Business research design

Business research design for MBA

1. Research Design

2. Meaning of Research Design Research design refers to a plan, structure and strategy of investigation so as to obtain answer to research questions or problems. In other words it is the detailed blueprint used to guide According to Sitelltiz et.al.(1962), “A research is the arrangement of conditions for collection, measurement and analysis of data in a manner that aims to combine relevance to the research purpose with economy in procedure”. In fact, research design is a blueprint or detailed plan for how a research study is to be completed. In other words it is a conceptual structure within which the research is conducted. It seeks the answers to the questions like- what approach to the problem should be taken? What methods will be used to tackle the problem? What strategies will be most effective?

3. Types of research design There are various types of research design have been suggested by different scholars. For instance, Selltiz et.al. ( 1962) suggested three broad categories of research design viz. ( i ) formulative or exploratory studies (ii) descriptive studies and (iii) testing causal hypothesis studies. Here, we are following the research designs as given below:

4. Descriptive Research: It is essentially a fact finding study, generally conducted to assess the behavior or characteristic of a given population or to describe the overall situation or events occurring at present . In other words, it simply portrays the facts and does not necessarily seek to explain relationship, test of hypothesis, make predictions or implication of the study . The investigators collect the data, classify, tabulate, present and analyze the data to describe about what exists, but they do not fully analyze and explain why phenomena behave as they do. For example, the studies like – descriptive study about the workers in a factory, community welfare activity and beneficiaries, state of physical health in rural area etc.

5. Government or sometimes the development agencies conduct the baseline survey of the particular district or region of the country to know about the situation or condition of the place in order to intervene some development program is also a descriptive research. It is an accumulation of a database, that is solely descriptive. Researchers, however, are not in agreement on what constitutes “descriptive research” and often broaden the term to include all terms of research except historical and experimental. Descriptive research contributes to any discipline primarily by building a foundation of facts.

6. The main purpose of the descriptive research may be categorized as follows: To collect detailed factual information that describes the existing phenomena. To identify problems, current conditions and practices. To make comparisons and evaluations. To determine what others are doing with similar problems on situation and benefit from their experience in making future plans and decisions .

7. Developmental Research: Development research is conducted for the purpose of assessing the development work or program launched by the government or any other agencies. It focuses to study the change in development work or program and the related indicators, which affects the change and their rate, direction and inter-related factors over a period of time. On the basis of rate of change, we can also predict for future trend of the program. Developmental research may be conducted in the following two ways: ( i ) Longitudinal Study (ii) Cross-sectional Study

8. Longitudinal Study: Longitudinal study is conducted to measure the change in nature and rate of change of certain phenomena at different stages of development. In this study, data are collected at two or more development phases from the same group of individual in order to evaluate the change of phenomena. Cross-sectional Growth Study: Cross-sectional study is designed to cross check the result of development phases. It measures the rate of changes by drawing samples from a cross-section of the particular place or society, where there has been intervened the development activities.  

9. Case study Research: The “case study” is very popular in research. It usually refers to a fairly intensive examination of a single unit. A unit may be a person or a family or a small social group of people or an institution or a community or a single company. In other words, it is essentially an intensive investigation of the particular unit under consideration. The objective of this method is to locate the factors that account for the behavior patterns of the given unit as an integrated totality. According to H. Odum , “ The case study method is a technique by which individual factor whether it be an institution or just be an episode in the life of an individual or a group is analyzed in its relationship to any other in the group”.

10. Case study method is a form of qualitative analysis where in careful and complete observation of an individual or a situation or an institution is done. But we must understand that a case study does not represent the total reality. It is just an example of the social reality or any other realities. Hence, a case study may be an intensive, integrated and insightful method of studying the social phenomenon. It can be used to illustrate a theory by providing an example. It enables us to explore, and understand the problems and issues in relation to a particular situation.

11. Causal-comparative Research: This research investigates the possible causes affecting a particular situation by observing the existing evidences and searching for possible consequences (factors) leading to the results. The main objective of this research is to study the situation or problem in order to explain the relationship between two or more variables. It is also known as ‘ex post facto’ research. In this method the investigator takes one or more independent variables and examines the results, seeking causes and consequences. It is basically conducted to find the clues about what might cause or contribute to the particular phenomenon. This research predicts the dependent variable on the basis of independent variables.

12. True Experimental Research There are so many types of experimental design. This section, therefore is confined to describing those most commonly used in social sciences: Post-test only Control Group Design (After only with Control): In some situations, when pre-test measurements were not possible, but are to be investigated the possible influence by exposing the dependent variable to experimental intervention (or treatment) and comparing the results with the control group, the post-test-only control group or after only with control group design is used. This design is diagrammed as follows:

13. The effect of experimental treatment is measured by the difference between O 1 – O 2 . Since the experimental group has been exposed to the treatment, we might expect that the observed difference (O 1 - O 2 ) might be due to the influence of experimental treatment. Experimental group Control group RA X O 1 O 2 Time

14. Pre-test – Post-test Control Group Design (Before and After with Control): A pre-test – post-test control group design or before – after with control group design is the classic experimental design. Thus there are two groups namely an experimental group and a control group, which does not receive any treatment. Between time T 1 (before) and T 2 (after), the experimental group is exposed to an experimental intervention (treatment) and the control group is left alone. At both time T 1 and T 2 , the experimental and control groups are measured in relation to the key dependent variable (or subject) that is of interest in the study.

15. Before T 1 After T 2 Experimental Group O 1 O 2 Control Group O 3 O 4 Intervention Diff.= T 1 – T 2 E diff . = O 2 – O 1 C diff . = O 4 – O 3 No Intervention RA

16. The measure of the subjects in the experimental group are tested before and after being exposed to the treatment by O 1 and O 2 at times T 1 and T 2 respectively. The measures for the same in the control group on the same subjects by O 3 and O 4 . Since the experimental group has been exposed to the treatment, we might expect that the difference in time between T 1 and T 2 might be due to the influence of the treatment, which is E diff . However, it is to be noted that this observed difference between T 1 and T 2 might also be due to factors other than experimental intervention. A change could occur due to passing of time or might be due to other set of possible factors. Hence to measure the real impact due to treatment, a control or comparison group is needed. Ideally, this group should be identical to the experimental group at time T 1 .

17. However, the control group has not been exposed to any experimental intervention. We can measure the change on the subjects over time T 1 and T 2 , which is denoted by C diff . The crucial thing to look at is whether the experimental group changed more than the control group. If it is changed significantly more, we normally would conclude that this is because of the experimental intervention. Of course, this conclusion is warranted only if both groups were same to start with. In order to ensure that they are the same to start with, people will be assigned randomly to the experimental and control groups.

18. Quasi-Experimental Design: In many field of research situations, it is certainly very costly or simply impossible and difficult to meet the random assignment criteria of a true experimental design. Despite of some difficulties in conducting true experimental studies due to its nature, researchers on the other hand want to avoid the problems of validity associated with non-experimental designs. In such case, it is reasonable to select quasi-experimental design. These designs do not have the restrictions of random assignment. At the same time, they do not adequately control for problems of threats to internal validity. Followings are the types of quasi-experimental designs:

19. The after only design (one-shot case study): In an after only design, a population is being exposed to an intervention and if the researcher wishes to study its impact on the population, then this design (or one-shot case study) is used. In this design, information on baseline (pre-test or before observation) is usually constructed on the basis of respondent’s recall of the situation before the intervention, or from information available in existing records. The change in dependent variable is measured by the difference between the ‘before’ (baseline) and ‘after’ observation.

20. X O Treatment or intervention Observation or measurement of dependent variable Technically this is not fair design for measuring the impact of an intervention as there is no proper baseline data to compare the ‘after’ observation with i.e. two sets of data are not strictly comparable.

21. The before-and-after design (One group Pre-test- Post test Design): The before-and-after design overcomes the problem of retrospectively constructing the ‘before’ observation by establishing it before the intervention is introduced to the study population. When the program has been completely implemented, it is assumed to have had its effect on the population, the ‘after’ observation is carried out to ascertain the impact attributable to the intervention. The difference between O 2 and O 1 (O 2 – O 1 ) is the measure of the influence of the experimental treatment. This design offers a comparison of the same group before and after intervention.

22. O 1 Before (Pre-test) X Treatment or intervention O 2 After (Post-test) The Time Series Design: This design is similar to the non-experimental pre-test post- test design except that there are repeated measurement observations before and after the program intervention (X). The advantage of this design is the multiple measurement observations.

23. For example, if we find that there is no difference between O 1 and O 2 and O 3 , but then a sudden increase occurs between O 3 and O 4 which is subsequently maintained in O 5 and O 6 , we can conclude with some degree of confidence that the sudden increase was probably due to the effect of the program intervention (X). Experimental Group Time O 1 O 2 O 3 X O 4 O 5 O 6