Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

multicollinearity - correlation between PCA components

edited January 2021 in JASP & BayesFactor

Hi,

I ran into some issues with components after performing PCA with orthogonal rotation. The component correlation in JASP is reported as 0, but when I copy-paste the values and correlate them in another analysis (R, excel, another Jasp file), the correlation between component 1 & 2 is quite high, certianly not 0 (pictures below). Is this just the nature of my data or is there something wrong with the analysis? I'm a bit confused and don't know how to proceed so any inputs are appreciated.

Thank you!



Comments

  • A related question but now I can't edit - does JASP automatically center the variables for PCA? I was told that it does but can't find any evidence of it in the tutorials, so just wanted to make sure

Sign In or Register to comment.