1
$\begingroup$

I am looking for a right correlation metric to measure correlation between two vectors. The dimensionality of the vectors is around 100. The first vector has real values as the elements and the second vector has 0,1 values.

Spearman rho is widely used as a correlation metric between two real-valued vectors, but my case is slightly different (0,1 values in one of the vectors). Whether Mann–Whitney U test is more appropriate in my case than Spearman Rho?

I appreciate your help.

Improve this question
$\endgroup$
1
  • 2
    $\begingroup$ If you are not limited Mann-Whitney and Spearman Rho, I would suggest you to consider boxplot / linear models as suggested by this post $\endgroup$
    – AshOfFire
    Commented Apr 23, 2018 at 8:59

1 Answer 1

Reset to default
1
$\begingroup$

First, Mann Whitney U is not a nonparametric alternative to correlation, it is an alternative to t-tests. It tests for differences between the two groups, not relationship.

Second, to your question, it looks like point-biserial correlation is what you want. As per Wikipedia:

The point biserial correlation coefficient (rpb) is a correlation coefficient used when one variable (e.g. Y) is dichotomous; Y can either be "naturally" dichotomous, like whether a coin lands heads or tails, or an artificially dichotomized variable. In most situations it is not advisable to dichotomize variables artificiall

Improve this answer
$\endgroup$

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