Research designs

Research Designs in Marketing Research

BY: Sweta Gupta

Research designs
A research design is a framework or blueprint for conducting the business research project. It details the procedures necessary for obtaining the information needed to structure or solve business research problems.

Research Design

The research design is the master plan specifying the methods and procedures for collecting and analyzing the needed information.

Components of a Research Design

Define the information needed
Design the exploratory, descriptive, and/or causal phases of the research

Specify the measurement and scaling procedures

Construct and pretest a questionnaire (interviewing form) or an appropriate form for

data collection Specify the sampling process and sample size Develop a plan of data analysis

Types of Research Design

Three traditional categories of research design: Exploratory Descriptive Causal/Experimental The choice of the most appropriate design depends largely on the objectives of the research and how much is known about the problem and these objectives.

A Classification of Research Designs

Research Design

Exploratory Research Design

Conclusive Research Design

Descriptive Research

Causal Research

Cross-Sectional Design

Longitudinal Design

Single CrossSectional Design

Multiple CrossSectional Design

Uncertainty Influences The Type Of Research

CAUSAL OR DESCRIPTIVE

COMPLETELY CERTAIN

ABSOLUTE AMBIGUITY

EXPLORATORY

Exploratory & Conclusive Research Differences

Exploratory
Objective:
To provide insights and understanding. Information needed is defined only loosely. Research process is flexible and unstructured. Sample is small and nonrepresentative. Analysis of primary data is qualitative.

Conclusive
To test specific hypotheses and examine relationships. Information needed is clearly defined. Research process is formal and structured. Sample is large and representative. Data analysis is quantitative.

Characteristics:

Findings /Results: Outcome:

Tentative.

Conclusive.

Generally followed by further exploratory or conclusive research.

Findings used as input into decision making.

A Comparison of Basic Research Designs

Exploratory
Objective:

Descriptive

Causal
Determine cause and effect relationships Manipulation of one or more independent variables Control of other mediating variables Experiments

Discovery of ideas Describe market and insights characteristics or functions Flexible, versatile Marked by the prior formulation of specific hypotheses Preplanned and structured design

Characteristics:

Often the front end of total research design Expert surveys Pilot surveys Secondary data Qualitative research

Methods:

Secondary data Surveys Panels Observation and other data

Degree of Problem Definition

Exploratory Research (Unaware of Problem) Our sales are declining and we dont know why. Would people be interested in our new product idea? Descriptive Research (Aware of Problem) Causal Research (Problem Clearly Defined)

What kind of people are buying Will buyers purchase more of our product? Who buys our our products in a new package? competitors product? Which of two advertising What features do buyers prefer campaigns is more effective? in our product?

10

Basic Research Objectives & Research Design

Research Objective Appropriate Design

To gain background information, to define terms, to clarify Exploratory problems and develop hypotheses, to establish research priorities, to develop questions to be answered To describe and measure marketing phenomena at a point Descriptive in time To determine causality, test hypotheses, to make if-then Causal/Experimental statements, to answer questions

11

Research Design: Exploratory Research

Exploratory research is most commonly unstructured, informal research that is undertaken to gain background information about the general nature of the research problem. Exploratory research is usually conducted when the researcher does not know much about the problem and needs additional information or desires new or more recent information.

12

Research Design: Exploratory Research

Exploratory research is used in a number of situations: To gain background information To define terms To clarify problems and hypotheses To establish research priorities

13

Research Design: Exploratory Research

A variety of methods are available to conduct exploratory research: Secondary Data Analysis Experience Surveys Case Analysis Focus Groups Projective Techniques

14

Uses of Exploratory Research

Formulate a problem or define a problem more

precisely Identify alternative courses of action Develop hypotheses Isolate key variables and relationships for further examination Gain insights for developing an approach to the problem Establish priorities for further research

Research Design: Descriptive Research

Descriptive research is undertaken to provide answers to questions of who, what, where, when, and how but not why. Two basic classifications: Cross-sectional studies Longitudinal studies

16

Methods of Descriptive Research

Secondary data analyzed in a quantitative as opposed to a qualitative manner
Surveys

Panels
Observational and other data

Descriptive Research: Cross Sectional Studies

Cross-sectional studies measure units from a
sample of the population at only one point in time. Sample surveys are cross-sectional studies whose samples are drawn in such a way as to be representative of a specific population. On-line survey research is being used to collect data for cross-sectional surveys at a faster rate of speed.

18

Cross-sectional Designs
Involve the collection of information from any given sample of

population elements only once. In single cross-sectional designs, there is only one sample of respondents and information is obtained from this sample only once. In multiple cross-sectional designs, 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. Cohort analysis consists of a 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.

Descriptive Research: Longitudinal Studies

Longitudinal studies repeatedly draw sample
units of a population over time. One method is to draw different units from the same sampling frame. A second method is to use a panel where the same people are asked to respond periodically. On-line survey research firms recruit panel members to respond to online queries.

20

Descriptive Research: Longitudinal Studies

Two types of panels:

Continuous panels ask panel members the

same questions on each panel measurement.

Discontinuous panels vary questions from

one time to the next. Longitudinal data used for: Market tracking Brand-switching Attitude and image checks

21

Use of Descriptive Research

To describe the characteristics of relevant groups, such as consumers, salespeople, organizations, or market areas. To estimate the percentage of units in a specified population exhibiting a certain behavior. To determine the perceptions of product characteristics. To determine the degree to which marketing variables are associated. To make specific predictions

Research Design: Causal/Experimental Research

Causality may be thought of as understanding a phenomenon in terms of conditional statements of the form If x, then y. Causal relationships are typically determined by the use of experiments, but other methods are also used.

23

Experiments
An experiment is defined as a research process that allows study of one or more variables which can be manipulated under conditions that permits collection of data that show the effect of such variables in unconfused fashion Independent variables: those over which the researcher has control and wishes to manipulate i.e. package size, ad copy, price. Dependent variables: those over which the researcher has little to no direct control, but has a strong interest in testing i.e. sales, profit, market share. Extraneous variables: those that may effect a dependent variable but are not independent variables.

24

Experimental Design
An experimental design is a procedure for devising an experimental setting such that a change in the dependent variable may be solely attributed to a change in an independent variable. Symbols of an experimental design: O = measurement of a dependent variable X = manipulation, or change, of an independent variable R = random assignment of subjects to experimental and control groups E = experimental effect
25

Basic Principles of Research Design

Principle of Replication Experiments should be repeated more than once Increases statistical accuracy & precision Principle of Randomization Protects against effects of extraneous factors Variations can be termed as Chance Principle of Local Control Effects of extraneous factors can be minimized by using best methods & measuring instruments

26

Uses of Casual Research

To understand which variables are the cause

(independent variables) and which variables are the effect (dependent variables) of a phenomenon To determine the nature of the relationship between the causal variables and the effect to be predicted METHOD: Experiments

Important Experimental Designs

Informal

Uses less sophisticated form of analysis Allows study of one variable at a time

Formal

Uses precise statistical methods for analysis Allows study of more than one variables at a time Interaction between variables can be studied

28

Informal Experimental Designs

After only Before - After Before - After with Control Group After Only with Control Group

29

Formal Experimental Designs

Completely Randomized Randomized Block Latin Square Factorial

30

Informal Experimental Designs

After Only Dependent variable is measured only after independent variable is introduced Purchase of Pepsi through coupon redemption in advertisement Not a good design because Before the event response is not measured Sometimes used for new products when Before measurements is known to be zero

31

Informal Experimental Designs

Before After Design
Experimental Group

Before measurement X1 Experimental variable Introduced X 1 Yes After measurement X2 Effect of Experimental variable =
( X 2 X1 )

Some of the aspects that may cause distortion are

History Effect Maturation Pre Test Effect Variety introduced by the researcher

32

Informal Experimental Designs

Before After with Control Group Design
Before measurement Exptal variable Introduced After measurement
1

Experimental Control Group Group Y1 X

Yes X2 No Y2
( X 2 X 1 ) (Y2 Y1 )

Net Effect of Experimental variable = Some problems are

Mortality Effect Interaction Effect

33

Informal Experimental Designs

After only with control group Experimental Group - X O1 Control Group O2 (O2-O1)

34

Experimental Designs - CRD

Completely Randomized Design

(Describe the experiment of regular & special trainings) Always use simple random for any selection Involves two principles: Replication & Randomization For analysis One way ANOVA is used Shortcoming: Can not control the effect of extraneous variable: Quality of training by different trainers

35

Control Group

Treatment B

Populatio n

Randmly Assigned

Expimtal Group

Independent Variable
36

Randomly select Sample

Treatment A

Experimental Designs - CRD

Experimental Designs - RBD

Randomized Block Design
Improvement over Completely Randomized Design:

Principle of local control is also applied ANOVA is used for analysis Describe the experiment: To measure the effect of four different tests on students having different IQ level

Divide students according to IQ level One student from each group is selected randomly Tests can be taken in random order: eliminates effect of fatigue or experience gained by taking repeat exams

37

Experimental Designs - RBD

Very Low IQ Student A Low IQ Average IQ High IQ Very High IQ Student E Student B Student C Student D

Form 1
Form 2 Form 3 Form 4

82
90 86 93

67
68 83 77

57
54 51 60

71
70 69 65

73
81 84 71

Experimental Designs - RBD

Four different forms of standardize test were given to

each of 5 students and the scores are as they obtained above. If each student separately randomize the order in which he or she took the test, we refer to the design as RB design . The purpose is to take care of extraneous variable (fatigue or experience) because of test repetition.

Experimental Designs - Latin Square

Latin Square Design
To evaluate the effect of three advertisement (A, B & C) in

a stores the researcher also want to know the interest in the store. Interest in the STORES Store Patronage High Medium Low High A C B Medium C B A Low B A C
40

Experimental Designs Latin Square

A statistical design that allows for the statistical

control of two non-interacting external variables in addition to the manipulation of the independent variables. For analysis- two way ANOVA

41

Experimental Designs - Factorial

Effect of two or more variables at various levels can be studied simultaneously and Interaction between variables can also be studied Describe the experiment of Flavours & Sugar Intensity for finding the choice of consumers for a new soft drink Helps in studying effect of two variables simultaneously Two way ANOVA is used for analysis
42

Experimental Designs Factorial

Control Variable Level I Level II Level III Experimental Variable Treatment A Cell 1 Cell 2 Cell 3 Treatment B Cell 4 Cell 5 Cell 6 Treatment C Cell 7 Cell 8 Cell 9 Treatment D Cell 10 Cell 11 Cell 12

Characteristics of sound measurement

Characteristics of a good measurement tool
Accurate counter or indicator of what is

being measured Easy and efficient to use

Three major criteria for evaluating a measurement tool

Validity

Reliability
Practicality
44

Validity
External & Internal validity Internal validity: One aspect
Ability of a research instrument to measure

what it is purported to measure Does the instrument really measure what its designer claims it does?

Classification of validity
Content validity

Criterion related validity

Construct validity
45

Validity
Content Validity
Degree to which the content of the items

adequately represents the universe of all relevant items under study To measure Corporate Image what knowledge, attitude and opinions are relevant to be included for measurement It is judgmental or panel evaluation

46

Validity
Criterion-Related Validity
Reflects the success of measures used for

prediction or estimation

Predictive Concurrent

Opinion questionnaire that can correctly

forecast the outcome of a union election has predictive validity An observational method that can correctly classify families into income classes has concurrent validity May sound very easy but for some variables it may prove to be difficult to secure correct figure for example Income of a family

47

Validity
Construct Validity
To measure attitude, aptitude, personality

tests and infer is very difficult No empirical validation seems possible This is typical construct validity

To evaluate construct validity both theory and measuring instrument need to be considered To determine the effect of ceremony on organizational culture

48

Reliability
Example: Weighing Scale It is Valid & Reliable if it measures your weight correctly If it consistently overweighs you by 3 Kgs, it is Reliable but not valid If it weighs erratically, it is neither valid nor reliable Reliable instruments are free from random

errors, and work well under different times and different conditions

49

Reliability
Three perspectives of reliability Stability, Equivalence & Internal consistency Stability

Producing consistent results with repeated measurements under similar environment Is concerned with personal and situational variations from one time to another Test-retest

Research Design: Dr. Dey 50

Reliability
Equivalence

How much errors can be introduced by different investigators or different samples Is concerned with variations at one point in time among observers or samples Parallel forms Tests the homogeneity among the items Split-half Kuder Richardson Formula 20 Cronbachs alpha

Internal Consistency

51

Practicality
Measurement process has to be Scientifically: Valid & Reliable Operationally: Practical Practicality has been defined as Economy Convenience Interpretability

52

Types of Experiments
Two broad classes: Laboratory experiments: those in which the independent variable is manipulated and measures of the dependent variable are taken in a contrived, artificial setting for the purpose of controlling the many possible extraneous variables that may affect the dependent variable Field experiments: those in which the independent variables are manipulated and measurements of the dependent variable are made on test units in their natural setting
Research Design: Dr. Dey 53

Problems with Experimental Designs

Lack of theoretical base: Unless enough

data, Hypothesis formulation may not be possible Time factor Cost of conducting experiments is high Administrative problems of coordination & execution
54

Experiments: Test Marketing

Test marketing is the phrase commonly used to indicate an experiment, study, or test that is conducted in a field setting. Two broad classes: To test the sales potential for a new product or service To test variations in the marketing mix for a product or service

55

Types of Test Markets

Standard test market: one in which the firm tests the product and/or marketing mix variables through the companies normal distribution channels Controlled test markets: ones that are conducted by outside research firms that guarantee distribution of the product through pre-specified types and numbers of distributors

56

Types of Test Marketscont.

Electronic test markets: those in which a panel of consumers has agreed to carry identification cards that each consumer presents when buying goods and services Simulated test markets: those in which a limited amount of data on consumer response to a new product is fed into a model containing certain assumptions regarding planned marketing programs, which generates likely sales volume

57

Test Markets

Test marketing is used in both consumer markets and industrial or B2B markets as well. Lead country test market: test marketing conducted in specific foreign countries that seem good predictors for an entire continent

58

Criteria for Selecting Test Market Cities

Representativeness: Do demographics match the total market? Degree of isolation: Cities are isolated markets; Or not isolated. Ability to control distribution and promotion: Are there pre-existing arrangements to distribute the new product in selected channels of distribution? Are local media designed to test variations in promotional messages?
59

Test Marketing
Pros: Allows most accurate method of forecasting future sales Allows firms the opportunity to pretest marketing mix variables Cons: Does not yield infallible results Are expensive Exposes the new product or service to competitors Takes time to conduct
60

Potential Sources of Error in Research Designs

Total Error

Random Sampling Error

Non-sampling Error

Response Error

Non-response Error

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

Interviewer Error

Respondent Error
Inability Error Unwillingness Error

Respondent Selection Error Questioning Error Recording Error Cheating Error

Errors in Business Research

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 business research project I. Random sampling error is the variation between the true mean value in the population and the true mean value in the original sample. II. Nonsampling error (can be attributed to sources other than sampling)
1) Nonresponse error: when some respondents

included in the sample do not respond (refusals or notat-homes)

Errors in Business Research

2) Response error:
Researcher errors

a) Surrogate information error: the variation between the info needed and sought by the researcher (e.g., instead of info on consumer choices, the researcher obtains info on consumer preferences because the choice process cannot be easily observed) b) Measurement error: the variation between the info sought and info generated (e.g., measuring perceptions rather than preferences)

Errors in Business Research

c) Population definition error: the variation between the actual population relevant to the problem at hand and the pop. as defined by the researcher (e.g., how to define a population of affluent households?)

Errors in Business Research

d) Sampling frame error: the variation between the population defined by the researcher and the population as implied by the sampling frame (e.g., the telephone directory used to generate a list of telephone numbers does not accurately represent the pop. of potential consumers due to unlisted, disconnected, and new numbers in service) e) Data analysis error: e.g., when an inappropriate statistical procedure is used

Errors in Business Research

Interviewer errors

a) Respondent selection error: respondents are selected other than those specified by the sampling design (e.g., a nonreader of a journal is selected rather than a reader to meet a difficult quota requirement) b) Questioning error: e.g., interviewer does not use the exact wording given in a questionn. c) Recording error: errors in hearing, interpreting, and recording the answers d) Cheating error: the interviewer fabricates the answers (e.g., does not ask about income, but then fills in the answer based on personal assessment)

Errors in Business Research

Respondent errors

a) Inability error: because of unfamiliarity, fatigue, boredom, faulty recall, question format, question content, etc. (e.g., a respondent cannot recall the brand of yogurt purchased four weeks ago) b) Unwillingness error:

No answer Intentionally wrong answer (e.g., declares himself as a reader of a prestigious magazine rather than a tabloid)

Thanks