Factor Analysis: A Short Introduction, Part 6–Common Problems

by Maike Rahn, PhD

In the previous blogs I wrote about the basics of running a factor analysis. Real-life factor analysis can become complicated. Here are some of the more common problems researchers encounter and some possible solutions:

  • The factor loadings in your confirmatory factor analysis are only |0.5| or less.

Solution: lower the cut-offs of your factor loadings, provided that lower factor loadings are expected and accepted in your field.

  • Your confirmatory factor analysis does not show the hypothesized number of factors.

Solution 1: you were not able to validate the factor structure in your sample; your analysis with this sample did not work out.

Solution 2: your factor analysis has just become exploratory. Something is going on with your sample that is different from the samples used in other studies. Find out what it is.

  • A few key variables in your confirmatory factor analysis do not behave as expected and/or are correlated with the wrong factor.

Solution: the good news is that you found the hypothesized factors. The bad news is that something is different about your sample compared with previous analyses. Find out what it is. You may be able to add valuable information to your field.

  • Your program indicates eight factors, but you think you really may have fewer factors in your data.

Solution:  force your factor analysis to show you other solutions, say from two to seven factors. See whether having fewer factors improves the interpretability of your results. Pursue the solution that gives you the most conclusive results based on theory.

  • You have a factor containing only variables with factor loadings of |0.3| or less.

Solution: you may have too many factors. Force the program to reduce the number of factors and rerun the analysis. When you understand your data better, drop the variables with factor loadings below your cut-off.

  • You have a factor with only two variables.

Solution: you may have too many factors. Force your program to reduce the number of factors and check whether your variables get incorporated into another factor. On the other hand, if they still stay with their previous factor, this factor may be very stable, and you may want to keep it separate. Of course, your results need to remain interpretable.

Reading material:

Finally, I would like to suggest some reading material. I like the factor analysis booklets from Sage’s Quantitative Applications in the Social Sciences. They are extensive and detailed, yet allow a novice to start at a beginner’s level and work her way up.

Kim, Jae-On and Mueller, Charles W (1978) Introduction to factor analysis. Series: Quantitative Applications in the Social Sciences. Sage Publications: Beverly Hills, CA.

Kim, Jae-On and Mueller, Charles W (1978) Factor analysis. Statistical methods and practical issues. Series: Quantitative Applications in the Social Sciences. Sage Publications: Beverly Hills, CA.


 

Principal Component Analysis
Summarize common variation in many variables... into just a few. Learn the 5 steps to conduct a Principal Component Analysis and the ways it differs from Factor Analysis.

Reader Interactions

Comments


Leave a Reply

Your email address will not be published. Required fields are marked *

Please note that, due to the large number of comments submitted, any questions on problems related to a personal study/project will not be answered. We suggest joining Statistically Speaking, where you have access to a private forum and more resources 24/7.