So I have a data set where I have 5 repeated measures of a variable, but also need to check for interactions with gender (categorical), and a continuous variable (let's say minutes of something). When I try to define the model through the repeated measures option on SPSS and then build the model, I can build the 2-way interactions, but can't build the 3-way interaction for all 5 repeated measures X gender X minutes. I was told that I could do it using this function within GLM but maybe I'm doing something wrong? I figured I'd ask if anyone has run into that issue before. Just for a little more information, my dataset is setup as repeated measures (so not long format). Do I have to convert the data to long format even though the only categorical variable is gender? Not sure if that's enough information, but help?
Hello Ricardo A. Wilhelm. IIRC, GLM forces all interactions involving within-subjects factors to be in the model. If you restructure the data to LONG format and use MIXED rather than GLM, you will have complete control over which terms are included in the model. See these UCLA pages if you need examples of repeated measures analysis via MIXED. HTH.
- https://stats.idre.ucla.edu/spss/seminars/repeated-measures/
- https://stats.idre.ucla.edu/spss/seminars/spss-mixed-command/
- Badges
- Science topic