An Empirical Investigation of the Stratified Adaptive Computerized Testing Model

Abstract

This study empirically investigated the validity and utility of the stratified adaptive computerized testing model (stradaptive) developed by Weiss (1973). The model presents a tailored testing strategy based upon Binet IQ measurement theory and Lord’s (1972) modern test theory. Nationally normed School and College Ability Test Verbal analogy items (SCAT-V) were used to construct an item pool. Item difficulty and discrimination indices were re-scaled to normal ogive parameters on 244 items. One hundred and two freshmen volunteers at Florida State University were randomly assigned to stradaptive or conventional test groups. Both groups were tested via cathode-ray-tube (CRT) terminals coupled to a Control Data Corporation 6500 computer. The conventional subjects took. a SCAT-V test essentially as published, while the stradaptive group took individually tailored tests using the same item pool. Results showed significantly higher reliability for the stradaptive group, and equivalent validity indices between stradaptive and conventional groups. Three stradaptive testing strategies averaged 19.2, 26.5, and 31.5 items per subject as compared with 48.4 items per conventional subject. A 50% reduction from conventional test length produced equal precision of measurement for stradaptive subjects. Item latency comparisons showed that those in the stradaptive group required significantly longer per item (about 11%) than conventional group members. It is

recommended that testing time rather than number of items be used as a dependent variable in future adaptive testing research.