Parallel Analysis and Velicer’s MAP Test
Searching the web for some relevant references about Generalizability Theory, I discover Brian O’Connor’s website. He’s developing several SPSS, SAS and Matlab programs to do some psychometrical analysis. I was particularly interested in his notes on parallel analysis and the reasons why we have to perform factor analysis on dichotomously scored items or Lickert items very carefully. Here is the related reference:
O’Connor, B.P. (2000). SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavior Research Methods, Instrumentation, and Computers, 32, 396-402.
There are several drawbacks with the usual R function that handles factor analysis, factanal, which I discuss in the following post: Exploratory Factor Analysis with principal factors.
Pas encore de commentaire.
Laisser un commentaire
Archives
- décembre 2008 (9)
- novembre 2008 (2)
- octobre 2008 (1)
- septembre 2008 (3)
- août 2008 (3)
- juillet 2008 (9)
- juin 2008 (2)
- mai 2008 (12)
- avril 2008 (6)
- mars 2008 (3)
- février 2008 (5)
- janvier 2008 (1)
Tags
artificial intelligence bayesian bioinformatics Biostatistics c clinical trials CTT epidemiology equating factor analysis Ggobi GLM graphs inter-rater agreement interaction IRT judgement latent localization logistic machine learning Mac OS X missing data mixed models multilevel networks nonlinear PCA perl programming psychiatry psychometrics python QoL questionnaire R regression reliability social network social sciences statistics survey visualization web XMLCatégories
La citation du jour
by William Shakespeare, Antony and Cleopatra