Group composition as a determinant of job analysis outcomes.

The purpose of this study was to determine if different groups of job incumbents would provide different job analysis outcomes. Patrolpersons from 12 university police departments served as subjects. Subject-matter-expert groups were formed on the basis of a multidimensional scaling (MDS) procedure, seniority level, and educational level. The job analysis technique used was the critical incident technique. Multivariate analysis of variance indicated that MDS groups differed on ratings and rank orderings of job dimension importance. These MDS group differences were probed via discriminant analysis, which revealed that the MDS groups differed on one discriminant function for the job dimension importance ratings and on two discriminant functions for the job dimension importance rank orderings. The MDS group's discriminant function differences were also related to supervisor rank orderings. Job analysis outcomes were not affected by subject-matter-expert groups of seniority and educational level. Results are discussed in terms of their implications for the job analysis practitioner. (PsycINFO Database Record (c) 2010 APA, all rights reserved)