Effects of Immediate Feedback and Pacing of Item Presentation on Ability Test Performance and Psychological Reactions to Testing

摘要

The study investigated the joint effects of knowledge of results (KR or no-KR), pacing of item presentation (computer or self-pacing), and type of testing strategy (50-item peaked conventional, variable-length stradaptive, or 50-item fixed-length stradaptive test, on ability test performance, test item response latency, information, and psychological reactions to testing. The psychological reactions to testing were obtained from likert-type items that assessed test-taking anxiety, motivation, perception of difficulty, and reactions to knowledge of results. Data were obtained from 447 college students randomly assigned to one of the 12 experimental conditions. The results indicated that there were no effects on ability estimates due to knowledge of results, testing strategy, or pacing of item presentation. although average latencies were greater on the stradaptive tests than on the conventional test, the overall testing time was not substantially longer on the adaptive tests and may have been a function of differences in test difficulty. Analysis of information values indicated higher levels of information on the stradaptive tests than on the conventional test. There was no statistically significant main effect for any of the three experimental conditions when test anxiety or test taking motivation were the dependent variables, although there were some significant interaction effects.