Dimensionality Assessment in Factor Analysis: An Empirical Kaiser Criterion

• In Event: 72.066 - Poster Session 16
  In Poster Session: 72.066-2 - Poster Session: Practical Issues in Measurement

Abstract

In factor analysis, most popular methods for dimensionality assessment such as the screeplot, the Kaiser criterion, or – the current gold standard – parallel analysis, are based on eigenvalues of the correlation matrix. To further understanding and development of factor retention methods, results from random matrix theory on empirical eigenvalue distributions are introduced. These results are used to develop an empirical version of the Kaiser criterion targeting typical research settings in which multiple scales are setup which are desired to be relatively short, but still reliable. Theoretical and simulation results will be presented to illustrate that parallel analysis is less effective in these settings and the performance of the new empirical Kaiser criterion will be evaluated.

Authors

• Johan Braeken, University of Oslo

• Marcel van Assen, Tilburg University