I feel like this is something I have seen a million times and yet it is totally slipping my mind. Suppose a series of scores has a strong negative correlation with some particular personality trait. Is there a way that I can adjust the scores in SPSS while controlling for the trait question?
Hi John, in SPSS, you can do an ANOVA. In the options menu, you can select the descriptives and the estimates. The descriptives show the unadjusted means while the estimates table in the shows adjusted means taking into account the information from the covariate (in your case, the personality trait).
I'm not aware of this option in the regression menu in SPSS.
I think the equivalent to Christian's answer in SPSS regression would be to simply fit the equivalent ANCOVA model as a regression. Best wishes, David
If you are just controlling, the ANCOVA is a solution. Alternatively, do a GLM or regression using Type I SS (and the nuisance variable first).
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS