Monte Carlo evaluation of implied orders as a basis for tailored testing

Abstract

TAILOR, the computer program which implements an approach to tailored testing outlined by Cliff (1975), was examined with errorless data by monte carlo methods. Three replications of each cell of a 3 × 3 table with 10, 20, and 40 items and persons were analyzed. Mean rank correlation coefficients between the true person and item order, specified by preassigned random numbers, and the computed order produced by the program averaged .96. The average proportion of items used was .48. A marked tendency was observed for the program to produce better results as the number of persons and items increased.