r2mlm
R-Squared Measures for Multilevel Models
Mairead Shaw, Jason Rights, Sonya Sterba, Jessica Flake
Last update: 20 December 2023
Software version identifier: 0.3.3, 0.1.0, 0.2.0, 0.3.0, 0.3.1, 0.3.2, 0.3.4, 0.3.5, 0.3.7
Copyright license: GNU General Public License, version 3.0
Generates both total- and level-specific R-squared measures from Rights and Sterba’s (2019) <doi:10.1037/met0000184> framework of R-squared measures for multilevel models with random intercepts and/or slopes, which is based on a complete decomposition of variance. Additionally generates graphical representations of these R-squared measures to allow visualizing and interpreting all measures in the framework together as an integrated set. This framework subsumes 10 previously-developed R-squared measures for multilevel models as special cases of 5 measures from the framework, and it also includes several newly-developed measures. Measures in the framework can be used to compute R-squared differences when comparing multilevel models (following procedures in Rights & Sterba (2020) <doi:10.1080/00273171.2019.1660605>). Bootstrapped confidence intervals can also be calculated. To use the confidence interval functionality, download bootmlm from <https://github.com/marklhc/bootmlm>.
