12.1 Introduction

12.1.1 What is a multilevel model?

Multilevel models (MLM) go by many names:

• General Linear Mixed Model
• Random Coefficients Model
• Hierarchical Linear Model

12.1.2 Two Faces of MLM

In traditional MLM we typically have a model for the means, and a model for the variance.

12.1.2.1 A Model for the Means

• Typicalled known as fixed effects
• Similar to the quantities tested in single-level models
• How the expected value of an outcome varies based on the values of predictors in our model

12.1.2.2 A Model for the Variances

• Typically known as random effects
• Similar to the assumptions used in single-level models
• Model for how residuals are distributed and vary across observations (persons, groups, and time)

12.1.3 Two-Level Longitudinal Data

Between-Person Variation

• Level 2 or inter-individual differences
• Time-invariant
• More of less than other people

Within-Person Variation

• Level 1 or intra-individual differences
• Time-varying
• Can only assess with longitudinal studies
• More or less than one’s average