided by a consideration of the resource constraints.In any marketing research project, money and time are limited . Other constraints include the availability of qualified personnel for the data collection.

Executing the Sampling Process

Execution of the sampling process requires a detailed specification of how the sampling design decisions with respect to the population , sampling frame, sampling unit, sampling technique and sample size are to be implemented .


Classification of Sampling Techniques .Non probability sampling relies on the personal judgment of the researcher rather than chance to select sample elements. The researcher can arbitrarily or consciously decide what elements to include in the sample.Non probability samples may yield good estimates of the population characteristics. However they do not allow for objective evaluation of the precision of the sample results.

In probability sampling , sampling units are selected by chance. It is possible to pre specify every potential sample of a given size that could be drawn from the population as well as the probability of selecting each sample. Every potential sample need not have the same probability of selection, but it is possible to specify the probability of selecting any particular sample of a given size. Confidence intervals which contain the true population value with a given level of certainty can be calculated .

Probability sampling techniques are classified as follows.1) element versus cluster sampling2) equal unit probability versus unequal probabilities


Sampling Techniques

Non Probability Probability

Convenience Sampling

Judgmental Sampling

Quota Sampling

SnowballSampling

Simple Random Sampling

Systematic Sampling

StratifiedSampling

ClusterSampling

Other SamplingTechniques

Proportionate

Disproportionate

68

3) unstratified versus stratified selection4) random versus systematic selection5) single stage versus multistage techniques

Non Probability Sampling Techniques :Convenience Sampling : Attempts to obtain a sample of convenient elements. The selection of sampling units is left primarily to the interviewer. Often respondents are selected because they happen to be in the right place at the right time.

Eg : 1)use of students , members or social organizations 2)mall intercept interviews without qualifying the respondents 3)people on streets

It is the lest expensive and least time consuming of all sampling techniques

Judgemental Sampling : it is a form of convenience sampling in which the population elements are selected based on the judgement of the researcher. The researcher exercising judgement or expertise chooses the elements to be included in the sample , because he or she believes that they are respresentative of the population of interest or are otherwise appropriate.

Eg : 1)test markets selected to determine the potential of a new product 2)department stores selected to test a new merchandising display screen

It is low cost , convenient and quick but it does not allow direct generalizations to a specific population , usually because the population is not defined explicitly. Judgemental sampling is subjective and its value depends entirely on the researcher’s judgement , expertise and creativity. It may be useful if broad population inferences are not required .

Quota Sampling : Two stage restricted judgmental sampling . The first stage consists of developing control categories , or quotas of population elements

To develop these quotas , the researcher lists relevant control characteristics and determines the distribution of these characteristics in the target population. . The relevant control characteristics which may include sex, age and race are identified on the basis of judgement .

Often the quotas are assigned so that the proportion of the sample elements possessing the control characteristics is the same as the proportion of population elements with these characteristics.

Snowball Sampling : An initial group of respondents is selected , usually at random. After being interviewed these respondents are asked to identify others who belong to the target population of interest.

Major objective of snowball sampling is to estimate characteristics that are rare in the population.

Eg : users of particular government or social services such as food stamps, whose names cannot be revealed,

Special census groups such as widowed males under 35.Used in industrial buyer –seller research to identify buyer seller pairs.


Major disadvantage is that it substantially increases the likelihood of locating the desired characteristic in the population. It also results in relatively low sampling variance and costs.

Graphical Illustration of Non Probability Sampling Techniques :

1) Convenience Sampling

A B C D E1 6 11 16 212 7 12 17 223 8 13 18 234 9 14 19 245 10 15 20 25

Group D happens to assemble at a convenient time and place. So all the elements in this group are selected . The resulting sample consists of elements 16,17,18,19 and 20 . No elements are selected from groups A, B, C and E.

2) Judgmental Sampling

A B C D E1 6 11 16 212 7 12 17 223 8 13 18 234 9 14 19 245 10 15 20 25

The researcher considers groups B , C and E to be typical and convenient. Within each of these groups one or two elements are selected based on typicality and convenience. The resulting sample consists of elements 8,10,11,13 and 24. Note that no elements are selected from groups A and D.

3) Quota Sampling

A B C D E1 6 11 16 212 7 12 17 223 8 13 18 234 9 14 19 245 10 15 20 25

A quota of one element from each group, A to E is imposed. Within each group, one element is selected based on judgement or convenience. The resulting sample consists of elements 3, 6, 13,20 and 22. One element is selected from each column or group.


4) Snowball sampling

A B C D E1 6 11 16 212 7 12 17 223 8 13 18 234 9 14 19 245 10 15 20 25

Elements 2 and 9 are selected randomly from groups A & B. Element 2 refers elements 12 and 13.Element 9 refers element 18. The resulting sample consists of elements 2, 9,12, 13 and 18. Note that no element is selected from group E.

Probability Sampling Techniques :

1) Simple Random SamplingEach element in the population has a known and equal and equal probability of being the sample actually selected. This implies that every element is selected independently of every other element. The sample is drawn by a random procedure from a sampling frame.

Eg : lottery system .The researcher first compiles a sampling frame in which each element is assigned a unique identification number.

2) Systematic Sampling : In systematic sampling , the sample is chosen by selecting a random starting point and then picking every ith element in succession from the sampling frame.

The sampling interval i is determined by dividing the population size N by the sample size n and rounding to the nearest integer.

Population = 10,000 elementsSample desired = 1000Sampling interval I = 100A random number between 1 and 100 is selected . If for example this number is 23 , the sample consists of elements 23, 123,223,323,423,523 and so on. It is similar to simple random sampling in that each population element has a known and equal probability of selection. It is different from SRS in that only the permissible samples of size n that can be drawn have a known and equal probability of being selected. For systematic sampling the researcher assumes that the population elements are ordered in some respect.It is less costly and easier than SRS because random selection is done only once.


3) Stratified Sampling : It is a two step process in which the population is partitioned into sub populations or strata. The strata should be mutually exclusive and collectively exhaustive in that every population element should be assigned to one and only one stratum and no population elements should be omitted.

Next elements are selected from each stratum by a random procedure usually SRS. Technically only SRS should be employed in selecting the elements from each stratum.In practice sometimes systematic sampling and other probability sampling procedures are employed.

It differs from quota sampling in that the sample elements are selected probabilistically rather than based on convenience or judgement.

Major objective of stratified sampling is to increase precision without increasing cost.

4) Cluster Sampling.

The target population is first divided into mutually exclusive and collectively exhaustive sub populations or clusters.

Then a random sample of clusters is selected , based on a probability sampling technique such as SRS. For each selected cluster , either all the elements are included in the sample or a sample of elements is drawn probabilistically.

If all elements in each selected cluster are included in the sample, the procedure is called one stage cluster sampling.If a sample of elements is drawn probabilistically from each selected cluster, the procedure is two stage cluster sampling.

Two stage cluster sampling can be either simple two stage cluster sampling involving SRS or probability proportionate to size (PPS) sampling.


Cluster Sampling

One Stage Sampling

Two StageSampling

MultistageSampling

Simple ClusterSampling

Probability Proportionate

To size sampling

72

Distinction between cluster sampling and stratified sampling

Cluster Sampling Stratified samplingOnly a sample of subpopulations (clusters ) is chosen

all the subpopulations( strata) are selected for further sampling

Objective is to increase sampling efficiency by decreasing costs

Objective is to increase precision

Form of cluster sampling is area sampling.Only one level of sampling for selecting the basic elements , the design is called single stage area sampling.If two or more levels of sampling take place the design is called two or multi stage area sampling

Stratified sampling is used when the researcher is particularly interested in certain specific categories within the total population . The population is divided into strata on the basis of recognizable or measurable characteristics of its members . eg : age, income , education etc. The total sample then is composed of members from each strata so that the stratified sample is really a combination of the number of smaller samples.

The various units comprising the population are grouped in clusters and the sample selection is made in such a way that each cluster has a known chance of being selected.

The use of stratified samples will lead to more accurate results only if the strata selected are logically related to the data sought.

A cluster sample is useful in two situations i) when there is incomplete data on the composition of the population ii)when it is desirable to save time and costs by limiting the study to specific geographical areas

Proportionate stratified sampling wherein the no of items in each stratum is proportionate to their number in the population . Disproportionate sampling wherein a proportionate sampling would cause the occurrence of very little data . Eg : study is to be conducted about the characteristics of car owners with the type of car being owned the basis for stratification. So if a proportionate sampling is used very little data about Chevrolet (2) and Toyota ( 3) might be obtained. So now a smaller number of maruti , ambassador and fiat owners would be included in the sample than their number in the universe warrants.

Major disadvantage is that cluster sampling leads to a substantial loss in precision , since units within each cluster tend to be rather heterogeneous

In probability proportionate to size sampling , the clusters are sampled with probability proportional to size. The size of cluster is defined in terms of the number of sampling units within that cluster.

So in first stage large clusters are more likely to be included than small clusters. In second stage the probability of selecting a sampling unit in a selected cluster varies inversely with the size of the cluster.


Thus the probability that nay particular sampling unit will be included in the sample is equal for all units, because the unequal first stage probabilities are balanced by the unequal second stage probabilities.

Strength and weakness of basic Sampling TechniquesTechnique Strengths Weaknesses

Non Probability Sampling

Convenience samplingLeast expensive, least time consuming , most convenient

Selection bias, sample not representative , not recommended for descriptive or causal research

Judgmental samplingLow cost, convenient , not time consuming

Does not allow generalization, subjective

Quota SamplingSample can be controlled for certain characteristics

Selection bias, no assurance of representativeness

Snowball SamplingCan estimate rare characteristics Time consuming

Probability Sampling

Simple random sampling (SRS)

Easily understood, results projectable

Difficult to construct sampling frame, expensive , lower precision, no assurance of representative ness

Systematic Sampling

Can increase representativeness, easier to implement than SRS, sampling frame not necessary

can decrease representativeness

Stratified SamplingIncludes all important sub populations , precision

Difficult to select relevant stratification variables, nof feasible to stratify on many variables, expensive

Cluster samplingEasy to implement, cost effective

Imprecise, difficult to compute and interpret results

Other probability sampling techniques.i) Sequential sampling : The population elements are sampled sequentially, data collection and

analysis are done at each stage and a decision is made as to whether addional population elements

should be sampled. Sequential sampling has been used to determine preferences for two competing

alternatives.

Eg : to establish the price differential between a standard model and a deluxe model of a consumer

durable.

In one study respondents were asked which of two alternatives they preferred , and sampling was

terminated when sufficient evidence was accumulated to validate a preference.

ii) Double Sampling : Also called two phase sampling, certain population elements are sampled

twice. In the first phase a sample is selected and some information is collected from all the elements

in the sample. In the second phase a sub sample is drawn from the original sample and additional

information is obtained from the elements in the sub sample.


It is useful when no sampling frame is readily available for selecting final sampling units but when

the elements of the frame are known to be contained within a broader sampling frame,

Choosing NonProbability Versus Probability Sampling

FactorsNon Probability Sampling

Probability Sampling

Nature of Research Exploratory Conclusive

Relative Magnitude of sampling and non sampling errors

Nonsampling errors are larger

Sampling errors are larger

Variability in the population Homogenous ( Low)

Heterogenous ( high)

Statistical considerations Unfavorable FavorableOperational Considerations Favorable Unfavorable

Internet Sampling

Definitions & Symbols


Internet Sampling

Online Intercept Sampling

Recruited Online Sampling

Other Techniques

Non Random Random Panel Non Panel

RecruitedPanels

Opt- in Panels

Opt – in ListRentals

75

Confidence intervals and other statistical concepts that play a central role in sample size determination are defined in the following list.

Parameter : It is a summary description of a fixed characteristic or measure of the target population. A parameter denotes the true value that would be obtained if a census rather than a sample were undertaken.

Symbols for population and sample statisticsVariable Population Sample

Mean μ X—

Proportion p

Variance σ 2 s2

Standard Deviation σ s

Size N n

Standard Error of the mean σX— Sx_

Standard error of the proportion σp Sp

Standardized variate(Z) (X -μ)/ σ (X-X—)/S

Coefficient of Variation "(C ) " σ/μ S/X—

Statistic : is a summary description of a characteristic or measure of the sample. The sample statistic is used as an estimate of the population parameter

Finite Population Correction : The finite population correction (FPC) is a correction for overestimation of the variance of a population parameter, for example a mean or proportion when the sample size is 10 % or more of the population size.

Precision Level : When estimating a population parameter by using a sample statistic, the precision level is the desired size of the estimating interval. This is the maximums permissible difference between the sample statistic and the population parameter.

Confidence Interval : The confidence interval is the range into which the true population parameter will fall, assuming a given level of confidence.

Confidence Level : The confidence level is the probability that a confidence interval will include the population parameter.


Sampling Process 1) How population is defined .

It is the specific group of people, firms, conditions , activities which form the pivotal point of research project. For developing a sample it becomes the primary duty of the researcher to define the population from which to draw the sample.

What is the population for your project ?

2) Sampling frame Developing a sampling frame. A sampling frame may be defined as the listing of the general components of the individual units that comprise the defined population.

Eg : a fertilizer manufacturing company wants to test its fertilizer efficiency with regard to crop yields. The sample units here can include fields.The sampling frame would be the set of coordinates used for selecting various agricultural fields.

3) Sampling methods . Whether probability or non probability sampling. i) Probability sampling if sampling error is to be estimated. ii) Non Probability sampling if the research is required to defend the randomness in the selection of sample units.iii)it is better to use non probability sampling when a frame does not exist nor one can be developed .iv)non probability sampling should be used if the researcher has limited knowledge about the statistical aspects of sampling. 4) Selecting Sample size Trade off between cost and sampling error , larger sample keeps error small but increases sampling cost. So how to select the correct sample size.

Probability Sampling : Degree of precision , knowledge of standard error and confidence intervals.

Non Probability Sampling : The central limit theorem or the principle of normal distribution is not applicable in the case of non probability sampling. It is the researchers intuition and judgment which makes the decision about the size of the sample.

The size of the sample in Non Probability sampling is generally 1/10th of the population size.


Sampling Metrics

Standard Errors of Mean & Proportion

i) Sampling of variables involves a numerical value such as average height, weight , expenditure here we estimate the standard error of the mean. Sigma x barii) Sampling of proportions is expressed in dichotomous terms. Eg : good or bad, heavy or light , high or low here standard error of the proportion is calculated.

Estimating standard error of the mean:The following formula is used :

Sx bar = (s/ sqrt (n)) * (sqrt((N-n)/N))

S x bar = standard error of the mean using sample dataS = standard deviation derived from the sample.n = sample sizeN = Population SizeExample : We are studying the spending habits of college students on entertainment. If there is a population of 6000 students and we take a sample of 500 students, we find that the average student spends Rs 100 per month on entertainment, This figure of Rs 100 is taken as an average for all 6000 students population . But is this assumption valid ? suppose if we take another sample of 500 students and survey them would it be valid . The results would actually be different . It is all because of sampling error ?Now Average spending is 100 .Standard deviations is say Rs 15

Estimate the standard error of the mean S x bar. = 15/sqrt(500) * Sqrt((600 – 500)/6000) = 0.64

The formula for constructing a confidence interval is :

X bar + - 1.96(S xbar) = 95 percent confidence interval

X bar + - 2.57(S x bar) = 99 percent confidence interval

X bar = sample mean

S xbar = standard error of the mean of the sample.

Substituting the sample data at 95 % confidence interval Xbar + - 1.96 (S xbar) = 95 % confidence interval(Rs 100 ) +- 1.96 ( Rs 0.64)(Rs 100) +- 1.26Rs 98.74 – Rs 101.26


This means that if a large number of probability samples ( size 500 ) were taken from the particular city college student body, 95 % of these samples would contain the actual mean of the universe within an interval of +- 1.96 standard errors of their mean values. This is want the 95 % confidence interval of Rs 98.74 to Rs 101.26 mean. It is assumed that this sample with a mean of Rs 100 is among those 95 percent.

99% interval is derived as Xbar + - 2.57 ( Sx bar) = 99 %Rs 100 +- 2.57(0.64)Rs 100 +- 1.64Rs 98.36 – Rs 101.64

What does this 99 % mean. It means that if a large number of samples ( size 500) are taken from the student body, 99% of these would contain the actual mean of the universe within an interval +- 2.57 of standard errors of their mean value.

So at 95% confidence interval 6000 * Rs 98.74 – 6000 * Rs 101.26Rs 592.440 – Rs 607.560

Standard error of proportion :

The formula for computing the standard error of proportion is

Sp = Sqrt ((( P.q)/n )*( (N-n)/N)))

Sp = standard error of the proportion P = proportion of sample possessing a certain attribute.

q or (1-p) = proportion of sample not having that attribute.

Example : A study of 5000 students was conducted by taking a sample of 400 students. It was regarding the students attitudes towards a company’s products. It was found that 88 % of the students (22% ), had favorable attitudes toward the company’s products while 312 students (78 %) held unfavorable ones. Based on this information, what would be the 95 % confidence interval depicting the attitude of the entire student body toward company’s products?

Standard error of the proportion is computed as follows

Sp = Sqrt((P.q)/n * (N-n)/N))

Sp = Sqrt((22.78/400 ) *(5000 – 400)/5000)) = 4.29 * 0.92 = 1.99 %Now the formula for 95 % confidence interval when attributes are involved is P =- 1.96(Sp)= 22 % + - 1.96(1.99 %)= 22% + - 3.9 %


= 18.1 % - 25.9 %Based on this 18 to 26 % confidence interval we can say that only 12 % to 26 % of the college students have a favorable attitude towards the company’s products.

Using standard error of the mean to determine the sample size

S xbar = s/ Sqrt(n)

Squaring both sides Sx bar square = s square / n

n = S 2 / S xbar 2

Example : Colgate Palmolive India Ltd wants to know how much does a male student spend annually on toothpastes. They selected Punjab agricultural university , ludhiana as population and took sample for the study . Assume that there were 18,000 students of which 8,300 were males. The company wants to be reasonably confident that the amount obtained will be within +- Rs 2 of the average male students real expenditure. How large a sample will be needed to achieve this precision. The researcher is fairly confident that the range of annual expenditures for toothpastes by male students at Punjab agricultural university is from Rs 0 to Rs 80.

Using the above formula

n = S 2/ S xbar 2

n= S 2 / Sx 2 n=(Rs 80 / 6) 2 / Sx 2

= (Rs 13.5 ) 2 / Sx 2

Here we divided S / 6 because in a normally distributed set of values , the range encompasses approximately three standard deviations . Thus 80 / 6 gives a reasonable estimate of 1 standard deviation.

The result is total estimated range/ 6 = estimated S

At 95 % confidence interval involves 1.96 Sx . in our example that means 1.96 Sx = +- Rs 2

Sx bar = Rs 2/ 1.96 = Rs 1.02

Substituting this value in original formula , we get

n= (Rs 13.50) 2 / Rs 1.02 2

= 182.25/ 1.04 = 175

Thus to be 95 % confident and the sample mean within +- Rs 2 of the population mean a sample of 175 male students must be contacted.

If colgate Palmolive needs greater precision, that is a smaller interval of +- Re 1 requires a sample of 701 students as under


1.96 Sx = Re1 So Sx = 0.51

n=(Rs 13.5)2 / 0.51 2 = 182.25 / 0.26 = 701Now small change in precision sought ( + - Re 1 vs +- 2 ) requires almost 4 fold increase in the size of the sample needed.

Now small change in precision size i.e if the n value generated by the above procedure i.e is more than 5 % of the population ( n / N > 5% ) the sample size should be revised downward . The downward revision would take the following form.

n alpha = n / 1 + n / N where n alpha = adjusted sample.

Now in the above example sample size was 175, this is only 2 % of the total population of 8300 students , so no downward adjustment is needed.

In case 701 is the sample size ( 8.5 % of the population ) downward adjustment is needed.

n alpha = 701 / 1 + 701 / 8300 = 646

Sample size in case of attributes


Measurement & Scaling

Fundamentals & Comparative Scaling

Measurement : Assigning numbers or other symbols to characteristics of objects according to certain prespecified rules.

Scaling : Creation of a continuum upon which measured objects are located.

What do we Measure ?• Perceptions, attitudes, preferences or any relevant characteristics.

• Characteristics of the Objects

Numbers in Marketing Research

• Numbers facilitate statistical analysis of resulting of data.

• Communication of Measurement rules & results.

Specification of Rules • Important aspect of measurement : Measurement Process must be Isomorphic. Isomorphic

means – one to one correspondence between the numbers and the characteristics being measured.

Scaling Example• Consider a scale from 1 to 100 for locating consumers according to the characteristic “ attitude toward department Stores”

Each respondent is assigned a number from 1 to 100 1 = extremely unfavorable100 = extremely favorable

Measurement & Scaling• Measurement is the actual assignment of a number from 1 to 100 to each respondent.

• Scaling is the process of placing the respondents on a continuum with respect to their attitude towards an object or process

Primary Scales of Measurement• Nominal Scale ; Ordinal Scale ; Interval Scale ; Ratio Scale

Examples• Nominal Scale - Numbers assigned to Runners

• Ordinal Scale - Rank order of winners

• Interval Scale - Performance rating on a 0- to 10 Scale

• Ratio Scale - Time to Finish in Seconds


Primary Scales of MeasurementScale Basic

CharacteristicsCommon Examples

Marketing Examples

Permissible StatisticsDescriptive Inferential

Nominal Numbers Identify & classify Objects

Social Security Nos Numbering of Football Players

Brand Numbers, Store types , Sex classification

Percentages,mode Chi Square , Binomial

Ordinal Numbers indicate the relative positions of the objects but not the magnitude of differences between them

Quality rankings, rankings of teams in a tournament

Preference rankings, market position,Social class

Percentile, median Rank-order correlation , Friedman ANOVA

Interval Differences between objects can be compared; zero point is arbitary

Temperature (Fahrenheit, centigrade)

Attitudes, opinions, index numbers

Range, mean, standard deviation

Product-moment correlations, t- tests, ANOVA, regression,factor analysis

Ratio Zero point is fixed; ratios of scale values can be computed

Length , weight

Age, income, costs, sales, market shares

Geometric mean, harmonic mean

Coefficient of variation

Examples in Marketing Research• Nominal Scale – Identifying respondents, brands, attributes, stores, & other objects.



• Ratio Scale - Time to Finish in Seconds*************************************************• Nominal Scale – Identifying respondents, brands, attributes, stores, & other objects.



• Ratio Scale - Time to Finish in Seconds


• Nominal Scale – The numbers in a nominal scale do not reflect the amount of the characteristic possessed by the objects.

•A high social security number does not imply that the person is in some way superior to those with lower social security number.

Example : Project : Department StoreNumbers 1 to 10 were assigned to the stores. Number 9 was assigned to SearsThis does not imply that Sears was inferior or superior to others.

But it is meaningful to say that 75 % of respondents patronized store 9.

• Ordinal Scale – Numbers are assigned to objects to indicate the relative extent to which the objects possess some characteristic.

• Indicates relative position,not the magnitude of differences between the objects.• The object ranked first has more characteristic as compared to the object ranked second.

Examples Quality rankings, rankings of teams in a tournament, socioeconomic class & occupational status.**************************Interval Scale

Used to measure relative attitudes, opinions, perceptions, and preferences

• Interval Scale – Numerically equal distances on the scale represent equal values in the characteristic being measured.

• Contains all the information of an ordinal scale, but it also allows you to compare the differences between objects.

Positive linear transformation of the form

y =a+bx will preserve the properties of the scale.x = original scale value ,y=transformed valueb is the positive constant, a is any constant.

Two interval scales that rate objects A,B,C,D as 1,2,3 & 4 or as 22,24,26 & 28 are equivalent.

• Interval Scale – Numerically equal distances on the scale represent equal values in the characteristic being measured.


• Contains all the information of an ordinal scale, but it also allows you to compare the differences between objects.

Positive linear transformation of the form

y =a+bx will preserve the properties of the scale.

x = original scale value ,y=transformed value

b is the positive constant, a is any constant.

Two interval scales that rate objects A,B,C,D as 1,2,3 & 4 or as 22,24,26 & 28 are equivalent.

Example : Temperature Scale

Attitudinal data obtained from rating scales are often treated as interval data.

Primary Scales of Measurement

Nominal Scale Ordinal Scale Interval Scale (Preference Ratings)

Ratio Scale

No Store Preference Rankings

1 to 7 11 to 17 $ spent last 3 months

1 Parisian 7 79 5 15 02 Macy’s 2 25 7 17 2003 K mart 8 82 4 14 0

4 Kohl’s 3 30 6 16 100

5 JC Penny 1 10 7 17 250

6 Neiman-Marcus

5 53 5 15 35

7 Marshal's 9 95 4 14 0

8 Sak’s Fifth Avenue

6 61 5 15 100

9 Sears 4 45 6 16 0

10 Wal-Mart 10 115 2 12 10

Comparison of Scaling Techniques


Scaling Techniques

Comparative Scales

Non Comparative

Scales

Paired Comparison

Rank OrderConstant

Sum

Q-Sort& other

procedures

Continuous Rating Scales

ItemizedRating Scales

Likert

Semantic Differential

Stapel

Scaling Techniques

Comparative Scales

Non Comparative

Scales

Paired Comparison

Rank OrderConstant

Sum

Q-Sort& other

procedures

Continuous Rating Scales

ItemizedRating Scales

LikertSemantic Differential

Stapel

Scaling Techniques

• Comparative Scales – Involve the direct comparison of stimulus objects.

Example respondents may be asked whether they prefer coke or Pepsi.

Comparative scales must be interpreted in relative terms and have only ordinal or rank order properties.Tend to reduce the halo or carryover effects from one judgment to another.To compare RC Cola to coke and pepsi the researcher would have to do a new study.

Non Comparative Scales : Monadic or metric scales each object is scaled independently or the others in the stimulus set.

Example : Respondents might be asked to evaluate coke on a 1-6 preference scale( 1= not at all preferred , 6= greatly preferred)

Paired Comparison

• Respondent is presented with two objects and asked to select one according to some criterion.

• The data obtained are ordinal in nature.


• Used frequently when stimulus objects are physical products.

We are going to present you with 10 pairs of shampoo brands, for each pair , please indicate which one of the two brands of shampoo in the pair you would prefer for personal use

Clinic Head & Shoulders SunsilkClinic 1 0 1

Head & Shoulders 0 1 1

Sunsilk 1 0 1

Number of times preferred

2 1 3

Rank Order Scaling

Respondents are presented with several objects simultaneously and asked to order or rank them according to some criterion.

Brand Rank Order1. Crest2. Colgate3. Aim4. Gleem5. Close up6. pepsodent

Constant Sum Scaling

In constant sum scaling , respondents allocate a constant sum of units such as points , dollars , chips among a set of stimulus objects with respect to some criterion.


Below are the attributes of eight toilet soaps. Please allocate 100 points among the attributes , so that your allocation reflects the relative importance you attach to each attribute.

Allows for fine discrimination among stimulus objects without requiring too much time.

It has two primary disadvantages. Respondents may allocate more or fewer units than those specified.

For example a respondent may allocate more or fewer units than those specified

Average response of three segmentsAttribute Segment 1 Segment 2 Segment 3Mildness 8 17 4

Lather 53 12 6

Packaging 15 9 10

Moisturizing 13 19 12

Sum

Q Sort & other Procedures

Qsort Scaling was developed to discriminate quickly among a

relatively large number of objects.

This technique uses a rank order procedure in which objects are

sorted into piles based on similarity with respect to some criterion.

Respondents are given 100 attitude statements on individual cards

and asked to place them into 11 piles, ranging from “ most higly

agreed with “ to “ least highly agreed with “

The number of objects to be sorted should not be less than 60 nor

more than 140;

60 to 90 objects is a reasonable range.

Magnitude estimation : Numbers are assigned to objects such

that ratios between the assigned numbers reflect ratios on the

specified criterion.


Example : respondents may be asked to indicate whether they

agree or disagree with each series of statements measuring attitude

towards department stores.

Non Comparative Scaling TechniquesScale Basic

CharacteristicExamples Advantages Disadvantages

Continuous rating scale

Place a mark on a continuous line

Reaction to TV commercials

Easy to construct Scoring can be cumbersome unless computerized

Itemized rating scales

Likert Scale Degree of agreement on 1(strongly disagree) to 5(strongly agree) scale

Measurement of attitudes

Easy to construct ,administer , and understand

More time consuming

Semantic Differential

Seven point scale with bipolar labels

Brand product and company images

versatile Controversy as to whether the data are interval

Stapel scale Unipolar ten-point scale, -5 to + 5 without a neutral point(zero)

Measurement of attitudes and images

Easy to construct; administered over telephone

Confusing and difficult to apply

Continuous rating scale : also called as graphic rating scale, respondents rate theobjects by placing a mark at the appropriate position on a line that runs from oneextreme of the criterion variable to the other.

How would you rate Sears as a department store

Version 1Probably the worst _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Probably the bestOnce the respondent has provided the ratings, the researcher divides the lineinto as many categories as desired and assigns scores based on thecategories into which the ratings fall.

Probably the worst _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Probably the best 0 10 20 30 40 50 60 70 80 90 100


Itemized rating scale : The respondents are provided with a scale that has a number or brief description associated with eachcategory. The categories are ordered in terms of scale position and the respondents are required to selectthe specified category that best describes the object being rated.

Likert scale : Rensis likert .

Requires respondents to indicate a degree of agreement or disagreement with each of a series of statementsabout the stimulus objects.Each scale item has 5 response categories , ranging from “ strongly disagree” to “ strongly agree “.

Example : Different opinions about sears store1 = strongly disagree2 = disagree3 = neither agree nor disagree4 = agree5 = strongly agree

Semantic Differential Scale : It is a 7 point rating scale with end points associated with bipolar labels that have semanticmeaning. In a typical application, respondents rate objects on a number of itemized, 7 pointrating scales bounded at each end by one of two bipolar adjectives such as cold and warm.

Sears is Powerful _:_:_:_:_:_:X-:_:_:WeakUnrealible_:_:_::_:_:_:_:_:_X:_:_:Reliable


Stapel scale : Jan StapelUnipolar rating scale with 10 categories numbered from -5 to +5 withoutneutral point(zero). This scale is usually represented vertically .Respondents are asked to indicate how accurately or inaccurately each termdescribes the object by selecting an appropriate numerical response category.

ExampleSelect a + number for the phrases you think describe the store accurately.The more accurately you thing the phrase describes the larger the + number you should chooseSelect a - number for phrases you think do not describe it accurately. The less accurately you think the phrase describes the larger the – number u should chooseForm Sears High Quality Poor Service+ 5 +5 -1 -1+4 +4 -3 -2

High Quality Poor Service

- 1 -1

-2 X -2

-3 -3

-4 -4

The staple scale can be administered over the telephone.

Major decisions the researcher must take while constructing the scales.

1. Number of scale categories to use

2. Balanced versus unbalanced scale

3. Odd or even number of categories

4. Forced versus non forced choice

5. The nature and degree of the verbal description

6. The physical form of the scale

Number of Scale Categories : Two conflicting considerations are involved in deciding the

number of scale categories. The greater the number of scale categories , the finer the

discrimination among stimulus objects that is possible. On the other hand respondents

cannot handle more than a few categories.


Balanced versus unbalanced scale: The number of favorable and unfavorable categories are

equal; in an unbalanced scale they are unequal.

Eg :

Balanced Scale Unbalanced Scale

Jovan Musk for men is Jovan Musk for men is

Extremely good Extremely good

Very good Very good

Good Good

Bad Somewhat Good

Very Bad Bad

Extremely bad Very Bad

The scale should be balanced in order to obtain objective data

If the distribution of responses is likely to be skewed , either positively or negatively ,an

unbalanced scale with more categories in the direction of skewness may be appropriate.

an unbalanced scale is used the nature and degree of unbalance in the scale should be taken

into account in data analysis.

Odd or Even Number of Categories

With an odd number of categories the middle scale position is generally designated as neutral or

impartial. The presence , position and labeling of a neutral category can have a significant

influence on the response. The likert scale is a balanced rating scale with an odd number of

categories and a neutral point. The decision to use an odd or even number of categories depends on

whether some of the respondents may be neutral on the response being measured. If a neutral or

indifferent response is possible from at least some of the respondents , an odd number or categories

should be used. If on the other hand the researcher wants to force a response or believes that no

neutral or indifferent response exists, a rating scale with an even number of categories should be

used.


Forced versus Nonforced Scales : On forced rating scales, the respondents are forced to express

an opinion because a “ no opinion “ option is not provided. In such a case respondents without an

opinion may mark the middle scale position. If a sufficient proportion of the respondents do not have

opinions on the topic, marking the middle position will distort measures of central tendency and

variance. In situations where the respondents are expected to have no opinion , as opposed to

simply being reluctant to disclose it , the accuracy of data may be improved by a no forced scale

that includes a “ no opinion “ category.

Nature & Degree of Verbal Description : The nature and degree of verbal description associated

with scale categories varies considerably and can have an effect on the responses. Scale categories

may have verbal numerical or even pictorial descriptions. Furthermore the researcher must decide

whether to label every scale category, some scale categories or only extreme scale categories.

Labeling can be done to reduce the scale ambiguity. The category description should be located as

close to the response categories as possible.

Physical Form or Configuration : A number of options are available with respect to scale form or

configuration. Scales can be presented vertically or horizontally . Categories can be expressed by

boxes, discrete lines, or units on a continuum and may not have numbers assigned to them. If

numerical values are used , they may be positive , negative or both.

Two unique rating scale configurations used in marketing research are

1) thermometer scale : the higher the temperature the more favorable the evaluation

2) smiling face scale : happier faces indicate more favorable evaluations.

Summary of Itemized Rating Scale Decisions1 Number of categories Although there is no single , optimal number, traditional guidelines

suggest that there should be between five and nine categories

2 Balanced versus unbalanced

the scale should be balanced to obtain objective data

3 Odd or even number of categories

if a neutral or indifferent scale response is possible from at least some of the respondents, an odd number of categories should be used

4 Forced versus nonforced In situations where the respondents are expected to have no opinion , the accuracy of data may be improved by a nonforced scale.

5 Verbal description an argument can be made for labeling all or many scale categories. The category descriptions should be located as close to the response categories as possible

6 Physical Form A number of options should be tried and the best one selected


Some commonly used scales in marketing :

Construct Scale Descriptors

1 Attitude very bad badneither bad nor good good

very good

2 ImportanceNot at all important

Not Important Neutral Important

Very Important

3 SatisfactionVery dissatisfied dissatisfied

neither dissatisfied nor satisfied satisfied

very satisfied

4 Purchase Intentdefinitely will not buy

probably will

might nor might not buy

probably will buy

definitely will buy

5purchase Frequency never rarely sometimes often

very often

Multi – item Scales

Scale Evaluation :


Develop a Theory

Generate an initial pool of items: theory ,secondary data and qualitative research

Select a reduced set of items based on qualitative judgment

Collect data from a large pretest sample

Perform statistical analysis

Develop a purified scale

Collect more data from a different sample

Evaluate scale reliability , validity and generalizability

Prepare the final scale

94

A measurement is a number that reflects some characteristic of an object. A measurement is

not true value of the characteristic of interest but rather an observation of it.

Measurement Accuracy : A measurement is not a true characteristic of interest but rather an

observation of it. Measurement error results in the measurement or observed score being different

from the true score of the characteristic being measured . The true score model provides a framework

for understanding the accuracy of measurement .

Xo = Xt + Xs + Xr

Xo = the observed score or measurement

Xt = the true score of the characteristic

Xr= random error.

Sampling Process


Scale Evaluation

Reliability ValidityGeneralizabili

ty

Test/ RetestAlternative

FormsInternal

ConsistencyContent Criterion Construct

Convergent

Discriminant

Nomological

95

1) How population is defined .

It is the specific group of people, firms, conditions , activities which form the pivotal point of research project. For developing a sample it becomes the primary duty of the researcher to define the population from which to draw the sample.

What is the population for your project ?

2) Sampling frame Developing a sampling frame. A sampling frame may be defined as the listing of the general components of the individual units that comprise the defined population.

Eg : a fertilizer manufacturing company wants to test its fertilizer efficiency with regard to crop yields. The sample units here can include fields.The sampling frame would be the set of coordinates used for selecting various agricultural fields.

3) Sampling methods . Whether probability or non probability sampling. i) Probability sampling if sampling error is to be estimated. ii) Non Probability sampling if the research is required to defend the randomness in the selection of sample units.iii)it is better to use non probability sampling when a frame does not exist nor one can be developed .iv)non probability sampling should be used if the researcher has limited knowledge about the statistical aspects of sampling. 4) Selecting Sample size Trade off between cost and sampling error , larger sample keeps error small but increases sampling cost. So how to select the correct sample size.

Probability Sampling : Degree of precision , knowledge of standard error and confidence intervals.

Non Probability Sampling : The central limit theorem or the principle of normal distribution is not applicable in the case of non probability sampling. It is the researchers intuition and judgment which makes the decision about the size of the sample.

The size of the sample in Non Probability sampling is generally 1/10th of the population size.


Sampling Metrics

Standard Errors of Mean & Proportion

i) Sampling of variables involves a numerical value such as average height, weight , expenditure here we estimate the standard error of the mean. Sigma x barii) Sampling of proportions is expressed in dichotomous terms. Eg : good or bad, heavy or light , high or low here standard error of the proportion is calculated.

Estimating standard error of the mean:The following formula is used :

Sx bar = (s/ sqrt (n)) * (sqrt((N-n)/N))

S x bar = standard error of the mean using sample dataS = standard deviation derived from the sample.n = sample sizeN = Population SizeExample : We are studying the spending habits of college students on entertainment. If there is a population of 6000 students and we take a sample of 500 students, we find that the average student spends Rs 100 per month on entertainment, This figure of Rs 100 is taken as an average for all 6000 students population . But is this assumption valid ? suppose if we take another sample of 500 students and survey them would it be valid . The results would actually be different . It is all because of sampling error ?Now Average spending is 100 .Standard deviations is say Rs 15

Estimate the standard error of the mean S x bar. = 15/sqrt(500) * Sqrt((600 – 500)/6000) = 0.64

The formula for constructing a confidence interval is :

X bar + - 1.96(S xbar) = 95 percent confidence interval

X bar + - 2.57(S x bar) = 99 percent confidence interval

X bar = sample mean

S xbar = standard error of the mean of the sample.

Substituting the sample data at 95 % confidence interval Xbar + - 1.96 (S xbar) = 95 % confidence interval(Rs 100 ) +- 1.96 ( Rs 0.64)(Rs 100) +- 1.26Rs 98.74 – Rs 101.26


This means that if a large number of probability samples ( size 500 ) were taken from the particular city college student body, 95 % of these samples would contain the actual mean of the universe within an interval of +- 1.96 standard errors of their mean values. This is want the 95 % confidence interval of Rs 98.74 to Rs 101.26 mean. It is assumed that this sample with a mean of Rs 100 is among those 95 percent.

99% interval is derived as Xbar + - 2.57 ( Sx bar) = 99 %Rs 100 +- 2.57(0.64)Rs 100 +- 1.64Rs 98.36 – Rs 101.64

What does this 99 % mean. It means that if a large number of samples ( size 500) are taken from the student body, 99% of these would contain the actual mean of the universe within an interval +- 2.57 of standard errors of their mean value.

So at 95% confidence interval 6000 * Rs 98.74 – 6000 * Rs 101.26Rs 592.440 – Rs 607.560

Standard error of proportion :

The formula for computing the standard error of proportion is

Sp = Sqrt ((( P.q)/n )*( (N-n)/N)))

Sp = standard error of the proportion P = proportion of sample possessing a certain attribute.

q or (1-p) = proportion of sample not having that attribute.

Example : A study of 5000 students was conducted by taking a sample of 400 students. It was regarding the students attitudes towards a company’s products. It was found that 88 % of the students (22% ), had favorable attitudes toward the company’s products while 312 students (78 %) held unfavorable ones. Based on this information, what would be the 95 % confidence interval depicting the attitude of the entire student body toward company’s products?

Standard error of the proportion is computed as follows

Sp = Sqrt((P.q)/n * (N-n)/N))

Sp = Sqrt((22.78/400 ) *(5000 – 400)/5000)) = 4.29 * 0.92 = 1.99 %Now the formula for 95 % confidence interval when attributes are involved is P =- 1.96(Sp)= 22 % + - 1.96(1.99 %)= 22% + - 3.9 %


= 18.1 % - 25.9 %Based on this 18 to 26 % confidence interval we can say that only 12 % to 26 % of the college students have a favorable attitude towards the company’s products.

Using standard error of the mean to determine the sample size

S xbar = s/ Sqrt(n)

Squaring both sides Sx bar square = s square / n

n = S 2 / S xbar 2

Example : Colgate Palmolive India Ltd wants to know how much does a male student spend annually on toothpastes. They selected Punjab agricultural university , ludhiana as population and took sample for the study . Assume that there were 18,000 students of which 8,300 were males. The company wants to be reasonably confident that the amount obtained will be within +- Rs 2 of the average male students real expenditure. How large a sample will be needed to achieve this precision. The researcher is fairly confident that the range of annual expenditures for toothpastes by male students at Punjab agricultural university is from Rs 0 to Rs 80.

Using the above formula

n = S 2/ S xbar 2

n= S 2 / Sx 2 n=(Rs 80 / 6) 2 / Sx 2

= (Rs 13.5 ) 2 / Sx 2

Here we divided S / 6 because in a normally distributed set of values , the range encompasses approximately three standard deviations . Thus 80 / 6 gives a reasonable estimate of 1 standard deviation.

The result is total estimated range/ 6 = estimated S

At 95 % confidence interval involves 1.96 Sx . in our example that means 1.96 Sx = +- Rs 2

Sx bar = Rs 2/ 1.96 = Rs 1.02

Substituting this value in original formula , we get

n= (Rs 13.50) 2 / Rs 1.02 2

= 182.25/ 1.04 = 175

Thus to be 95 % confident and the sample mean within +- Rs 2 of the population mean a sample of 175 male students must be contacted.

If colgate Palmolive needs greater precision, that is a smaller interval of +- Re 1 requires a sample of 701 students as under


1.96 Sx = Re1 So Sx = 0.51

n=(Rs 13.5)2 / 0.51 2 = 182.25 / 0.26 = 701Now small change in precision sought ( + - Re 1 vs +- 2 ) requires almost 4 fold increase in the size of the sample needed.

Now small change in precision size i.e if the n value generated by the above procedure i.e is more than 5 % of the population ( n / N > 5% ) the sample size should be revised downward . The downward revision would take the following form.

n alpha = n / 1 + n / N where n alpha = adjusted sample.

Now in the above example sample size was 175, this is only 2 % of the total population of 8300 students , so no downward adjustment is needed.

In case 701 is the sample size ( 8.5 % of the population ) downward adjustment is needed.

n alpha = 701 / 1 + 701 / 8300 = 646

Sample size in case of attributes


1) ABC so is a leading firm of dry cleaners with branches in delhi and the large towns in western U.P . The company has been thinking of setting up a new shop in agra which already has four other dry cleaning shops, one of which is a branch of a national company at par with the ABC. ABC is concerned about the reaction of customers and potential customers to the opening of its branch in Agra, and is of the opinion that one of the three possibilities can take place

i) increased market shareii) no change in the present operationsiii) reduced market share.

The company assigns a value to each of these outcomes.

Outcome Value in Rs (lakhs)

Outcome Probability

Increased Share 3 Increased Share 0.2No Change 1 No Change 0.5

Reduced Share -2 Reduced Share 0.3

Probabilities assigned to the outcomes after discussion’s with the managers .

The expected monetary value can be calculated as follows EMV ( Expected monetary Value ) = ( Rx 3,00,00 x 0.2 ) + ( Rs 1,00,000 X 0.5 ) + ( Rs -2,00,000 X 0.3) = Rs 60,000 + Rs 50,000 – Rs 60,000 = Rs 50,000~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~2) ABC co whose major competitor is XYZ co, has to decide whether or not to introduce a new product. Although completely reliable information regarding XYZ’s plans is not available , the marketing manager of ABC has assumed certain probabilities on the basis of past experience. He has prepared the following conditional pay off ?XYZ Company Strategy Probability C1 = 0.6 C2 = 0.4

Rs (Million) Rs (Million)Strategy S1 6 10 S2 4 15

C1 : indicates that XYZ co introduces new productC2 : indicates that XYZ co does not introduce new productS1 Indicates that ABC decides to introduce new productS2 indicates that ABC decides not to introduce new product

The question is : what is the expected value of perfect information ?We first calculate the EMV of the two strategies : EMV of the two strategies : EMV of S1 = (6x 0.6) + (10 X 0.4 ) = 3.6 + 4 = Rs 7.6 million


EMV of S2 = (4 X 0.6) +(15X 0.4) = 2 .4 + 6 = Rs 8.4 Million On the basis of these calculations, ABC should choose strategy S2, which is the optimal strategy.From the pay off matrix given above , it is possible to calculate a tentative value for perfect information. Assume that the ABC co is considering approaching a consultancy firm to provide additional information relating to XYZ’s plans .

IT should be noted that ABC cannot pay any arbitary amount demanded by this consultancy firm. It should therefore fix up an upper limit beyond which it would like to pay. How is this limit determined ?

Assume that ABC Co is fully acquainted with the plans of its competitior XYZ co . In such a case ABC will choose S1 and gain a payoff of Rs 6 million when XYZ introduces a new product. This is because S2 will yield only Rx 4 million. Likewise , ABC will choose S2 and gain a pay-off of Rx 15 million when it is certain that XYZ will not introduce a new product. This is because s1 will yield only Rx 10 million.Now from the information given above , it is known that XYZ will choose C1 during 60 percent of the time and C2 during 40 percent of the time . In view of this , the expected monetary value under certainty will be as follows : EMV ( C ) = (6 x 0.6 ) + ( 15 X 0.4 ) = 3.6 + 6 = Rs 9.6 millionNow the expected monetary value of perfect information is to be determined . This is the difference between the expected monetary value under certainty and expected monetary value of the optimal strategy calculated under conditions of uncertainity (i.e) EMV ( UC) . Thus the expected monetary value of perfect information is :

EMVPI = EMV( C ) - EMV(UC) = Rs ( 9.6 – 8.4 ) million = Rs 1.2 million3)A marketing manager of a soft drink manufacturing company is seriously considering whether to undertake a special promotion . The two options before him are i) Run a special promotion.ii)Do not run a special promotion.The following table gives the probabilities assigned by the marketing manager to the three possible outcomes Viz . very favourable consumer’s reaction , favourable consumer’s reaction and unfavourable consumer’s reaction.Possible consumerReaction

Alternative courses of action Probabilities of Consumer Reactions

A1 Rs A2 RsVery favourable 1,00,00,000 0 0.7Favourable 10,00,000 0 0.1Unfavourable -50,00,000 0 0.2

Calculate the Expected monetary value of perfect information ?


mimd lexicon[1]

Documents

research segmentation

market research

type of research

qualitative research

marketing problems

marketing environment

marketing opportunities

understanding of marketing