Item analysis of unipolar item response data
Psychometrics and Research Methodology
Department of Psychology
Faculty of Social and Behavioral Sciences
Project running from: 1 December 2002 – 10 September 2008
Promotor(es): Prof. dr W.J. Heiser, Dr. M. de Rooij
The project aims at contributing to the development of a full-blown item analysis of unipolar (single-peaked) items. Correspondence analysis will be used as a method for the multidimensional representation of item response data, and a coefficient of single-peakedness will be developed that measures the strength of the nonlinear relationship between item responses and person scores (or item scores). This coefficient will also be used to define a measure of reliability. The new methodology will be tested on a number of clinical test data, as well as on simulated data.
Date of defence: 26 May 2011
Title of thesis: Item analysis of single-peaked response data: The psychometric evaluation of bipolar measurement scales. ISBN: 978-94-6169-080-7.