A general and simple method for obtaining R2 from generalized linear mixed‐effects models

By Shinichi Nakagawa, H. Schielzeth. 2013

9,998 citations recorded by the source corpus.

Subfields: Ecology, Evolution, Behavior and Systematics, Nature and Landscape Conservation, Statistics and Probability, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology, Evolution, Behavior and Systematics

In Nature and Landscape Conservation

In Statistics and Probability

In Evaluation & Benchmarks

Citations in the corpus

Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change (2016, positive): “Marginal R (Nakagawa and Schielzeth, 2013).”
Quantifying physiological influences on otolith microchemistry (2015, methodology): “Akaike information criterion corrected for small sample sizes (AICc), r2 values (Nakagawa & Schielzeth 2013) and ‘dredge’ (MuMIn package) terms were used for model selection, as detailed in Appendix S3.”
A statistical framework to explore ontogenetic growth variation among individuals and populations: a marine fish example (2015, methodology): “The amount of variance explained by models was assessed using two R2 metrics (calculated on models fit with REML) defined for mixed-effects models by Nakagawa and Schielzeth (2013).”
Extension of Nakagawa & Schielzeth's R2GLMM to random slopes models (2014, methodology): “R2 has been extended to apply to generalized linear models (GLMs) (Maddala 1983) and linear mixed effects models (LMMs) (Snijders & Bosker 1994) [reviewed by (Nakagawa & Schielzeth 2013)].”
Artificial night lighting rather than traffic noise affects the daily timing of dawn and dusk singing in common European songbirds (2014, methodology): “We report R2 to assess the predictive quality of the models, following Nakagawa and Schielzeth (2013) (Supplementary Table S1).”
Formidable females and the power trajectories of socially integrated male vervet monkeys (2017, methodology): “We ran the analyses in R, using in the ‘ lme4 ’ package (Bates, Maechler, Bolker, & Walker, 2015) to model outcomes and the ‘ MuMIn ’ package (Barton & Barton, 2015) to generate marginal and conditional R 2 values for the ﬁ xed effects and whole models, respectively (Nakagawa & Schielzeth, 2012).”
An R2 statistic for fixed effects in the generalized linear mixed model (2017, methodology): “However, Nakagawa and Schielzeth [25] point out that R2 statistics corresponding to individual levels in a model may decrease with the addition of a fixed predictor in larger models.”
Landscape‐level crop diversity benefits biological pest control (2018, methodology): “We calculated marginal (fixed effects) and conditional (random and fixed effects) coefficients of determination (R2) to assess overall model fit (Tables S3–S5; function “r.squaredGLMM”; Nakagawa & Schielzeth, 2013).”
Tree diversity is key for promoting the diversity and abundance of forest‐associated taxa in Europe (2020, methodology): “For each model, we calculated the proportion of the deviance explained by the fixed factors only (marginal R²) or by the random and fixed effects (conditional R²), according to Nakagawa and Schielzeth (2013) (piecewiseSEM package; Lefcheck 2016).”