multilevel-modeling

What problem does it solve?

This Skill helps you analyze nested or repeated-measures data by estimating within-person and between-person variance using multilevel/mixed-effects models, including ICC to decide whether clustering matters.

Core Features & Use Cases

• Random intercepts and slopes for hierarchical data: Fit models where units (e.g., observations) are nested within persons or groups, and allow predictors to have person-specific effects.
• ICC and variance decomposition: Compute intraclass correlation to quantify how much outcome variability is attributable to between-group differences.
• ESM-friendly within-person centering: Separate within-person deviations from between-person differences for experience sampling or diary designs, reducing confounding.
• Model comparison and inference workflows: Perform likelihood ratio tests, use Satterthwaite-style df via lmerTest, and support common contrast workflows through emmeans.
• Three-level extensions: Handle structures like observations within days within persons for intensive longitudinal data.

Quick Start

Use the multilevel-modeling skill to fit a random-intercept model, compute ICC, apply within-person centering for ESM predictors, and compare candidate random-effect structures for your outcome.