I'm trying to code a clmm model in R to analyze 1-5 Likert data. Respondents in different cities were asked to take a survey twice: once before a treatment, then again afterwards (Treatment). The model aims to assess whether and how the treatment interacts with demographic factors, namely respondent age (Age) and socioeconomic status (SES). Thus, I believe I need a random effects structure that will accommodate both the nested nature of subjects (City), and the repeated measures (Subject). I'm thinking my model should look something along the lines of: clmm(Likert~Treatment*Age+Treatment*SES+(1|City/Subject), but I'm largely self-taught when it comes to mixed effects models and often struggle with defining random effects. Thanks!


1 Answer 1


The proposed model is:

clmm(Likert ~ Treatment*Age + Treatment*SES + (1|City/Subject)

This model will fit fixed effects for:

  • Treatment
  • Age
  • SES
  • the two way interaction between Treatment and Age
  • the two way interaction between Treatment and SES
  • random intercepts for City
  • random intercepts for Subject varying within City (ie Subject is nested within City.

Based on the description in the post, this seems like a perfectly reasonable approach.

  • $\begingroup$ Does this answer your question ? If so please consider marking it as the accepted answer. If not, please let us know why. Also, if you haven't already, please consider upvoting it. $\endgroup$ Jul 24, 2021 at 12:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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