The package 'r2glmm' in R calculated the "semi-partial R squared" of each fixed predictor in a mixed effects model. The package describes this as

"The semi-partial R squared statistic corresponds to a select subset of fixed predictors in a fitted model and measures the relative increase in association between the dependent and independent variables resulting from the inclusion of the specified subset."

I am wondering if this is synonymous with the squared semi-partial correlation for each predictor?

Note that the package calculates the Marginal R squared of the model as a whole. I wonder if it does this for the fixed predictors as well, which would explain the difference.


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.