1
$\begingroup$

I am computing the Spearman correlation between the ranks produced by different teachers for the same exam session. Now, I have seen in the case of Pearson correlation, that the squared correlation is reported in the end. Do I need to do the same for the Spearman correlation?

Thanks in advance

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

If you are using either correlation (Pearson's, in the case of continuous data, or Spearman's, in the case of ordinal data) to express the degree to which two variables covary, then you can just present the coefficient.

On the other hand, if you are interested in expressing how much variance in one variable another variable explains (or, because I think this is more common in the context of evaluating a statistical model, e.g. multiple linear regression), then you can present the squared coefficient, or the coefficient of determination.

Improve this answer
$\endgroup$
1
  • $\begingroup$ I've seen a lot of conflicting answers when it comes to use the coefficient of determination for Spearman's Rho. Could you please elaborate on why you think it is appropriate to use the coefficient of determination for Spearman's correlation? $\endgroup$
    – mribeirodantas
    Commented Dec 11, 2021 at 19:52

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.