Accession Number:



Determining Composite Validity Coefficients for Army Jobs and Job Families

Descriptive Note:

Final rept. Mar-Sep 2002

Corporate Author:


Report Date:


Pagination or Media Count:



The broad goal of the present research and the first study completed in response to the September 2001 Expert Panel recommendations is to compute composite validity coefficients. using criterion data derived from the 1987 - 1989 Skill Qualifications Test program, for the 7-test ASVAB for 150, 17, and 9 job family structures. These are the structures underlying ongoing classification research. The specific research objectives are as follows 1. To compute the 7-test ASVAB LSE least squares estimate composite validity coefficients for the first-tier 150 job family structure. These correlation coefficients are corrected, first, for unreliability of the criterion and, then, for restriction in range effects due to assignment from an Army input population to MOS samples. The coefficients are computed for both back biased and cross unbiased validities of LSE composites. 2 To compute ASVAB composite validity coefficients for the youth population in the 150 job family structure. This involves a correction for the Army input and then a separate restriction in range correction due to selection from the youth population into the Army. Again, the coefficients are computed for bath back and cross validities. 3. To compare mean validity coefficient results obtained for the 150 job families with those obtained earlier for the 66 MOS families. Although there was a substantial overlap in MOS between the two data sets, the 66 MOS study was computed on data that was collected several years earlier than was the 150 family study. 4. To compute the weighted aggregation of test composite validity coefficients for the aggregated MOS corresponding to each of the 17 job family composites of the second tier and for each of the 9 interim composites. Validities are first corrected for the Army input population and then corrected for the youth population for both back and cross samples.

Subject Categories:

  • Psychology
  • Personnel Management and Labor Relations

Distribution Statement: