Search

Occupational norms exist for some tests, and differences in the distributions of test scores by occupation are evident. Sampling error (SE), situationally specific factors (SSFs), and differences in job content (DIJCs) were explored as possible reasons for the observed differences. SE was explored by analyzing 742 validity studies performed by the U.S. Employment Service using the General Aptitude Test Battery (GATB). The effects of SSF variance were explored by correlating GATB scores from pairs of validity studies for 175 occupations. Because the combination of SE and SSF variance is not sufficient to account for all the variance observed, DIJCs were also considered. It was found that test scores, either in individual or aggregated (mean) form, were significantly correlated with job data from the Position Analysis Questionnaire (PAQ), and regression equations could be developed to estimate occupational norms. Using such equations, it is possible to estimate norms for jobs on the basis of over 2,500 PAQ occupational profiles or by collecting PAQ data for new jobs. This procedure enhances the possible usefulness of tests, especially for those for which extensive occupational norms are not currently available. Two flowcharts and four tables present data from the analyses. A 24-item list of references is included. (SLD)