Report R3.13 Measurement in Research Seminar Leaders: Jaydeep Mukherjee, Adwaita Govind Menon, D. P. Dash, Gopal K. Nayak Faculty and Fellow Scholars, XIMB This research training seminar was conducted as a group discussion. The participants discussed various issues pertaining to measurement in research. Measurement is an essential activity in research. It is the process of assigning numbers to entities which are being studied. Precision and correctness of measurement are vital in research. Measurement depends a lot on the entities to be measured and the context in which the measurement is being undertaken. The same entity may be measured differently by different researchers. However, sometimes it becomes necessary in research to develop standard measures. Therefore, reliability, unbiasedness, etc., become important qualities in such cases. Still, interpretations and inferences drawn after measurement can vary and they shape the future course of research. The subjectivity involved is a critical consideration to be taken into account during the process of measurement. The issues of measurement were then discussed vis-à-vis a research project being conducted currently by some of the participants. It involves developing a scale for measuring teaching effectiveness in a business school context. One of the difficult aspects was to measure the “attitude” of the teacher towards students' learning. In this context, the use of attitudinal scales such as Thurstone and Likert scales was discussed. But, the way these scales deal with subjectivity of the responses remains to be clarified. It was stressed that the purpose of measurement needs to be properly defined. It was suggested that a well defined research framework be drawn before the data collection phase. Clarity of concept and knowledge on the subject are necessary for addressing issues of measurement in research. Measurements usually involve mapping. Mapping is the process of representing one conceptual domain in terms of another conceptual domain. It refers to a systematic set of correspondences that exist between the constituent elements of the source and the target domains. It requires analogical reasoning. The process of measurement in research consists of conceptualisation and operationalisation. Conceptualisation involves the formation of concepts. It is important to relate concepts to theories. The researcher might come up with a new concept, but such initiatives are rare in research. The process of taking a conceptual definition and making it more usable by linking it to one or more specific, concrete indicators could be termed as operationalisation. This is usually done through numbers that reflect empirical reality. After operationalisation, the process of measurement involves determining the level of measurement (nominal, ordinal, interval, ratio), and then assessment of reliability and validity. The role of estimates in research was discussed. To estimate is to judge and form an opinion regarding the value (intrinsic or extrinsic) of an object from imperfect data. This was examined in the context of financial management. Financial management is a field where estimates are used in the absence of actual figures. It was suggested that researchers can make a choice regarding the use of estimates. The use of "shadow estimates" and "surrogates" in financial research was cited. A specific example would be the shadow price estimate. In finance and economics, it refers to the use of linear programming techniques in a situation where price cannot be charged or where the price does not reflect the effort made in producing the good. It is an attempt to achieve optimum allocation of resources in the absence of an effective pricing system. In financial markets, it refers to the arbitrary assignment of money values to non-marketed products. Surrogate refers to a substitute. The use of surrogates and shadow estimates aid the process of measurement in research. The issue of subjectivity in the use of estimates was discussed. Estimates not being the exact figures might lead to errors. There could be a chance of differing estimates for the same entity. It was concluded that the issue needs careful attention of the researcher. The role of modelling and mapping in research were also discussed. Modelling refers to the process of generating a model as a conceptual representation of some phenomenon. Typically a model refers only to some aspects of the phenomenon in question. It is highly likely that two models of the same phenomenon may be different. It is essential to know the actual purpose of the modeller and the assumptions that have been made. Models in general are domain specific. A model is used when it is helpful in understanding a particular phenomenon in a specific context. It helps the researcher to follow the logic of the phenomenon under study. However, there could be context-independent models too, which can be transferred across domains. The discussion then shifted to the issue of reflexivity. Reflexivity is a difficult topic in the measurement literature. Reflexivity could be termed as the researcher’s awareness of the ways in which the researcher's involvement with a particular study influences, acts upon, and informs the research. It could refer to the ways in which the researcher’s values, experiences, interests, beliefs, etc., have shaped the research. It might also refer to the assumptions (about the world and knowledge) that the researcher has made in the course of the research. It helps the researcher to think about the implications of such assumptions for the research and its findings. Reported by Adwaita Govind Menon, with inputs from D. P. Dash and Jacob D. Vakkayil (16 March 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 |
