Scale analysis and the measurement of social attitudes

This paper discusses and compares the methods of attitude scale construction of Thurstone (method of equal-appearing intervals), Likert (method of summated ratings), and Guttman (method of scale analysis), with special emphasis on the latter as one of the most recent and significant contributions to the field. Despite a certain lack of methodological precision, scale analysis provides a means of evaluating the uni-dimensionality of a set of items. If the criteria for uni-dimensionality are met, the interpretation of rank-order scores is made unambiguous, and efficiency of prediction from the set of items is maximized. The Guttman technique, however, provides no satisfactory means of selecting the original set of items for scale analysis. Preliminary studies indicated that both the Likert and the Thurstone methods tend to select scalable sets of items and that their functions in this respect are complementary. A method of combining the Likert and Thurstone methods in order to yield a highly scalable set of items is outlined. Sets of 14 items selected by the method have, in the two cases where the technique has been tried, yielded very satisfactory scalability.     

THE MEASUREMENT OF WORK: HIERACHICAL REPRESENTATION OF THE MULTIMETHOD JOB DESIGN QUESTIONNAIRE

To study the changing nature of work, researchers need measures of work that are valid and comprehensive. One potentially useful measure of work is the Multimethod Job Design Questionnaire (MJDQ; Campion, 1988), which was developed to assess 4 general approaches to work design (i.e., motivational, mechanistic, biological, perceptual-motor). Although the MJDQ holds promise as a general measure of work, little information is available regarding its psychometric properties. This study examines the MJDQ, using alternative hierarchical factor structures to represent work at varying levels of abstraction. Little support was found for the 4-factor structure corresponding to the work design approaches underling the MJDQ or for various hierarchical factor structures that simultaneously depicted general and specific aspects of work. However, a 10-factor first-order model received good support and may provide a useful basis for scoring and interpreting the MJDQ in future research.     

ORGANIZATIONAL DEVELOPMENT INTERVENTIONS: A META-ANALYSIS OF THEIR EFFECTS ON SATISFACTION AND OTHER ATTITUDES

This meta-analysis examined the effects from 126 studies that employed organizational development (OD) interventions to modify satisfaction and/or other attitudes. Raju and Burkes's (1983) Taylor series approximation 1 (TSA 1) validity generalization procedure was used to conduct the meta-analysis. Multi-faceted interventions have been more effective in modifying satisfaction and attitudes than were OD interventions that used a single human-processes or technostructural technique. For specific interventions, team building and lab training were the most effective means of changing satisfaction and other attitudes. Also, OD seems to affect attitudes more than satisfaction. Caution is encouraged in generalizing these findings, given (1) the non-random selection of techniques, (2) the frequent moderating effects of both the participant's organizational level and the methodological rigor of the studies, and (3) this study's frequent failure to account for a substantial amount of the variability of the effect sizes.     

PROBLEMS WITH THE USE OF PROFILE SIMILARITY INDICES IN THE STUDY OF CONGRUENCE IN ORGANIZATIONAL RESEARCH

Profile similarity indices (PSIs) have become widely used in studies of congruence (i.e., fit, matching, similarity, agreement) in organizational research. PSIs combine two sets of measures, or profiles, from corresponding entities (e.g., the person and organization, supervisor and subordinate, organization and environment) into a single score intended to represent their overall congruence. Unfortunately, PSIs are conceptually ambiguous, discard information essential to testing congruence hypotheses, conceal the source of the difference between entities, and impose a highly restrictive set of constraints on the coefficients relating the measures comprising the PSI to the outcome. This article shows how polynomial regression analysis may be used to avoid problems with PSIs while capturing the underlying relationships PSIs are intended to represent. Limitations and extensions to the procedure are discussed.