Tools for Dyadic Multilevel Models R package dyadMLM version 0.1.0 Provides tools for dyadic multilevel modeling with linear and generalized linear mixed-effects models. It validates and prepares long-format cross-sectional and intensive longitudinal data, including ecological momentary assessment designs, for distinguishable and exchangeable dyads. It also supports datasets containing multiple observed dyad compositions. It constructs composition-aware, model-ready variables for Actor-Partner Interdependence Models (APIMs), Dyadic Score Models (DSMs), and Dyad-Individual Models (DIMs). Prepared data can be used with model engines such as 'glmmTMB' and 'brms' for Gaussian and non-Gaussian outcomes, including counts, proportions, and skewed continuous responses. Post-estimation tools compare compatible fitted models and back-transform exchangeable sum-and-difference random-effect covariance structures into member-level quantities. The APIM and DSM specifications and their relationships follow Iida et al. (2018) <doi:10.1177/0265407517725407>; the multilevel sum-and-difference random-effects implementation for exchangeable dyads adapts del Rosario and West (2025) <doi:10.1177/25152459251351286>. Software Depends: R (>= 4.1) Imports: dplyr (>= 1.1.0), pillar, rlang (>= 0.4.0), tidyselect, tibble (>= 3.0.0) Suggests: brms, glmmTMB, htmltools, knitr, lavaan, marginaleffects, rmarkdown, spelling, testthat (>= 3.1.7) Pascal Küng <kueng.pascal@gmail.com> Comprehensive R Archive Network (CRAN) Pascal Küng [aut, cre, cph] (ORCID: <https://orcid.org/0000-0001-7346-9414>) MIT + file LICENSE (https://CRAN.R-project.org/package=dyadMLM/LICENSE) 2026-07-30 application/tgz https://CRAN.R-project.org/package=dyadMLM doi:10.32614/CRAN.package.dyadMLM en-US