Research Design

TYBMS

Prof. Hemant Kombrabail

RESEARCH DESIGNS The next step after stating the management problem, research purpose, and research hypotheses and questions, is to formulate a research design. The starting point for the research design is, in fact, the research questions and hypotheses that have been so carefully developed. In essence, the research design answers the question: How are we going to get answers to these research questions and test these hypotheses? The research design is a plan of action indicating the specific steps that are necessary to provide answers to those questions, test the hypotheses, and thereby achieve the research purpose that helps choose among the decision alternatives to solve the management problem or capitalize on the market opportunity DEFINITIONS OF RESEARCH DESIGN: (1) According to David J. Luck and Ronald S. Rubin, "A research design is the determination and statement of the general research approach or strategy adopted/or the particular project. It is the heart of planning. If the design adheres to the research objective, it will ensure that the client's needs will be served." (2) According to Kerlinger "Research design in the plan, structure and strategy of investigation conceived so as to obtain answers to research questions and to control variance." (3) According to Green and Tull "A research design is the specification of methods and procedures for acquiring the information needed. It is the over-all operational pattern or framework of the project that stipulates what information is to be collected from which source by what procedures." The second definition includes three important terms - plan, structure and strategy. The plan is the outline of the research scheme on which the researcher is to work. The structure of the research work is a more specific scheme and the strategy suggests how the research will be carried out i.e. methods to be used for the collection and analysis of data. In brief, research design is the blueprint of research. It is the specification of methods and procedures for acquiring the information needed for solving the problem. Questionnaires, forms and samples for investigation are decided while framing research design. Finally, the research design enables the researcher to arrive at certain meaningful conclusions at the end of proposed study. STEPS IN PLANNING THE RESEARCH DESIGN: There are four broad steps involved in planning the research design as explained below: (1) Determining work involved in the project: The first step in planning research design is determining the work involved in the projectand designing a workable plan to carry out the research work within specific time limit. The work involved includes the following: (a) To formulate the marketing problem (b) To determine information requirement (c) To identify information sources (d) To prepare detailed plan for the execution of research project.

1

TYBMS

Prof. Hemant Kombrabail

This preliminary step indicates the nature and volume of work involved in the research work. Various forms require for research work will be decided and finalised. The sample to be selected for the survey work will also be decided. Staff requirement will also be estimated. Details will be worked out about their training and supervision on field investigators, etc. In addition, the questionnaire will be prepared and tested. This is how the researcher will prepare a blue-print of the research project. According to this blueprint the whole research project will be implemented. The researcher gets clear idea of the work involved in the project through such initial planning of the project. Such planning avoids confusion, misdirection and wastage of time, money and efforts at later stages of research work. The whole research project moves smoothly due to initial planning of the research project. (2) Estimating costs involved: The second step in planning research design is estimating the costs involved in the research project. MR projects are costly as the questionnaire is to be prepared in large number of copies, interviewers are to be appointed for data collection and staff will be required for tabulation and analysis of data collected. Finally, experts will be required for drawing conclusions and for writing the research report. The researcher has to estimate the expenditure required for the execution of the project. The sponsoring organisation will approve the research project and make suitable budget provision accordingly. The cost calculation is a complicated job as expenditure on different heads will have to be estimated accurately. The cost of the project also needs to be viewed from the viewpoint of its utility in solving the marketing problem. A comprehensive research study for solving comparatively minor marketing problem will be uneconomical. (3) Preparing time schedule: Time factor is important in the execution of the research project. Planning of time schedule is essential at the initial stage. Time calculation relates to the preparation of questionnaire and its pre-testing, training of interviewers, actual survey work, tabulation and analysis of data and finally reports writing. Time requirement of each stage needs to be worked out systematically. Such study will indicate the time requirement of the whole project. Too long period for the completion of research work is undesirable as the conclusions and recommendations may become outdated when actually available. Similarly, time-consuming research projects are not useful for solving urgent marketing problems faced by a company. Preparing time schedule is not adequate in research design. In addition, all operations involved in the research work should be carried out strictly as per time schedule already prepared. If necessary remedial measures should be adopted in order to avoid any deviation in the time schedule. This brings certainty as regards the completion of the whole research project in time.

2

TYBMS

Prof. Hemant Kombrabail

(4) Verifying results: MR findings need to be dependable to the sponsoring organisation. Researcher may create new problems before the sponsoring organisation if the research work is conducted in a faulty manner. Such unreliable study is dangerous as it may create new problems. It is therefore, necessary to keep effective check on the whole research work during the implementing stage. For this suitable provisions need to be made in the research design. After deciding the details of the steps noted above, the background for research design will be ready. Thereafter, the researcher has to prepare the research design of the whole project. He has to present the project design to the sponsoring agency or higher authorities for detailed consideration and approval. The researcher can start the research project (as per design) after securing the necessary approval to the research design prepared.

TYPES OF RESEARCH DESIGNS A research design is like a roadmap—you can see where you currently are, where you want to be at the completion of your journey, and can determine the best (most efficient and effective) route to take to get to your destination. We may have to take unforeseen detours along the way, but by keeping our ultimate objective constantly in mind and using our map we can arrive at our destination. Our research purpose and objectives suggest which route (design) might be best to get us where we want to go. but there is more than one way to "get there from here." Choice of research design is not like solving a problem in algebra where there is only one correct answer and an infinite number of wrong ones. Choice of research design is more like selecting a cheesecake recipe—some are better than others but there is no one which is universally accepted as "best." Successfully completing a research project consists of making those choices that will fulfill the research purpose and obtain answers to the research questions in an efficient and effective manner. Choice of design type is not determined by the nature of the strategic decision faced by the manager such that we would use research design A whenever we need to evaluate the extent of a new product opportunity, or design B when deciding on which of two advertising programs to run. Rather, choice of research design is influenced by a number of variables such as the decision maker's attitude toward risk, the types of decisions being faced, the size of the research budget, the decision-making time frame, the nature of the research objectives, and other subtle and not-so-subtle factors. Much of the choice, however, will depend upon the fundamental objective implied by the research question: • To conduct a general exploration of the issue, gain some broad insights into the phenomenon, and achieve a better "feel" for the subject under investigation (e.g.. What do customers mean by "good value"?). • To describe a population, event, or phenomenon in a precise manner where we can attach numbers to represent the extent to which something occurs or determine the degree two or more variables covary (e.g., determine the relationship between age and consumption rate).

3

TYBMS

Prof. Hemant Kombrabail

• To attribute cause and effect relationships among two or more variables so that we can better understand and predict the outcome of one variable (e.g., sales) when varying another (e.g., advertising).

RESEARCH DESIGN

CONCLUSIVE RESEARCH DESIGN

CAUSAL RESEARCH

EXPLORATORY RESEARCH DESIGN

DESCRIPTIVE RESEARCH

CROSS-SECTIONAL DESIGN

LONGITUDINAL DESIGN

SINGLE CROSS-SECTIONAL DESIGN

MULTIPLE CROSS-SECTIONAL DESIGN

This classification is frequently used and is quite popular. Before we discuss each of these design types, a cautionary note is in order. Some might think that the research design decision suggests a choice among the design types. Although there are research situations in which all the research questions might be answered by doing only one of these types (e.g., a causal research experiment to determine which of three prices results in the greatest profits), it is more often the case that the research design might involve more than one of these types performed in some sequence. The overall research design is intended to indicate exactly how the different design types will be utilized to get answers to the research questions or test the hypothesis.

4

TYBMS

Prof. Hemant Kombrabail

A further cautionary note is needed to warn the reader that while it may appear that if sequencing is done the sequence would be exploratory, descriptive, then causal, that is not always the case. For example, some companies may do an annual survey of consumers to determine the frequency with which certain behaviors are performed (e.g., washing dishes by hand) followed by exploratory research that probes to gain an in-depth understanding of the circumstances surrounding that behavior (i.e., descriptive then exploratory rather than exploratory then descriptive). It is not hard to imagine a research design that might sequence as exploratory, then descriptive, then exploratory again; or causal, then descriptive. It is important to remember that because a research design is a plan of action to obtain answers to the research questions, it is those questions that suggest which design types are necessary and the sequence of conducting those design types, if a sequence is needed. An example later in this chapter will be used to illustrate this point. With these cautions in mind we will now discuss the design types in greater detail. 1. Exploratory Research: A marketing research study may be either exploratory or full scale. Exploratory research is conducted when the researcher does not know how and why certain phenomenon occurs. Here, the hypothetical solutions or actions are explored and evaluated by the decision-maker, e.g. evaluation of quality of service of a bank/hotel/airline. Here, the quality cannot be assessed directly as tangible features are not available. The purpose of exploratory research is to know the unknown. Exploratory research determines fruitful alternatives that the executive would not have perceived. This also narrows down the scope of the investigation. Exploratory research is undertaken to get the answer to the following question. "What alternative courses of action might solve the problem and thereby reach the final objective?" This research is unstructured in character. Exploratory studies are important. They may provide adequate information on a decision situation or may greatly facilitate the design of formal research studies. To explore means to find out or discover. Exploratory research is conducted in order to find out causes/reasons behind a specific marketing problem. It is the starting point in all types of research projects. The purpose of exploratory research is to define the marketing problem precisely, collect required information/data relating to the problem and identify alternative courses of action in order to deal with the marketing problem. For example, advertising campaign of a company may not give positive results in terms of sales. The failure of campaign may be due to many possible causes relating to advertising copy, media selected, faulty pre-testing, faulty illustration give, wrong timing of ad. campaign or finally inadequate follow-up measures to the ad. campaign. In the exploratory research, the possible causes will be identified, the most appropriate causes will be selected, hypothesis will be developed and research activity will be conducted accordingly. Survey of consumers, retailers, sales executives and sales-force will be useful for exploratory research. On some occasions, small scale sales survey may provide useful data for exploratory research.

5

TYBMS

Prof. Hemant Kombrabail

In exploratory research, the stress/focus is on the discovery of ideas/causes. For example, sales may be declining for the last six months. Quick study may be conducted to find out the causes/factors responsible. Such causes will be listed. Here, an exploratory study/research may be conducted in order to find out the most likely cause so as to introduce suitable remedial measures. The objective of exploratory research is to generate/discover new ideas. The secondary/published data can be used for exploratory research as such data are easily available. If the services of respondents are used (for survey purpose), they should be given full freedom to express themselves. The same rule should be made applicable to focus groups. Exploratory research is useful for the study of marketing problems about which sufficient information/details are not available. Exploratory study needs to be flexible in its approach. Exploratory Research is used:  To define the problem more precisely  To identify relevant courses of action i.e. find the most likely alternatives, which are then turned into hypotheses.  Isolate key variables and relationships for further examinations.  Gain insights for developing an approach to a problem.  Establish priorities for further research. Once a hypothesis is formulated, research is used to determine if the hypothesis was correct. Hypothesis is tentative answers to questions that serve as guides for most research projects Exploratory research may also be involved when the perceived problem is much less general; it is used to develop the most promising hypotheses. The findings of exploratory research should be regarded as tentative or as an input to further research. Typically, such research is followed by further exploratory or conclusive research. The analysis of primary data is qualitative. In general, exploratory research is meaningful in any situation in which the researcher does not have enough understanding to proceed with the research project. 2. Conclusive Research Design Conclusive Research Design 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. Conclusive Research is designed to assist (he decision maker in determining, evaluating and selecting the best course of action to take in a given situation. As shown in the figure conclusive research designs may be either descriptive or causal and descriptive designs may be either cross-sectional or longitudinal. (A) Descriptive Research: Descriptive research is undertaken when the researcher desires to know the characteristics of certain groups such as age, sex, occupation, income or education. The

6

TYBMS

Prof. Hemant Kombrabail

objective of descriptive research is to answer the "who, what, when, where and how" of the subject under study/investigation. Descriptive studies are normally factual and simple. However, such studies can be complex, demanding scientific skill on the part of researcher. Descriptive studies are well structured. It tends to be rigid and its approach cannot be changed often and again. In descriptive studies, the researcher has to give adequate thought to framing research questions and deciding the data to be collected and the procedure to be used for this purpose. Data collected may prove to be inadequate if the researcher is not careful in the initial stages of data collection. Descriptive research designs are used for some definite purpose. Descriptive research cannot identify cause and effect relationship. Descriptive research is designed to describe the present situation or the features of a group or users of a product. In marketing, such research is undertaken to know the characteristics of certain groups or users of a product such as age, sex education, income etc. Such research studies are based on secondary data or survey research. The major objective of descriptive research is to describe something - usually market characteristics or functions A major difference between exploratory and descriptive research is that descriptive research is characterized by the prior formulation of the hypotheses. Thus, the information needed is clearly defined. As a result, descriptive research is preplanned and structured. It is typically based on large representative samples A formal research design specifies the methods for selecting these sources of information and for collecting data from those sources. Uses of Descriptive Research: Descriptive research is conducted for the following reasons 1. To describe the characteristics of relevant groups, such as consumers, salespeople, or organizations, or market areas. For e.g. we could develop a profile of the "heavy users" (frequent shoppers) of prestigious department stores such as Shoppers Stop. 2. To estimate the percentage of units in a specified population exhibiting a certain behavior e.g. the percentage of heavy users of prestigious department stores who also patronize discount department stores. 3. To determine the " perceptions of product characteristics. For e.g. how do households perceive the various department stores in terms of salient factors of the choice criteria? 4. To determine the degree to which marketing variables are associated. For e.g.: to what extent is shopping at department stores related to eating out? 5. To make specific predictions. For e.g. what will be retail sales of Shoppers stop (specific store) for fashion clothing (specific product category) in the Mumbai area (specific region)?

7

TYBMS

Prof. Hemant Kombrabail

6. To collect demographic information of consumers/users of a product under study. 7. For finding out views and attitudes of customers, e.g. how many customers prefer branded goods or ISI marked goods. 8. Make predictions about future marketing trends, consumer needs or expectations or possible sales after ‘n’ years. 9. To discover the relationship between certain variables, e.g. sale of toothpaste among rural population and urban population or rate of savings among low, middle and higher income groups. Descriptive research can be divided into the following two categories: (a) Cross-sectional studies Cross-sectional study is a study involving a sample of elements from the population of interest at a single point of time. It is a study concerned with a sample of elements from a given population. Such sample may deal with households, dealers, retail stores and other entities. Information/data on a number of characteristics are collected from the sample elements. Such data are analyzed for drawing conclusions. Cross sectional studies include field studies and surveys. Field studies are conducted is the life situations such as schools, factories, institutions, etc. Here, the inter-relations among variables are studied under real setting. The crosssectional analysis involves counting the simultaneous occurrence of the variables of interest. Field studies have certain merits and limitations. The important merit is that such studies are close to real life and cannot be criticised on the ground that they are away from real settings or are artificial. Field studies are also socially significant. However such studies are scientifically inferior to laboratory and field experiments. There is also lack of precision in the measurement of variables. Cross-sectional study is possible through survey. Survey research is wide in scope. Detailed information can be collected from a sample of large population. This method is also economical as more information can be collected per unit of cost. The time required for sample survey is also less than a census-inquiry. However, in survey research, more importance is given to information collection and not to in-depth analysis. Secondly, survey research needs more time and money when conducted on a large scale. (b) Longitudinal studies Longitudinal studies are based on panel methods and panel data. A panel is a sample of respondents who are interviewed not only once but thereafter from time to time. Here data to be collected relate to same variables but the measurements are taken repeatedly. For example, purchase of grocery products by families/ households at regular intervals. Such data will reflect/indicate change in the buying behaviour of families/households. There are many advantages and limitations of panel data. Panel data are suitable when the researcher undertake detailed analysis. Similarly, panel data are more comprehensive as compared to data collected from individual families. Finally, panel data collected is more accurate as compared to data collected through survey. These advantages of panel data improve the quality of research findings and conclusions.

8

TYBMS

Prof. Hemant Kombrabail

There are certain limitations of panel data. For example, panels used for data collection may not be representative samples. Panel members may not be co-operative or may leave the panel membership. As a result, the representative character of the original sample may be adversely affected. Secondly, panel members may report wrong data. Their interest may reduce gradually and they may supply information in a casual manner. Their sense of participation/responsibility may reduce. This will affect the quality of data and also of findings. (B) Casual Research: Casual research design is the third type of research design. As the name indicates, casual design investigates the cause and effect relationship between two or more variables. This design measures the extent of relationship between the variables. Casual research designs attempt to specify the nature of functional relationship between two or more variables. Casual research is useful to show the impact of one variable on the other. For example, price and market demand relationship or relationship between market competition and sales performance. Even the positive/negative effect of advertising on sales can be studied through casual research. The relationship between the casual factors can be studied through casual research. In addition, the variables which create effect on other variables can be studied in depth through casual research. Data for casual research can be collected through field survey with the help of a questionnaire or by conducting laboratory experiments / controlled experiments. Laboratory experiments are possible in the case of testing of new products or package design. The casual research design is based on reasoning. The designs for casual research can be divided into three categories: (a) Historical (b) Survey (c) Experimental. IMPORTANCE / UTILITY OF RESEARCH DESIGN Research design is important as it prepares proper framework within which the research work/activity will be actually carried out Research design acts as a blue print for the conduct of the whole research project. It introduces efficiency in investigation and generates confidence in the final outcome of the study. Research design gives proper direction and time-table to research activity. It keeps adequate check on the research work and ensures its completion within certain time limit. It keeps the whole research project on the right track. Research design avoids possible errors as regards research problem, information requirement and so on. It gives practical orientation to the whole research work and makes it relevant to the marketing problems faced by the sponsoring organisation. Finally, it makes the whole research process compact and result-oriented. A researcher

9

TYBMS

Prof. Hemant Kombrabail

should not go ahead with his research project unless the research design is planned properly.

10