0
$\begingroup$

I have three different groups. I did a PCA and plotted PCA1 and PCA2 as a scatter. One of the groups looks different to the others. However, afterwards I realised that I applied PCA to each group separately, when it should be applied to all at the same time. If I apply to all at the same time, the groups look the same in the PCA plot - i.e. the blue is aligned with the red and green.

Is there any possible value in applying PCA separately and the difference found here, or is this simply a random effect of incorrect application of PCA?

$\endgroup$

1 Answer 1

1
$\begingroup$

It depends on your question : do you want to find the population's factorial structure (assuming all groups came from the same population) or compare the groups' factorial structure. Doing PCA, a exploratory analysis, could be more related to the first option, so you should pool all groups in a single PCA. For the second option, you could use a single factor analysis including multigroup analysis, like multigroup PCA.

There is many options available to you. Your research question should guide you.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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