DID YOU KNOW? DTIC has over 3.5 million final reports on DoD funded research, development, test, and evaluation activities available to our registered users. Click HERE
to register or log in.
DESCRIBING AIRMAN PERFORMANCE IN THE ADMINISTRATIVE CAREER LADDER BY IDENTIFYING PATTERNS OF TRAIT RATINGS,
PERSONNEL RESEARCH LAB LACKLAND AFB TEX
Pagination or Media Count:
Trait ratings were used to account for the variance in airman performance reports and in overall experimental performance ratings. Airmen in the administrative career ladder, DAFSCs 70230, 50, 70, and 70490, across all commands, were rated by supervisors on overall performance and on 65 traits. Current overall airman performance reports APRs were obtained from base records. Among the 2,606 sets of ratings with complete data, 1,083 individuals were evaluated twice, representing personnel rated by two supervisors. Broken down by skill levels, the smallest N was 140, for 9-level men who had been rated twice. Using data undifferentiated by skill, in which a man might appear twice if so rated, trait ratings accounted for 70 per cent of the variance in experimental performance ratings and about 43 per cent of the variance in APRs, after grade was removed as a predictor. When data were sorted by skill level, prediction held up in all skills except DAFSC 70270, where it dropped to 60 per cent. Patterns of traits which were more predictive of performance in one skill level than another were found, and these patterns could be sensibly interpreted in terms of the expected demands of the jobs. In a cross-validation against different raters, the predictive advantage of selected patterns was found to be statistically significant for the 5-, 7-, and 9-skill levels. The study is discussed in terms of its implications for criterion development, particularly in respect to its place in the sequence of current criterion research studies. Author
APPROVED FOR PUBLIC RELEASE