One of my cases has a value of -14.07, putting it at the bottom of the distribution. But it would make more sense if its value were positive. I read that the sign of the factor is meaningless, does it also apply to the scores?


closed as unclear what you're asking by Michael Chernick, Sycorax, Jeremy Miles, kjetil b halvorsen, mdewey Nov 20 '18 at 13:42

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.


It would be helpful if you could share more about the details or your problem. What is the model you are fitting? How are you fitting the factor analysis? What software are you using?

However, my instinct is that the sign does not matter. The reason is because, when you do a factor analysis, you have some observed variables, and you are trying to explain the variation in those observed variables using some unobserved factors. Your data has units. But the underlying factor you estimate to explain your data has no meaningful units -- all that matters is the relative difference between one factor value an another, not the absolute value of the factor. In this sense, the values of the underlying factor that are positive or negative are essentially arbitrary.

I believe this post answers your question more thoroughly.

Does the sign of scores or of loadings in PCA or FA have a meaning? May I reverse the sign?


Not the answer you're looking for? Browse other questions tagged or ask your own question.