Upload
udit-singh
View
14
Download
0
Tags:
Embed Size (px)
Citation preview
April 19, 2023
2
Topic Outline Nature of, Overview and Classification of Design Developing an appropriate research design Experimental research design: Basic Designs Types & validity of experimental design –
external & internal
Topic & Structure of the lesson
April 19, 2023
3
Learning Outcomes
On completion of this chapter you should be able to understand:
• Understand the major descriptors of research design
• Understand the major types of research designs
• Understand the relationships that exist between variables in causal designs and the steps for evaluating those relationships
RESEARCH DESIGNS:OVERVIEW
April 19, 2023
4
It is a statement of only the essential elements of a study, those that provide the basic guidelines for details of the project.
A research design is like a description of a `model'
RESEARCH DESIGN: DEFINITION
April 19, 2023
5
Research design constitute the blueprint for the collection, measurement and analysis of data.
The essentials of research design:An activity- and time – based plan.A plan always based on the research question.A guide for selecting sources and types of information.A framework for specifying the relationship among the study’s
variables.A procedural outline for every research activity.
Why Research Design is needed?
April 19, 2023
6
Clarity
Relevance
Ease in Analysis and Interpretation
Economy:
Classification of Research Design
April 19, 2023
7
Research Design
ConclusiveResearch
Design
Causal Research
DescriptiveResearch
LongitudinalDesign
Cross –SectionalDesign
MultipleCross –SectionalDesign
SingleCross –
Sectional Design
ExploratoryResearch
Design
Classification of Research Designs:
April 19, 2023
8
Exploratory research (ER) are the simplest, most flexible and most loosely structured designs. As the name suggests, the basic objective of the study is to explore and obtain clarity on the problem situation.
Sample selected is small & non-representative. The primary data are qualitative. Findings are tentative.
Conclusive research is more formal and structured than ER. It is based on large, representative samples, and data are subjected to quantitative analysis. Objective of these studies is to provide a comprehensive and detailed explanation of phenomena under study.
Findings are conclusive. Used in managerial decisions. These may be either descriptive or causal.
DIFFERENCES BETWEEN EXPLORATORY AND CONCLUSIVE RESEARCH
April 19, 2023
9
Exploratory Conclusive
Objective: To provide insights and understanding.
To test specific hypotheses and examine relationships
Characteristics:
Information needed is defined only looselyResearch process is flexible and unstructuredSample is small and non-representativeAnalysis of primary data is qualitative. Tentative
Information needed is clearly defined.Research process is formal and structuredSample is large and representative.Data analysis is quantitative. Conclusive
Findings/Results:
Outcome: Generally followed by further exploratory or conclusive research
Findings used as input into decision making
Exploratory Research Design
April 19, 2023
10
Secondary resource analysis: Secondary sources of data give information –in terms of details of previously collected findings in facts and figures – which has been authenticated and published.
Case method: it is intricately designed and reveals a comprehensive and complete presentation of facts, as they occur, in a single entity. This could be an individual, an organisation or an entire country.
Exploratory Research Design
April 19, 2023
11
Expert opinion survey: valuable insights obtained from experts which might be based on their experience in the field or based on academic work done on the concept.
Focus group discussions: a carefully selected representative sub set of the larger respondent gather to discuss together, in a short time frame, the subject/topic to be investigated.
Descriptive Studies
Descriptions of population characteristics
Descriptions of population characteristics
Estimates of frequency of characteristics
Estimates of frequency of characteristics
Discovery of associations among variables
Discovery of associations among variables
Descriptive Research Design
April 19, 2023
13
Cross-sectional research designs: two criteria1.carried out at a single moment in time, therefore
the applicability is temporal specific2.Conducted on a sub-section of the respondent
population
VariationsSingle/multiple cross- sectional designs Cohort analysis
Descriptive Research Design
April 19, 2023
14
Longitudinal studies: 1.The study involves selection of a
representative group as a panel.2.There are repeated measurement of the
researched variable on this panel over fixed intervals of time.
3.Once selected the panel composition needs to stay constant over the study period.
Experimental Design
April 19, 2023
16An experiment is generally used to infer a causality. In an experiment, a researcher actively manipulates one or more causal variables and measures their effects on the dependent variable of interest.
Independent variables: Independent variables are also known as explanatory variables or treatments. The levels of these variables are manipulated (changed) by researchers to measure their effect on the dependent variable.
Test units: Test units are those entities on which treatments are applied.
Dependent variables: These variables measures the effect of treatments (independent variable) on the test units.
Experiment: An experiment is executed when the researcher manipulates one or more independent variables and measures their effect on the dependent variables while controlling the effect of the extraneous variables.
Extraneous variables: These are the variables other than the independent variables which influence the response of test units to treatments.
Examples: Store size, government policies, temperature, food intake, geographical location, etc.
Validity
April 19, 2023
17
Validity: The degree of confidence researchers and managers can have in the results of the study.
Two Goals 1. draw valid conclusions about the impact of independent
variable on the study group, Secondly, should be able to generalize the findings to a larger
population of interest.
The first objective is related with what is known as internal validity. The second is related with external validity.
Internal validity: Internal validity tries to examine whether the observed effect on a dependent variable is actually caused by the treatments (independent variables) in question.
Some major classes of variables that may affect internal validity/generate experimental error: i) History (ii) Maturation (iii) Testing Effects (iv) Instrumentation (v) Selection Bias (vi) Test Unit Mortality
Validity
External validity: External validity refers to the generalization of the results of an experiment. The concern is whether the result of an experiment can be generalized beyond the experimental
In most of the experiments the data are collected through sampling process and not from population.
Factors Affecting External Validity: The environment at the time of test may be different from the environment of
the real world where these results are to be generalized.Population used for experimentation of the test may not be similar to the
population where the results of the experiments are to be applied.Results obtained in a 5–6 week test may not hold in an application of 12
months.Treatment at the time of the test may be different from the treatment of the
real world.
It is desirable to have an experimental design that has both external and internal validity.
Methods to Control Extraneous Variables
April 19, 2023
19
Randomization
Matching
Use of experimental designs
Statistical control
Types of Experimental Designs
April 19, 2023
21
Types: Basic Designs & Statistical DesignsBasic Designs: Considers the impact of
only one independent variable at a time1. After-Only Design; 2. Before-After Design3. Before-After With Control Design4. Simulated Before-After Design5. After-Only with Control6. Solmon Four-Group Design
Types of Experimental Design
April 19, 2023
22
Statistical Design: Allows the evaluation of the effect of more than one independent variable
1. Randomized Block Design2. Latin Square design3. Factorial Design
Various Symbols in Experimental designs:MB = pre-measurement of the dependent variable i.e. before the introduction/manipulation of the independent variable.MA = post-measurement of the dependent variable i.e. during the introduction/manipulation of the independent variable.X = treatment; the actual introduction or manipulation of the independent variableR = designation/notation that the group is selected randomly
Basics Experimental Designs
April 19, 2023
23
After-Only design: involves manipulating the independent variable and following this with a post-measurement, or symbolically: ‘X MA’
Ex. Ford Motors company spent $500,000 on such exp. In Dallas & in San Deiego. Sent engraved invitations to women to attend dealer showroom ‘parties’ and served wine & cheese. Latest Clothing Fashions displayed by models & Ford automobiles were shown in ‘no pressure’ situation.
Subsequent purchase by those who attended the party was one measure to measure the success of the experiment.
Advantages/Disadvantages: Typical most new-products test markets example. Results difficult to interpret & subject to numerous errors. Requires substantial market knowledge & subjective judgment. – Should be used with care.
Basics Experimental Designs
April 19, 2023
24
Before After Design: It involves a pre-measurement in addition: ‘MB X MA’
The result of interest: (MA – MB) i.e. considerable advantage over After-Only.
Ex. To estimate the effect of price increase on market share.
Researcher must be alert to the possibility that extraneous variable caused the result than the independent variable. Hence unless the researcher is confident that extraneous variables are not operating or that he/she can control their effect, before after design should be avoided.
Above two tests – Quasi-experimental Design
Before-After with Control Design
April 19, 2023
25
Involves the addition of a control Group to the Before-After Design: R MB1 X MA1
R MB2 MA2Can control all potential errors except mortality & interaction.Measure of interest: (MB1 – MA1) – (MB2 – MA2)Ex. A Firm wishes to test the impact of a P-O-P display. Ten retail
stores selected for inclusion in the treatment group, another ten for the control group. Sales measured in each group of stores, before and after the new P-O-P display. The change in sales of the two group is compared. Controls any initial inequalities between the sales group.
In cases where the interaction is unlikely and control for history and selection errors is important, the before-after with control group design is the best design in terms of cost and error control.
Simulated Before After Design
April 19, 2023
26Controls pre-measurement & interaction errors in
experiments dealing with attitude and knowledge of human subjects. Uses separate groups for the pre & post measurement: R MB
R X MAMeasure of interest= (MA – MB)Different individuals receives the pre- & post
measurements, there can be no pre-measurement or interaction effect.
Ex. Advertising Research. Large sample-Questionnaire- attitude towards the product (pre-measurement) – Advertisement (Change in the independent variable) – second sample of respondents, same questionnaire (post-measurement). Thus difference in the two scores can contribute to the effect of advertising campaign.
After-Only with Control design
April 19, 2023
27
If it is likely that the groups are initial equal on the variable of interest then there is no reason to go to the expense of pre-measurement. Instead an after-only with control design can be used.
R X1 MA1R MA2
Measure of interest: (MA2 – MA1)
Solomon Four-Group Design
April 19, 2023
28
Consists of Four Groups, two treatment + two control and six measurements: two pre-measurements + Four post-measurements:
Experiment Group1: R MB1 X MA1Control Group1: R MB2 MA2Experiment Group2: R X MA3Control Group2: R MA4
It controls all sources of experimental errors except measurement timing & reactive error which is not subject to control by designs. No single method of analysis makes use of all six measurement methods simultaneously
All four groups are pre‑selected in such a way that they are equivalent i.e. they are selected on a random basis. This means before measurement should be the same in all four groups except for random variations.
The four‑group six-study design may be taken as a model for marketing experiments, has little practical value.
The expense of selecting four groups randomly and making six studies among these groups make this design impractical for most marketing studies.
Participants’ Perceptions
No deviation perceived
Deviations perceived as unrelated
Deviations perceived as researcher-induced
Statistical Designs
April 19, 2023
30
Statistical designs allow for statistical control and analysis of external variables.
Completely randomized design
Randomized block design
Latin square design
Factorial design
Statistical designs
April 19, 2023
31
Completely randomized Design:
Treatments are applied to the experimental units entirely by a chance process.
Statistical tool used – ANOVA one way
Statistical Designs
Randomized Block Design(Two Way ANOVA): In RBD the experimental units are blocked, that is, grouped
or stratified, on the basis of extraneous or blocking variable. Ex. Assume that a total sample of 800 males and 400 females
is available. Individuals are assigned to blocks based on their gender, producing one block of 400 females and one block of 800 males. The individuals within each block are randomly assigned to treatment groups. The use of ANOVA then allows the researcher to determine the impact of the commercial on the overall group as well as its impact on the males and females subgroups.
April 19, 2023
32
Factorial Design
Used to measure the effect of two or more independent variables at same time and to measure the interaction effect of the variables. Interaction occurs when the simultaneous effect of two or more variables is different from the sum of their effects taken one at a time.
Ex. Ones favorite color might be gray & favorite desert might be ice-cream. However it does not follow that he/she would prefer gray ice-cream.
April 19, 2023
33
Factorial Design
Ex. Consider the problem of determining the proper concentration of sugar & flavor in a soft drink.
One approach may be to make a batch of optimum mixture and have a sample consumer taste it and indicate and order of preference.
Another approach may be: makeup several batches with different level of sugar content and flavor constant. Consumer may then taste and indicate a preference. Later on sugar could be held constant and flavor varied.
Later approach may indicate that heavy sugar and heavy sugar were preferred. May not be valid always. The fact may be when the flavor is strong, the sugar may be less desirable.
So its important to test various levels of sugar content combined with various levels of flavor
Factorial Design
Suppose four degrees/levels each were selected as possible characteristics of the final product. Sixteen combinations can be made as follows & be given to sample of consumers.. With say some preference from 1-10… Hypothetically:
Second degree of sugar content & third degree of flavor are preferred over others. The combination of these two is the product formula as per FD.
Flavor Intensity
Sugar Content
1 2 3 4
1 A – 4.9 B- 6.0 C – 5.0 D 3.6
2 E – 6.1 F – 7.3 G – 5.1 H – 3.8
3 I – 8.1 J - 9.2 K – 8.3 L – 4.6
4 M – 6.2 N – 6.4 O – 6.2 P – 3.2
Latin Square design
With the Latin Square design one can control variation in two directions.
The design requires that extraneous or blocking variables be divided in to an equal no. of blocks or levels, such as drugstores, supermarkets and discount stores. The independent variables be divided in to the same no. of levels, such as high price, medium price & low price.
-Treatments are arranged in rows and columns -Each row contains every treatment. -Each column contains every treatment. -The most common sizes of LS are 5x5 to 8x8 Advantages of the LS Design 1. You can control variation in two directions. 2. Hopefully you increase efficiency as compared to the RBD.
April 19, 2023
36
Latin Square Design
April 19, 2023
37
Disadvantages of the LS Design: 1. The number of treatments must equal the number of replicates. 2. The experimental error is likely to increase with the size of the square.
References:
1. Business Research Methods – Cooper, Schindler; Tata Mc Graw Hills
2. Marketing Research – G C Beri; Tata Mc Graw Hills.3. Business Research Methods – William G Zikmund; Thomson.4. Marketing Research – Tull, Hawkin; PHI
April 19, 2023
39