# Why would the results of PCA differ from a confirmatory factor analysis?

I have conducted a confirmatory factor analysis (CFA) to test the fit of a model with 5 factors and 5 items per factor. I used the modification indices to alter the model until I obtained statistics indicating an acceptable fit of the model to my data.

As others have done in my area I then conducted a principal components analysis (PCA) with a varimax rotation to check that items loaded on the factors identified by the CFA; however, not all of them did.

• Is it unusual for the results of a PCA to be inconsistent with a confirmatory factor analysis?
• Could this be because I have a small sample size ($n=96$)?
• As the CFA model has 'acceptable fit', should I simply work with the scale structures from those results and not do the PCA?

In both case, but especially CFA, $N=96$ is a very limited sample size. Although some authors have suggested a ratio individuals:items of 5 to 10, this is merely the number of dimensions that is important. In your case, the estimation of your parameters will be noisy, and in the case of PCA you may expect fluctuations in your estimated loadings (just try bootstrap to get an idea of 95% CIs).