THE RELATIONSHIP OF SPECIFIC MENTAL-ABILITY MEASURES COMPARED TO A GENERAL MENTAL-ABILITY MEASURE TO QUALITY AND QUANTITY PERFORMANCE ON A CLERICAL JOB SAMPLE

Barrett (1993) presented evidence that mental ability measures designe d specifically for the prediction of performance in tasks and jobs hav e validity at least equal to that of tests of general mental ability, while having advantages such as lower adverse impact. The current stud y demonstrates that the match in specificity of predictor and criterio n can be attained through simple changes in the scoring of standard me ntal and clerical ability tests. A sample of 51 university students co mpleted a battery of pre-employment tests and, a week or two later, wo rked on a clerical job sample. The results indicated that quality of p erformance (number of errors on the task) was best predicted using the number of errors made on the predictor tests while quantity of work w as only predicted by the number correct. A test of general mental abil ity did not correlate with either criterion. These results contradicte d the conclusions of Schmidt (1993) that specific ability tests have n o incremental validity above a general ability test.