I'm struggling to find a way to analyse my data. The design is:

  • Two response variables: liking, perceived threat (of immigrant groups)

  • random factor: target group (economic migrants, refugees)

  • fixed factors: continuous variable A*continuous variable B, quadratic terms of A & B


Thus, participants indicate their liking and perceived threat towards both economic migrants and refugees. My aim is to test whether the fixed factors predict the response variables across the target groups, while accounting for participants' two responses being nested in participants. I have so far analysed this data in lmer for a single response variable, using this code:

m1 <- lmer (liking ~ A*B + A² + B² + (1|subject) + (1|target group)

Please correct me if this is wrong and if it doesn't account for participants' responses for both target groups being nested in them, I'm very new to mixed models.


My question now is: how can I include a second response variable (i.e., perceived threat)? I did have a look around and saw that mcglm can handle multivariate designs, but I struggled to convert the lmer code into a mcglm code. Could someone help me with that perhaps? Also, I think mcglm would return a Chi-square test - does anyone know if it's possible for mcglm to return a multivariate F-test?

Any help would be very much appreciated!


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.