ABILITY ESTIMATION FOR CONVENTIONAL TESTS

Five different ability estimators-maximum likelihood [MLE (theta)], weighted likelihood [WLE (theta)], Bayesian modal [BME (theta)], expected a posteriori [EAP (theta)] and the standardized number-right score [Z (theta)]-were used as scores for conventional, multiple-choice tests. The bias, standard error and reliability of the five ability estimators were evaluated using Monte Carlo estimates of the unknown conditional means and variances of the estimators. The results indicated that ability estimates based on BME (theta), EAP (theta) or WLE (theta) were reasonably unbiased for the range of abilities corresponding to the difficulty of a test, and that their standard errors were relatively small. Also, they were as reliable as the old standby-the number-right score.