Partial correlations: What are the uses of them?

[sampier85]

They never seem to get reported in psychology research papers? I read some descriptions of them and they seem so useful in terms of statistical control.

So Friends, If they are not used, are there any alternative approaches for statistical control?

I would love you if you could list them and provide me with more info...

Regards,
Sam

 

[spunky]

So Friends, If they are not used, are there any alternative approaches for statistical control

multiple regression coefficients can usually be interpreted for that.

partial (and semi partial) correlations are useful as an intermediate step to other things. you can obtain an R-squared in regression through the appropriate linear combination of the semipartial correlations among predictors, for instance. there are also ways in which you can obtain the eigenvalues from canonical correlations through the use of partial and semipartial correlations. i mean, dont get me wrong, they are pretty useful but (at least in my opinion) they are somewhat limited in what they can do so there are more appropriate methods (like regression) on which people can work with adding statistical controls to their variables.