HOME | CURRENT | ARCHIVES | FORUM Research World, Volume 3, 2006 Online Version

The point is not that adequate measurement is ‘nice’. It is necessary, crucial, etc. Without it we have nothing. (Korman, 1974, p. 194)

Validation is an unending process ... Most psychological measures need to be constantly evaluated and reevaluated to see if they are behaving as they should. (Nunnally & Bernstein, 1994, p. 84)

Concepts are the basic units of theory development (Zikmund, 2003, Chapter 3: "Theory Building"). Once the researcher has defined the concepts involved in theory development, the next challenge is to measure these concepts by developing appropriate scales.

Developing a sound scale is a difficult and time consuming process. While developing a new measure one could rely on the practices adopted by researchers earlier. A review of scale development practices conducted by Hinkin (1995) highlights the significance of the various methods used for scale development.

Schwab (1980) suggests that the development of a measure could be done in three stages:

i) Item development, i.e., generation of individual items--focus on capturing the domain of interest to ensure "content validity."
ii) Scale development, i.e., combination of individual items to form the scale--focus on the assessment of the psychometric properties of the scale.
iii) Scale evaluation, i.e., psychometric examination of the new measure--focus on "construct validity."

The seminar leader dwelt on the various aspects of measurement and development of instruments for survey research. He focused on the construction and use of an instrument for measuring the marketing environment and promotion mix of rural enterprises--his area of doctoral research.

Promotion includes any activity that a firm uses to communicate with customers, using a "promotional mix," which involves a combination of five elements: mass media advertising, direct promotion, personal selling, sales promotion, and public relations (Armstrong & Kotler, 2004, p. 399). After outlining his area of research, the seminar leader presented his research questions:

i) What is the promotion mix adopted by rural enterprises; is there any similarity and/or difference between that adopted by different types of industry (categorised as per Government of India)?
ii) What is the role of environment (both internal and external) in their promotion mix?

The internal and external factors which influence the promotion mix were discussed. The internal factors were size of the organisation, product category, quality differentiation of products (QDP), and rate of innovation by the firm. The external factors were competition in the industry, target market, and usage pattern by consumers.

Hypotheses:
H1: The proportion of push and pull strategy in the promotion mix will vary with the type of industry.
H2:  Promotion mix will vary with internal and external environment of the industry.

The "sampling universe" selected was the rural enterprises in western India and the "sampling frame" was the Khadi and Village Industries Commission (KVIC) directory. As per the definition used in the present study, rural enterprise means any enterprise located in rural area or township (as specified by the Government of India), which produces goods or renders  services with or without the use of power. The fixed capital investment per artisan or worker should not exceed INR 50,000 (in plant, machinery, land, and building). A sample of 30 respondents was considered for scale development, selected randomly from the sampling frame. The study included three different product categories: khadi, detergents & chemicals, and handmade paper.

Of the external and internal factors, four factors--rate of innovation, quality differentiation of products, competition in the industry, and the target market--were chosen for measurement. Positively worded items wherein favourable attitudes were ranked higher on the scale were included to reduce "response pattern bias" and "systematic error," which may arise due to some imperfect aspect of the research design. To measure the factors a 5-point Likert scale was developed, with a neutral measure and the ratings moving from negative to positive.

The next step in the process is to to ensure "reliability" of the new measure by checking for internal consistency of the factors. The most common internal consistency measure is Cronbach's alpha, which is usually interpreted as the mean of all possible "split-half coefficients." Reliability tests conducted yielded Cronbach alpha (standarised) values for the various factors: rate of innovation (0.8333), quality differentiation of products (0.7478), competition in the industry (0.8057), and target market (0.8627). All the alpha values are above 0.7, which indicate satisfactory internal consistency. If the alpha values were less than 0.7, then some of the items would have to be deleted and the instrument checked for "content validity."

The use of a small sample size and not using test-retest reliability for internal consistency were cited as limitations. Non-usage of the LISREL technique for "confirmatory factor analysis" was cited as another limitation in the scale development process. The use of LISREL would have led to better reliability of the scale.

Some of the findings from the field were presented. Organisations with high rate of innovation, higher number of product categories, higher intensity of competition, and greater diversity in usage pattern of consumers laid greater emphasis on the "push" strategy (free sampling and personal selling with the use of Internet in the handmade paper industry).

There was some discussion on the assumptions involved in quantitative data analysis. For example, it was suggested that the characteristics of the "statistical population" need to be well defined in order to ensure the reliability of the findings.

References

Armstrong, G., & Kotler, P. (2004). Marketing: An introduction, 7th Edition. Upper Saddle River, NJ: Pearson Prentice Hall.

Hinkin, T. (1995). A review of scale development practices in the study of organisations. Journal of Management, 21(5), 967-988.

Korman, A. K. (1974). Contingency approaches to leadership. In J. G. Hunt & L. L. Larson (Eds.), Contingency approaches to leadership (pp. 189-195). Carbondale: Southern Illinois University Press.

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory, 3rd Edition. New York: McGraw-Hill.

Schwab, D. P. (1980). Construct validity in organization behavior. In B. M. Staw, & L. L. Cummings (Eds.), Research in organizational behavior, Vol. 2. (pp. 3-43 ). Greenwich, CT: JAI Press.

Zikmund, W. G. (2003). Business research methods, 7th Edition. London: Thomson Learning.

Reported by C. D. Kuruvilla, with inputs from J. D. Vakkayil and  A. G. Menon (8 February 2006).


Copyleft The article may be used freely, for a noncommercial purpose, as long as the original source is properly acknowledged.