Hi,

fist of all I was wondering whether pwcorr is okay to use when I would like to calculate the correlation between two dummy variables? Further I am not quite sure whether my type of interpreting the results is correct. Suppose we have two dummy variables. one is age, which takes on the values 1 = Female 0 = Male and the variable Riddle with 1 = Solves riddle and 0 = Does not solve riddle. Suppose I run pwcorr and got a significant (on the 5% level) correlation coefficient of -0.15. Do I have to interpret the result as a probability? In other words: The probability that a women solves the riddle is 15% lower than the average probability of men solving the riddle?

Thanks!