r2mlm

r2mlm computes total- and level-specific R-squared measures for multilevel models to quantify variance explained across hierarchical data structures.

Key Features:

• Total- and level-specific R-squared: Calculates both total R-squared and level-specific R-squared values for multilevel models.
• Variance decomposition: Distinguishes each source of explained variance corresponding to different types of outcome variances.
• Rights and Sterba (2019) framework: Implements the variance decomposition framework of Rights and Sterba (2019) for multilevel R-squared calculation.
• Random effects support: Handles multilevel models that include random intercepts and/or random slopes.
• Computation automation: Automates the computations required to obtain the R-squared measures under the specified framework.
• Graphical visualizations: Produces graphics that accompany R-squared measures to facilitate comparison and interpretation of multilevel model effects.

Scientific Applications:

• Hierarchical data analysis: Quantifies explained variance in hierarchical datasets commonly encountered in the social and behavioral sciences.
• Model comparison: Compares multilevel models by partitioning explained variance across levels and variance components.
• Interpretation of random effects: Provides level-specific insights into variance attributable to random intercepts and slopes.

Methodology:

Implements the Rights and Sterba (2019) variance decomposition framework for multilevel models with random intercepts and/or slopes to compute total- and level-specific R-squared measures.