Evaluations of Implied Orders as a Basis for Tailored Testing Using Simulations.
Abstract
TAILOR is a computer program that uses the implied orders concept as the basis for computerized adaptive testing. The basic characteristics of TAILOR, which does not involve pretesting, are reviewed here and two studies of it are reported. One is a Monte Carlo simulation based on the Birnbaum model and the other nuses a matrix of item responses to the Stanford-Binet. In the Birnbaum model study, a variety of conditions were simulated and it was found that TAILOR typically used responses to about half the items and achieved validities with true score within a few points of the validity of the complete test. Item discrimination parameters affected the efficiency of TAILOR. The Binet study used correlations between scores based on one bank of Tailored items and another independent, parallel set and found results similar to those in the Birnbaum simulation. It appears that TAILOR, like other adaptive testing systems, can aid efficiency where item discriminations are high or, equivalently, ability variance is large. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1977
- Accession Number
- ADA047931
Entities
People
- Douglas Mccormick
- Norman Cliff
- Robert Cudeck
Organizations
- University of Southern California