Employment test validation for minority and nonminority telephone company service representatives.

Abstract

Samples of minority (black) and nonminority, newly hired telephone company service representatives participated in a test validation study. Tests administered during employment were related to specially developed proficiency criteria. Generally, individual and composite test and criterion averages obtained by the ethnic samples differed significantly, but validity coefficients were comparable. Regression equation comparisons indicated that common test standards could be used to evaluate minority and nonminority job applicants.

Employment Test Validation for Minority and Nonminority Production Workers

“…In all three cases (combined, minority, nonminority), the same tests (Rule Measurements and Specifications Sheet) were selected providing additional evidence of the lack of differential validity. Following the selection of the tests, scores o n the Rule Measurements and Specifications Sheet tests were combined by summing, thus ignoring their respective beta weights (see Gael and Grant, 1972;Gael, Grant, and Ritchie, 1975). This decision was influenced by practical considerations regarding future test use in employment offices.…”

“…THE fair use of tests in employment settings necessitates an examination of their validity. As Gael and Grant (1972) have noted, "employment test bias, as is the case with test utility, cannot be gauged unless the tests are related to meaningful job standards" (p. 135). Kirkpatrick, Ewen, Barrett, and Katzell (1968) have pointed out that the judgment of "fair test use" should be made with respect to two major considerations.…”

Work Sample Testing

“…Work sample tests were developed as criteria to validate selection tests for telephone service representatives by Gael and Grant (1972). The employees had to deal with a number of typical telephone calls and the resulting administration such as record preparation and filing.…”

Evaluating Test Fairness

“…In fact, several studies carried out in the 1970s support the finding that minority groups are rated less highly on objective criterion measures (e.g. Campbell, Crooks, Mahoney and Rock 1973;Gael and Grant 1972;Gael, Grant and Richie 1975a, 1975b, cited in Schmidt 1993. This body of evidence has led Schmidt to conclude 'The cumulative research on test fairness shows that the average ability and cognitive skill differences between groups are directly reflected in job performance' (Schmidt 1993, p. 5).…”