Examining the role of opportunity to learn in the knowledge test-training performance relationship.

摘要

Past research has shown that knowledge tests provide reasonable levels of validity and incremental validity (over general cognitive ability), but they often yield large subgroup differences. This study examined the role of opportunity to learn (OTL) in the knowledge test-training performance relationship. Specifically, the study was based on the supposition that to the extent a given knowledge test covers content to which an examinee has had the opportunity to learn, then good (or poor) performance on that test may be considered predictive of relative success (or failure) in future knowledge acquisition situations. The study assessed OTL as a moderator of the knowledge test-training performance relationship.;Two samples from the Armed Services were used in the study. The first---Subgroup Differences Sample---consisted of 2,650 recruits. The second---Validation Study Sample---consisted of 2,507 trainees from twelve Army and Air Force technical training courses at three different bases. Science and technical (S&T) knowledge was measured using alternative versions of the S&T tests of the Armed Services Vocational Aptitude Battery (ASVAB)---Auto & Shop Information, Electronics Information, General Science, and Mechanical Comprehension. Opportunity to learn was assessed using a self-report questionnaire, which assessed exposure to test content and the quality of this exposure. Hierarchical regression analyses were used to examine the relationships.;Results provided little evidence to suggest that OTL acted as a moderator of the knowledge test-training performance relationship. Instead, the results suggested that OTL may influence these relationships by acting as a suppressor of the knowledge test-training performance relationship. Based on these results, alternative test scores were derived that either minimized or maximized the relationship between OTL and the associated knowledge test scores, and the validities, incremental validities (over general cognitive ability), and subgroup differences associated with these test scores were then compared. When compared to the test scores that maximized the OTL-knowledge test relationship, the test scores that minimized this relationship yielded (a) somewhat higher validities when OTL yielded incremental validity over the associated test score, (b) slightly lower incremental validities (over general cognitive ability), and (c) lower subgroup differences, especially gender subgroup differences. Implications and recommendations are discussed.