The distribution of my residuals were slightly negatively skewed, and when I mirrored and transformed my outcome, the histogram still showed a slight negative skew. Can this be normal, or did I overlook a step in my transformation process?
Thank you for all the lightning fast responses!
1 comment:
I would guess either (a) that the transformation was performed on the original variable, not the mirrored (or reflected) variable or (b) the residuals analysis was performed with the original residuals, not the ones computed from the regression conducted with the transformed variable.
So, if your variable were selfworth, you would first reflect (or mirror or reverse score) the variable by using
transform --> compute --> selfworthrev = 5 - selfwrth
then transform the skew by using
transform --> compute --> selfworthrevsqrt = sqrt(selfwrthrev)
Then re-run the analysis, asking to save the unstandardized residuals and predicted values, then re-run your tests of assumptions.
Post a Comment