I sampled women and selected 30 high conformity women and 30 low conformity, matched for age, weight and activity level amongst other things. I split each group into half, with half on an exercise routine rated five stars by 75% of past users, and the other half rated five stars by 25% of people. I measured how well they rated their plan on seven days and took the sum, and generally speaking, the low conformity women rated the bad routine slightly lower than the good routine, whereas the high conformity women rated the bad routine slightly higher than their counterparts, but rated the good routine MUCH higher.

I kept conformity as a random factor, because I used an externally validated scale, but considered my routine fixed. I ran an ANOVA but my main effects weren't significant. My interaction was, though. I can't run post-hoc tests within SPSS because I don't have enough levels. What is the best way to further explore the interaction?

(I took the sum rather than average because this is a replication of an experiment already performed, where the sum of points scored in a game were used as the outcome variable.)

Cheers for your time.


Your Answer

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

Browse other questions tagged or ask your own question.