I am running ANCOVA on SPSS using post-operation score (continuous) as Dependent variable, pre-operation score (continuous) as Covariate, and 2 groups (categorial 1&0) as Fixed factors. This is the usual scenario. But can I add another 2 variables one is categorial (gender) and one continuous (age) to the Fixed Factors all at same time? My idea is to check which one of these factors (groups, gender, age) has significant effect and higher partial η2.

  • $\begingroup$ Is your question about how to do this in SPSS or is it about whether this is permissible, statistically? If the former it should be asked on an SPSS site. $\endgroup$ – Peter Flom - Reinstate Monica Dec 27 '13 at 15:52
  • $\begingroup$ I know how to do it on SPSS, my question is this permissible statistically $\endgroup$ – Emaddin Kidher Dec 27 '13 at 16:05

Given your comment, yes, this is fine to do, provided the other assumptions are met. You can have any number of independent variables in a linear model and they can be continuous or categorical or any combination.

Of course, you have to have sufficient N to avoid overfitting; you have to check for collinearity, etc, but there is no inherent limit.

  • $\begingroup$ Thank you very much Peter, really appreciated. What I like about your answers is also how straight to the point and short without going into complex details thank you again $\endgroup$ – Emaddin Kidher Dec 27 '13 at 16:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.