Improved Estimation Procedures for Item Response Functions.

Abstract

This report summarized the main findings of a research which focused on the estimation of parametric item response curves under the assumption that univariate ability parameters are sampled from some parametric population distribution. Both maximum likelihood and Bayesian approaches have been studied and compared to the more conventional approaches where abilities are treated as fixed parameters. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 1984
Accession Number
ADA147677

Entities

People

  • R. K. Tsutakawa

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Algorithms
  • Bayesian Networks
  • Behavioral Sciences
  • Computational Science
  • Computations
  • Demography
  • Education
  • Educational Psychology
  • Military Research
  • Models
  • Probability
  • Psychology
  • Schools
  • Social Sciences
  • Statistics
  • United States
  • Universities

Fields of Study

  • Mathematics

Readers

  • Psychometric Testing or Psychological Assessment.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference