Task Level Job Performance Criteria Development
AMERICAN INSTITUTES FOR RESEARCH WASHINGTON DC
Pagination or Media Count:
This study investigated the possibilities for improving the identification of the requirements for jobs by studying performance of job incumbents on separate tasks. Three specialties were selected for study 291X0, Telecommunications Operations Specialist 304X4, Ground Radio Communications Equipment Repairman 431X1C, Aircraft Maintenance Specialist, single- and dual-engine jet. Incumbents, peers, and supervisors rated the performance of the incumbents on a selected set of tasks. In addition, job inventories and an experimental test battery were administered to the incumbents. Data of record were also obtained from Air Force files to provide such items as incumbent grade, service time, sex, education at enlistment, and Aptitude Index scores. Correlations were run between raters, correlating performance on separate taks, and between raters, correlating performance on 6 overall dimensions of appraisal. Cross-rater reliabilities were low, but significant, on task assessments, and in the r .40 range on overall ratings. Similarly low correlations were found for nontask predictors, such as grade, service time, and aptitude indexes. All types of obtained measures, except data on the origins of training and on task performance satisfaction, were put into regression problems to account for the 6 overall performance ratings made by peers and supervisors. The data suggest that different factors were important for different kinds of work, and for different dimensions of performance appraisal. By far the most enlightening finding was that difficult tasks in terms of learning time were better measured on performance.
- Personnel Management and Labor Relations