Home  /  Resources & support  /  Training  /  Organizational training  /  Multilevel/mixed models using Stata

Organizational training

Multilevel/mixed models using Stata


This course introduces multilevel/mixed modeling for nested and longitudinal data and its implementation in Stata. Mixed models contain both fixed effects, analogous to regression coefficients, and random effects, effects that vary across clusters. Participants will learn how to use mixed models to answer research questions about the observation- and cluster-level data and how to disaggregate these effects. Introductory theory, estimation, model building, and diagnostics will be discussed and demonstrated through many examples.

Are you more interested in panel-data models with unobserved individual-level heterogeneity, endogenous variables, or lagged variables? Check out our Panel-data analysis using Stata course.

The course will be interactive, use real data, and offer ample opportunity for specific research questions and for working exercises to reinforce what is learned.

Request this course

Course topics

  • The nested data problem and potential solutions
  • The multilevel/mixed-effects model (MLMM) with mixed
    • Theory and intuition
    • Random-intercept models
    • Random-coefficient (random-slope) models
    • Estimation methods: maximum likelihood, restricted maximum likelihood, generalized least squares, and small-sample inference
    • Model comparison: Wald test, likelihood-ratio test, information criteria
    • Using residuals and diagnostic plots to check assumptions
  • Longitudinal data analysis
    • Exploring and visualizing longitudinal data with the xt suite of commands
    • Random-intercept models with xtreg
    • Growth curve models with mixed
    • Alternative covariance structures
  • More complex models
    • Three-plus level models
    • Crossed-effects models
    • Binary and count responses with the me suite of commands
    • Estimation via adaptive Gaussian quadrature
  • Report results from a multilevel modeling analysis


Basic knowledge of standard linear regression and a working knowledge of Stata.


This course is available in-person or virtually. In-person training courses generally run for eight hours per day and include morning and afternoon breaks and a lunch break. Virtual training courses are typically divided into three- to four-hour daily sessions. You can arrange a convenient schedule with your instructor.

Request this course