Raven, J., & Fugard, A. (2008, May 23). What’s Wrong with Factor-Analysing Tests Conforming to the Requirements of Item Response Theory? WebPsychEmpiricist. Retrieved May 23, 2008 from http://wpe.info/papers_table.html.

Although many of those who are familiar with Item Response Theory (IRT) are well aware that factor analysing matrices of correlations between the items constituting such tests tends to yield misleading results, this is not as well known as it should be. In fact, endless researchers have come to seriously misleading conclusions as a result of applying factor analysis to such tests. The current paper illustrates just how misleading these results can be by factor analysing computer-generated data simulating that which would be obtained from the use of that ultimate form of an IRT test – a tape measure or meter stick – to measure height.

Download the pdf