Tailored Testing Theory and Practice: A Basic Model, Normal Ogive Submodels, and Tailored Testing Algorithms

Abstract

In this report, selection theory is used as a theoretical framework from which mathematical algorithms for tailored testing are derived. The process of tailored, or adaptive, testing is presented as analogous to personnel selection and rejection on a series of continuous variables that are related to ability. Proceeding from a single common-factor model, the author derives the two- and three-parameter normal ogive item response functions as submodels. For both of these submodels, algorithms are developed for sequential item selection, ability estimation, and test termination in the context of adaptive ability testing. It is shown that the adaptive testing method based on these algorithms is formally identical to a previously developed Bayesian sequential tailored testing procedure.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1983
Accession Number
ADA133385

Entities

People

  • Vern W. Urry

Tags

Communities of Interest

  • Biomedical
  • C4I
  • Human Systems

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Computational Science
  • Databases
  • Distribution Functions
  • Estimators
  • Human Factors Engineering
  • Human Resources
  • Information Science
  • Management Personnel
  • Military Research
  • Normal Distribution
  • Personnel Management
  • Personnel Selection
  • Random Variables
  • Security
  • Test Methods

Fields of Study

  • Mathematics

Readers

  • Computational Fluid Dynamics (CFD)
  • Operations Research
  • Psychometric Testing or Psychological Assessment.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms