Correlation with clustered standard errors

[Annabell B]

Hi everybody, I am using Stata SE 14.0. I would like to run a correlation between variables, but adjust for groups (my sample consists of students from different schools, hence I would like to adjust for "school type" when running the Pearson's correlation). For the then following multiple regression I use the cluster (school) command, but I am clueless on how to get the same information (Pearson's coefficient & p-values) for correlations? I need to report them for the journal submission, and it will look inconsistent to report un-clustered correlations, as they are very different from the regression results. Any help would me much appreciated! Many thanks in advance!

[Carlo Lazzaro]

Annabell (please note the preference for full names on this forum. Thanks),
can't you rely on -estat vce, corr- after -regress-?
And, what's the contribution of a Pearson's correlation if you have already performed a regression?