Simulation; Testing

A computerized flexilevel test was implemented and its ability estimates were compared with those of a Bayesian estimation based computerized adaptive test (CAT) as wen as with known true ability estimates. Results showed that when the flexilevel test was terminated according to Lord's criterion, its ability estimates were highly and significantly correlated with those of the Bayesian based CAT and with the known true thetas. Matched t-tests between the flexilevel and the Bayesian CAT and with the known true thetas were not significantly different from one another. The primary implication of this study is that it is feasible to implement a computerized adaptive test without having to create a calibrated item pool containing 200 or more items, AS is the case with Item Response Theory (IRT) based CAT systems. Neither the software nor the hardware needed for flexilevel CAT methods present the ratb4r stciou