Realistic Work Sample Tests: A Review

Abstract

INFORMATION with the highest validity seems to have a pobnt-topoint correspondence with the criterion. For instance, in a review article of the scorable application blank, Asher (1972) concluded that biographical information showed substantially higher predictive validity when job proficiency was the criterion, than other predictors including tests of intelligence, personality, interest, perception, motor skill, and mechanical ability.The generalization seemed t o be that factual and verifiable historical information about applicants was the best predictor of future performance in specific positions. This implication was based on comparisons with tests designed t o measure "simple" dimensions. Simple in this context means a pure or single dimension of behavior.If there is merit in a theory about predictive power in a pointto-point connection between the predictor and criterion space, then tests of single dimensions should be less powerful predictors than more complex tests such as work sample tests designed to be a miniature replica of the criterion task. The intention was not to include in this review ready-made standardized tests, but only work samples especially created for specific criterion tasks.In a search of the literature, work sample tests were classified into the categories of either motor or verbal. A work sample was identified as "motor" if the task was a physical manipulation of things as, for example, tracing a complex electrical circuit, operating a sewing machine, making a tooth from plaster, or repairing a gear box. A work sample test was classified as "verbal" if there was a problem situation that was primarily language-oriented or people-oriented. For instance, the in-basket problem required the subject to deal with a set of materials which a company manager

Gender Differences in Work Sample Assessments: Not All Tests Are Created Equal

“…Meta-analytic reviews have indicated that work samples have high criterion-related validity for predicting job performance (Roth, Bobko, & McFarland, 2005;Schmidt & Hunter, 1998). Furthermore, as work samples have a higher degree of fidelity to tasks performed on the job than traditional individual differences predictors (Asher & Sciarrino, 1974), they are generally associated with more favorable applicant reactions (e.g., Hausknecht, Day, & Thomas, 2004;Macan, Avedon, Paese, & Smith, 1994).…”

Work Sample Tests: Their Relationship with Job Performance and Job Experience

“…Based on these results, Roth et al (2005) emphasized that work sample validity could be noticeably lower than previously thought and therefore some organizations may be overestimating the validity properties of work samples. Despite the lower validity estimates that were obtained in more recent studies, which have overcome important limitations related with conceptual and methodological problems (Asher & Sciarrino, 1974) on the one hand, and the limited scope on the other, in general, the directionality of the results of the studies converge on an indication that work sample tests are valid predictors of job performance.…”

Simulations for Personnel Selection

“…In my first week of graduate school, I found myself assigned to write the inaugural paper of my career on the subject of work samples. Though I knew almost nothing about personnel selection at that time, my research led me to Asher and Sciarrino (1974), an article that immediately had a strong intuitive appeal and still remains a foundational element in my philosophy of personnel selection. In this article, the authors introduce the concept of "point to point correspondence," a concept suggesting that prediction is enhanced when predictors are designed to be miniature replicas of the criterion space.…”