Multilevel Models for Cross Sectional and Longitudinal Data: Summer 2015 (ICPSR)


Upcoming Workshops:

Thank you for visiting my course notes. Here are some upcoming opportunities to learn from me and my colleagues in person:

WorkshopDatesLocationInstructor
Diagnostic MeasurementMay 22-25, 2017Omni Hilton Head Oceanfront Resort, Hilton Head Island, South CarolinaJonathan Templin
Introduction to Longitudinal Multilevel ModelsMay 30-June 2, 2017Omni Hilton Head Oceanfront Resort, Hilton Head Island, South CarolinaLesa Hoffman

Multilevel Models for Cross Sectional and Longitudinal Data

ICPSR Summer Workshop at the University of Colorado Boulder
July 13, 2015 – July 17, 2015

Presented by:

Dr. Jonathan Templin
Department of Educational Psychology
University of Kansas
email: jtemplin@ku.edu
http://jonathantemplin.com

Materials created by Dr. Lesa Hoffman
Associate Professor of Quantitative Methods, Child Language Program,
University of Kansas

Course Textbook:
Hoffman, L. (2014). Longitudinal Analysis: Modeling Within-Person Fluctuation and Change. Taylor and Francis
http://www.pilesofvariance.com/

Course Overview

Multilevel models are known by many synonyms (i.e., hierarchical linear models, general linear mixed models). The defining feature of these models is their capacity to provide quantification and prediction of random variance due to multiple sampling dimensions (across occasions, persons, or groups). Multilevel models offer many advantages for analyzing longitudinal data, such as flexible strategies for modeling change and individual differences in change, the examination of time-invariant or time-varying predictor effects, and the use of all available complete observations. This workshop will serve as an applied introduction to multilevel models for longitudinal data, including studies of individual change (i.e., growth curve models), individual fluctuation (i.e., daily diary designs), and multiple dimensions of within-person time (i.e., measurement burst designs).

The primary software package utilized for instruction will be SAS, but some reference examples using SPSS and STATA may also be provided. The course will also include daily opportunities for hands-on practice and individual consultation. Participants should be familiar with the general linear model (e.g., ANOVA and regression), but no prior experience with multilevel models or knowledge of advanced mathematics (e.g., matrix algebra) is assumed.

Overall Course Files
Syllabus:Syllabus
All Activities Files ZIPActivities
All Examples SAS SyntaxSAS Examples
All Examples SPSS SyntaxSPSS Examples
All Examples Stata SyntaxStata Examples
Schedule of Topics
TimeReading(s)Presentation Document(s)
Monday, July 13th
9:00-10:15
Ch. 1 & 3Lecture 1: Introduction to Multilevel Models
10:30-11:30Example 1: General Linear Models and Repeated Measures ANOVA
11:30-1:15Individual Lab Time and Lunch Time
1:15-2:30Ch. 4Lecture 2: Describing Within-Person Fluctuation over Time
2:45-4:00Example 2: Alternative Covariance Structure Models for Fluctuation

Excel File
4:00-5:00Lab Time: Work with your own data or do Activity 1 (Getting Data Ready for Longitudinal Models)
Tuesday, July 14th
9:00-10:15 Ch. 5 & 6Lecture 3: Describing Within-Person Change over Time
10:30-11:30Example 3: Random Effects Models for Change

Excel File
11:30-1:15Individual Lab Time and Lunch Time
1:15-2:30Lecture 3, continued
2:45-4:00Example 3, continued
4:00-5:00Lab Time: Work with your own data or do Activity 2 (Fitting Unconditional Longitudinal Models)
Wednesday, July 15th
9:00-10:15 Ch. 7Lecture 4: Time-Invariant Predictors in Longitudinal Models
10:30-11:30Example 4: Time-Invariant Predictors of Change

Excel File
11:30-1:15Individual Lab Time and Lunch Time
1:15-2:30Ch. 8Lecture 5: Time-Varying Predictors for Fluctuation
2:45-4:00Example 5: Time-Varying Predictors for Fluctuation

Excel File
4:00-5:00Lab Time: Work with your own data or do Activity 3 (Fitting Conditional Models with Level-2 Predictors)
Thursday, July 16th
9:00-10:15Lecture 6: Multilevel Models for Clustered Data
10:30-11:30Example 6: Two-Level Clustered Data – Students within Schools

Excel File
11:30-1:15Individual Lab Time and Lunch Time
1:15-2:30Lecture 6 and Example 6, continued
2:45-4:00Ch. 9Lecture 7: Multivariate Models for Change
4:00-5:00Lab Time: Work with your own data or do Activity 4 (Fitting Conditional Models with Level-1 Predictors)
Friday, July 17th
9:00-10:15Example 7: Multivariate Models for Change
10:30-11:30Ch. 11Lecture 8: Three-Level Longitudinal Models
11:30-1:15Individual Lab Time and Lunch Time
1:15-2:30Example 8: Three-Level Longitudinal Models
2:45-5:00Individual Lab Time