Adjusting for multiple proportion comparisons

[Patrick Abi Nader]

Hi All,

I am using Stata 14.
I have to do 21 different proportion tests to compare a binary variable with a 7 level categorical variable. I want to do pairwise comparisons for each of the 7 levels using multproc a user written command (by Roger Newton). Is there an efficient way to extract the p-value from each proportion test and put it into a new dataset? I would prefer not to manually run each test, write down the p value, run the next test, etc, and create a new file.

Thank you

[ben earnhart]

from the help file:


multproc and smileplot save the following results in r():

Scalars
r(puncor) uncorrected critical P-value
r(pcor) corrected critical P-value
r(npvalues) number of P-values
r(nreject) number of P-values rejected

Macros
r(method) the method() option

[Patrick Abi Nader]

Hi Ben,

Thank you for your answer.

I am trying to create the dataset that multproc would use to create the variables you mentioned.

From the help file:
multproc takes as input a dataset with one value for each of a set of multiple statistical tests of multiple null hypotheses -- including a variable containing p-values for these tests and an uncorrected overall critical p-value specified by the user -- and carries out a multiple-test procedure.

Patrick