Distinguishable and Exchangeable Dyads: Bayesian Multilevel Modelling
Cross-Sectional and Intensive Longitudinal APIM and DIM
Abstract
This tutorial presents applied Bayesian multilevel modelling for dyadic data,
covering both distinguishable and exchangeable dyads. It develops cross-sectional
and intensive longitudinal variants of the Actor-Partner Interdependence Model
(APIM) and the Dyad-Individual Model (DIM), demonstrates their algebraic
equivalence, and discusses implications for interpretation. Using simulated
data, it provides fully reproducible code to rebuild the models and assess
parameter recovery. The tutorial offers comprehensive brms
implementations, including random-effects structures, heterogeneous residual
variances, compound-symmetry residuals, and optional AR(1) processes.
Quick Guides
The companion PDFs provide direct access to the model code, assumptions, and reporting tables without scrolling through the full tutorial: distinguishable dyads and exchangeable dyads.
Suggested Citation
Küng, P. M. (2026). Distinguishable and Exchangeable Dyads: Bayesian Multilevel Modelling (Version v2.0.9). University of Zurich. https://doi.org/10.5281/zenodo.17400655.
Keywords
dyadic data; APIM; DIM; Bayesian multilevel modelling; brms; cmdstanr; longitudinal diary data; compound symmetry; leave-one-out cross-validation