Item selection and hypothesis testing for the adaptive measurement of change

Matthew D. Finkelman, David J. Weiss, Gyenam Kim-Kang

Research output: Contribution to journalArticle

21 Scopus citations

Abstract

Assessing individual change is an important topic in both psychological and educational measurement. An adaptive measurement of change (AMC) method had previously been shown to exhibit greater efficiency in detecting change than conventional nonadaptive methods. However, little work had been done to compare different procedures within the AMC framework. This study introduced a new item selection criterion and two new test statistics for detecting change with AMC that were specifically designed for the paradigm of hypothesis testing. In two simulation sets, the new methods for detecting significant change improved on existing procedures by demonstrating better adherence to Type I error rates and substantially better power for detecting relatively small change.

Original languageEnglish (US)
Pages (from-to)238-254
Number of pages17
JournalApplied Psychological Measurement
Volume34
Issue number4
DOIs
StatePublished - Jun 1 2010

Keywords

  • Change
  • Computerized adaptive testing
  • Individual change
  • Kullback-Leibler information
  • Likelihood ratio
  • Measuring change

Fingerprint Dive into the research topics of 'Item selection and hypothesis testing for the adaptive measurement of change'. Together they form a unique fingerprint.

  • Cite this