Optimizing the Use of Response Times for Item Selection in Computerized Adaptive Testing

Despite common operationalization, measurement efficiency of computerized adaptive testing should not only be assessed in terms of the number of items administered but also the time it takes to complete the test. To this end, a recent study introduced a novel item selection criterion that maximizes Fisher information per unit of expected response time (RT), which was shown to effectively reduce the average completion time for a fixed-length test with minimal decrease in the accuracy of ability estimation. As this method also resulted in extremely unbalanced exposure of items, however, a-stratification with b-blocking was recommended as a means for counterbalancing. Although exceptionally effective in this regard, it comes at substantial costs of attenuating the reduction in average testing time, increasing the variance of testing times, and further decreasing estimation accuracy. Therefore, this article investigated several alternative methods for item exposure control, of which the most promising was a simple modification of maximizing Fisher information per unit of centered expected RT. The key advantage of the proposed method is the flexibility in choosing a centering value according to a desired distribution of testing times and level of exposure control. Moreover, the centered expected RT can be exponentially weighted to calibrate the degree of measurement precision. The results of extensive simulations, with item pools and examinees that are both simulated and real, demonstrate that optimally chosen centering and weighting values can markedly reduce the mean and variance of both testing times and test overlap, all without much compromise in estimation accuracy.