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)

Open PDF

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

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Age Groups
  • Business Administration
  • Central Processing Units
  • Computer Programs
  • Computers
  • Data Science
  • Education
  • Information Science
  • Marine Corps
  • Military Research
  • New York
  • Psychology
  • Regression Analysis
  • Reliability
  • Social Sciences
  • Statistics
  • Training

Readers

  • Psychometric Testing or Psychological Assessment.
  • Regression Analysis.