Variance partitioning/ R-squared computation in multilevel (or mixed effects) models:

The r2mlm R package (Shaw, Rights, Sterba, & Flake, 2020), which is available on CRAN, contains numerous functions to compute R-squared for various types of multilevel models. See details here:


Model assessment for latent variable models:

npmmApproximation: This R function outputs implied fixed effects, random effect (co)variances, and residual variance components for a given nonparametric multilevel regression mixture model. (R code; for further details, see supplemental materials of Rights & Sterba [2016] The relationship between multilevel models and nonparametric multilevel mixture models: Discrete approximation of intraclass correlation, random coefficient distributions, and residual heteroscedasticity)

modelavgIRT: R function that reads in item response theory model person scores and their associated standard errors from each of a set of candidate models and outputs model-averaged person scores and standard errors. (R code; for further details, see Rights, Sterba, Cho, & Preacher [2018] Addressing model uncertainty in item response theory person scores through model averaging)

regMixR2: This R function computes and outputs R-squared measures and analytic decompositions of variance for single-level and multilevel regression mixture models. (R code; for further details, see supplemental materials of Rights & Sterba [2018] A framework of R-squared measures for single-level and multilevel regression mixture models)


Accounting for parcel allocation variability in structural equation models:

PAVranking: This R function quantifies and assesses the consequences of parcel-allocation variability for model ranking of structural equation models. Available in the semTools package on CRAN (

poolMAlloc: This R function employs an iterative algorithm to choose the number of random item-to-parcel allocations needed to meet user-defined stability criteria for a fitted structural equation model. Available in the semTools package on CRAN (