Measurement of Intangibles: Problems of Reliability and Validity

Abstract

Theorists in the social sciences often use concepts that are formulated at rather high levels of abstraction. These are different from the variables that are the stock -in -trade of empirical researchers in these fields. The problem of bridging the gap between theory and research is then seen as one of measurement errors. Measurement is a process involving both theoretical as well as empirical considerations. From an empirical standpoint, the focus is on the observable response - whether it takes the form of a mark on a self-administered questionnaire, the behaviour recorded in an observational study, or the answer given to an interviewer. Theoretically, interest lies in the underlying observable (and directly unmeasurable) concept that is represented by the response. Measurement focuses on the crucial relationship between the empirically grounded indicator(s) -that is, the observable response - and the underlying unobservable concept(s). Intangibles are often measured indirectly (referred to as quasi-quantitative measurement). Such a measurement has to encounter the problem of measurement error as well as race the twin problems of ensuring the reliability and validity of such measures. If the complexity of reality, in particular the qualitative nature of information that we possess about that reality, is to be captured in our theories that we develop, then it is necessary to construct methods of analysis which take into

account the many variables whose inter-relations underline reality as well as any qualitative properties of those variables. The paper discusses the issues of reliability and validity of such measures. The paper gives specific examples of measurement intangibles in two different contexts- measurement by surrogate indicators and measurement by quasi-quantitative methods. The first example considers the case of tacit knowledge as measured by a surrogate indicator, that is, the way in which the scientific and technical manpower in the laboratories of the Council of Scientific and Industrial Research (CSIR), India are deployed across different functional areas like research and development, industrial liaison, planning and monitoring, infrastructure services, workshops including glass blowing and animal house, engineering services and in the pilot plants for long periods. The second example considers the case of measurement of effectiveness of research units in CSIR laboratories where quasi-quantitative measurement technique has been employed. A measurement model has been developed for measuring research unit effectiveness. The fit of the model, its convergent validity and its discriminant validity have been assessed.