Report R3.9 Survey Research: Application of Quantitative Methods Seminar Leader: Damodar Suar, Humanities & Social Sciences, IIT Kharagpur ds[at]hss.iitkgp.ernet.in The seminar sought to discuss the process of survey research and the application of quantitative methods therein. In course of the seminar quantitative and qualitative methods for research were contrasted with each other. The broad steps involved in survey research were identified and the importance of quantitative methods was discussed. Issues such as choices before a researcher, need for conceptualisation of models, and the need for quantification of data were also raised. The different types of variable and their importance in research were also discussed. It became clear that every step in the process of research requires the researcher to make choices from a gamut of options available. The seminar started with the age-old feud between researchers using quantitative analysis and those using qualitative analysis, condemning each other’s practices as irrelevant and meaningless. Quantitative or qualitative analysis would not exist on their own and need to support each other. Good research is in general a mixture of quantitative and qualitative analyses. Such research is called triangulative research. It may be difficult to get research papers published without including some quantitative analysis. Moreover, it was argued, with the passage of time the qualitative aspects of theories tend to fade away and what lingers on is the quantitative component. The example of Maslow’s theory was cited as an exception, whose quantitative analysis has faded away with the passage of time. An attempt was made to identify the possible reasons for quantification and its popularity: 1. Statistical methods are perceived as "highly scientific" 2. Generalisation principles in statistical methods are easily understood 3. Computer based tools are available for handling large amounts of quantitative data 4. Correct use of statistical techniques provides a high degree of precision in results The seminar leader emphasised the importance of quantitative techniques in survey research. He outlined the different steps of a typical survey research project: 1. Selection of topic 2. Formulation of research question 3. Identification of constructs and variables 4. Preparation of questionnaire and interview schedule 5. Sampling 6. Pilot survey 7. Data collection 8. Preparation of master sheet 9. Statistical analysis 10. Interpretation 11. Report writing The seminar leader pointed out that there was a possibility of quantitative or numerical techniques being used at many steps in the research process. In this context meta-analysis was presented as a method for quantitative review of literature. When this is possible, it could guide the researcher in selecting research topics. The discussion moved towards quantification issues regarding what is generally considered as soft data, pertaining to subjective concepts such as love, hate, etc. In this connection, the difference between concept, construct, and variables was discussed. Concepts are rendered measurable by constructs and constructs are decomposed into operational variables. Examples: Concept: Intelligence One construct may be: IQ (accordingly, the variables will be: numerical ability, verbal ability, etc.) Another construct may be: EQ (accordingly, the variables will be: self-awareness, self-regulation, etc.) Quantification of variables such as self-awareness, self-regulation, etc., using scales for measurement of behavioural intentions was discussed. We also talked about practical guidelines that could aid easy quantification such as use of close-ended questions in the questionnaire. Different types of variable were discussed too. The dependent, independent, intervening, and intravening variables were introduced to the participants. The moderating and mediator variables were discussed as a sub-classification of intervening and intravening variables. A moderator variable is one which affects the direction and/or strength of the relation between an independent or predictor variable and a dependent or criterion variable. Specifically within a correlation analysis framework, a moderator is a third variable that affects the zero-order correlation between two other variables. A given variable may be said to function as a mediator when it accounts for the relation between the predictor and the criterion. It could also be stated that a moderator variable is one that influences the strength of a relationship between two other variables, and a mediator variable is one that explains the relationship between the two other variables. The need for the formation of a conceptual model was stressed. Ideally, the model could be a bird’s eye view of the literature that would provide a foundation to develop the research work. The different steps in a survey research process require the researcher to make certain choices that determine the options available subsequently. It is a process of consciously narrowing down to what exactly the researcher seeks. Choice of the unit of study and interpreting the behaviour of variables decide the outcome of the survey research. Choosing a sample size is critical for survey research. Larger the sample size, more it is representative of the population (not straight away; e.g., heterogeneous population). A pilot survey is vital to check the working of the questionnaire. Data collection by trained personnel and its interpretations through statistical methods are important in survey research. The seminar leader cautioned the participants regarding the idiosyncrasies of researchers. The personal bias that the researcher brings in without being aware of it could have a bearing on the process of research. The researcher needs to be aware of the same. The researcher needs to narrow down from broad tentative points of investigation to specific researchable questions. For a survey research project to be fruitful, there has to be some clarity about the audience to be addressed and the exact purpose of the research. Reported by Adwaita Govind Menon, with inputs from D. P. Dash and Jacob D. Vakkayil (20 January 2006). Copyleft Xavier Institute of Management, Xavier Square, Bhubaneswar 751013, India Research World (ISSN 0974-2379) http://www1.ximb.ac.in/RW.nsf/pages/Home |
