Sorry if this is a basic question, just can't seem to find the answer I'm looking for.
For this experiment, we collected participant sex info (between-subj. categorical IV), disgust sensitivity scores (continuous moderator), and participant ratings of male and female target faces (two-level repeated measures DV). Participants saw and rated both male and female faces in the same session.
I first ran a simple Repeated Measures ANOVA in SPSS, with participant sex as the between subj. factor and target sex as the within-subj. factor. Target Sex and the Target Sex X Participant Sex interaction were significant. Then I wanted to test if disgust sensitivity moderated any of the effects, so I ran a Mixed ANCOVA with disgust sensitivity included as a covariate. All effects dropped to non-significance. I'm having trouble interpreting these results. We have theoretical reason to believe that these effects should be moderated by disgust sensitivity.
Is there a better way to test for moderation with this design? Linear regression seems like the best bet, but how do I deal with a two-level within-subj (repeated measures) DV in linear regression? Is there a macro (e.g., PROCESS) that would be appropriate for this? Is multilevel modeling necessary?
Thank you
Hi Jp
1. Several tests are available in:
Landau, S. and Everitt, B. S.(2004). A Handbook of Statistical analyses using SPSS.
2.Howitt, D. and Cramer, D.(2008). Introduction to SPSS.
2. The attached documents may be helpful in developing these tests.
Regards,
Zuhair
Since you have repeated measures, you need to used a mixed effects model or an ANCOVA (with target gender coded as a repeated-measures factor).
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA