Army Synthetic Validity Project: Report of Phase 3 Results. Volume 2. Research Instruments.
Abstract
The Army Synthetic Validity Project developed and evaluated a series of alternative procedures for (a) analyzing critical components of jobs, (b) obtaining expert judgments of validities of individual attributes for predicting critical performance components, (c) establishing prediction equations for specific jobs when criterion-related validation data were not available, (d) estimating criterion-referenced performance standards for specific jobs, and (e) specifying scores on the predictor battery necessary to achieve the desired performance standard, given the bivariate distribution between predictor scores and performance scores. The results and conclusions obtained for each of these activities follow. As a consequence of results obtained in earlier phases of the project, the attribute model and the job behavior method were set aside and the questionnaire based on tasks performed became the tool of choice. While all methods provided reliable descriptions, the task questionnaire yielded greater discriminability across MOS and had higher acceptability among the judges.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1991
- Accession Number
- ADA235636
Entities
People
- Jane M. Arabian
- John P. Campbell
- Lauress L. Wise
- Norman G. Peterson
- R. G. Hoffman
Organizations
- American Institutes for Research