Construct Explication through Factor or Component Analysis: A Review and Evaluation of Alternative Procedures for Determining the Number of Factors or Components

Abstract

Several years ago, in a paper entitled “Construct validity and personality assessment” Jackson (1979) argued that the greatest impact of the construct approach to psychological measurement had been felt in the area of personality. In describing his own work, and that of others, he demonstrated the value of this approach for understanding personality and for personality scale construction. As in personality assessment work, much of the research conducted by organizational psychologists relies on questionnaire measures developed to assess complex constructs. It is interesting to note that, around the same time as Jackson's comments, Schwab (1980, p.34) lamented that “organizational behavior has suffered” from a lack of attention to construct validation. In the intervening 20 years, attention to the construct and its measurement has increased in organizational research. Nonetheless, one could not argue that the construct approach has gained center stage in this area or that concerns about conceptual and measurement adequacy in organizational research have been ameliorated dramatically. Indeed, in recent discussions of questionnaire measures developed for use in organizations, Hinkin (1995, 1998) suggested that concerns about adequate construct measurement represented “perhaps the greatest difficulty in conducting survey research” (Hinkin, 1995, p. 967).