I'm running multivariate mixed effects linear meta-analytic models in R. One of my predictor variables is categorical, but each observation generally has multiple categorical values. I'm not sure how to best incorporate this variable as a fixed effect in the model.

To make it more concrete, the variable is "activity type" and the data look something like this:

Observation A: recreation, hunting

Observation B: hunting

Observation C: recreation, roads

... etc. There are 7 categories in total across >150 observations.

I could include each activity type as its own binary variable (perhaps w/ interaction terms to account for multiple activities), but I would like some way to assess the overall effect of "activity type" if possible. Is there a way to have a matrix as a predictor variable?

All suggestions welcome!


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.