I have a data set that contain 75 variables of football players . These 75 variables basically measures two different types of information. 30 of those variables related to bio metric information and other 45 variables related to individual performing abilities.

I want to identify the correlation between the bio-metric information and individual performing abilities.

My approach for this is ,

first do the principle component analysis separately for the variables that measure bio metric information and the variables that measure individual performing ability. And later use those principle components to calculate the correlation.

I want to know whether my approach is correct or not. Are there any statistical methods that are more suitable to this situation ?

Thank you .

  • 1
    $\begingroup$ Canonical correlation analysis can be used to identify association between groups of variables. $\endgroup$ – AlexK May 15 at 23:36
  • $\begingroup$ @AlexK Thank you very much. That is the best method that can apply here. Also do you think that the method i proposed using principle component analysis also useful in this situation ? $\endgroup$ – student_R123 May 16 at 2:25
  • 1
    $\begingroup$ Why reduce your original variables to some new variables that have no interpretation when you can just use the original variables? And were you planning to just get one PC per group? What if they explain very little of the variation in the original variables? I don't see how this can be useful. $\endgroup$ – AlexK May 16 at 5:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.