Adjusting non parametric correlation for covariates

#1
Hello!
I am dealing with the following problem:
I am testing if there is a correlation among a continuous IV and a group of 4 Continuous DV.
Since I tested the variables for normal distribution and the hypothesis was rejected I performed a spearman correlation. Though I would need to adjust this test for two variables, one is continuous and the other is categorical.
Is there a way in stata to perform it?

I also would need an advice about whether it is appropriate or not to perform the bonferroni correction, i have not done it yet because the group of DV are not necessarely linked among them, what do you think?

Thank you in advance!
 

Karabiner

TS Contributor
#3
Seemingly you want to perform ANCOVA or a multiple linear
regression analysis. There's no need for data to be normally
distributed, but if your sample size is small, then the residuals
from the analysis should be normal (or, more precisely:
should be from a normally distributed population).

With kind regards

K.
 
#4
Thank you for your reply!
Isn't ANCOVA appropriate when one IV (that is categorical) groups the DV ?
in my case both DV and IV are continuous and I would like to adjust the association for a categorical covariate.
At the moment I just explored the correlation between the two continuous variables using a Spearman (rather than a Pearson correlation).
Please correct me if I am wrong