R2.merMod: merMod method for R2

In multilevelTools: Multilevel and Mixed Effects Model Diagnostics and Effect Sizes

Description

For pseudo R2 by cluster, the squared correlation between observed and predicted values for each cluster unit is returned. For the overall model, the marginal and conditional R2 are calculated as described in the references.

Usage

## S3 method for class 'merMod'
R2(object, cluster = FALSE, ...)

Arguments

object	A model estimated by lmer.
cluster	A logical whether to calculate individual pseudo R2 values by cluster unit (if TRUE) or the marginal and conditional R2 for the overall model (if FALSE, the default).
...	Added for consistency with generic. Not currently used.

Value

a named vector with the marginal and conditional R2 values, if CLUSTER = FALSE, otherwise, a data table with the pseudo R2 for each cluster unit.

References

For estimating the marginal and conditional R-squared values, see: Nakagawa, S. and Schielzeth, H. (2013) <doi:10.1111/j.2041-210x.2012.00261.x> "A general and simple method for obtaining R2 from generalized linear mixed-effects models" and also: Johnson, P. C. (2014) <doi:10.1111/2041-210X.12225> "Extension of Nakagawa & Schielzeth's R2GLMM to random slopes models"

Examples

library(JWileymisc)
data(aces_daily, package = "JWileymisc")
m1 <- lme4::lmer(PosAff ~ STRESS + (1 + STRESS | UserID),
  data = aces_daily)

R2(m1)
R2(m1, cluster = TRUE)

hist(R2(m1, cluster = TRUE)$R2)

rm(m1)

multilevelTools documentation built on March 13, 2020, 2:07 a.m.