Black/White Differences in the Prediction of Success in Air Force and Marine Corps Technical Training Programs.
Abstract
Johnson-Neyman methodology is used to investigate differential prediction by race in Marine Corps and Air Force occupational specialty areas. For the Air Force data set, meaningful Johnson-Neyman regions of differences were found in 23 of 28 comparisons. In all cases the regressions for blacks were flatter than for whites and the black/white interaction was disordinal. In 12 cases the cut-score for course qualification was within the Johnson-Neyman region and in 10 of the 12 cases the bias was negative for blacks. It is also noted that if the cut-scores had been set substantially higher the bias would have been positive for blacks in all cases. It is suggested that this analysis explains why earlier studies which averaged bias across values of x yielded mixed results. It is also suggested that the consistent result obtained here results from the lower predictability found in the black subpopulation. Keywords: ASVAB(Armed Services Vocational Aptitude Battery) Differential Prediction, Air Force and Marine Differential Prediction.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 13, 1985
- Accession Number
- ADA169817
Entities
People
- Melvin R. Novick
- Walter M. Houston
Organizations
- University of Iowa