Structural Equation Modeling

A 5-Day Seminar Taught by Paul Allison, Ph.D.

Read reviews of this course


Structural Equation Modeling (SEM) is a statistical methodology that is widely used by researchers in the social, behavioral and educational sciences.  First introduced in the 1970s, SEM is a marriage of psychometrics and econometrics. On the psychometric side, SEM allows for latent variables with multiple indicators. On the econometric side, SEM allows for multiple equations, possibly with feedback loops. In today’s SEM software, the models are so general that they encompass most of the statistical methods that are currently used in the social and behavioral sciences.

This 5-day seminar assumes no previous knowledge of SEM, and covers almost a full semester’s worth of material. It is designed to make you a knowledgeable, effective and confident user of methods for structural equation modeling. Each day will include 1 to 2 hours of supervised, practical exercises that will help you achieve mastery of these methods.

Here Are a Few Things You Can Do With Structural Equation Modeling

  • Test complex causal theories with multiple pathways.
    • Estimate simultaneous equations with reciprocal effects.
    • Incorporate latent variables with multiple indicators.
    • Investigate mediation and moderation in a systematic way.
    • Handle missing data by maximum likelihood (better than
    multiple imputation).
    • Analyze longitudinal data.
    • Estimate fixed and random effects models in a comprehensive framework.
    • Adjust for measurement error in predictor variables.

How This Seminar Differs From Paul Allison’s 2-Day seminar “Introduction To Structural Equation Modeling”

This course includes all the material in the 2-day seminar, but in more detail, especially regarding models for categorical data. It also covers many other topics such as missing data, bootstrapping, formative indicators, interactions, and longitudinal data analysis. Lastly, much more time is devoted to exercises.


Computing

This is a hands-on course with lots of opportunities to practice SEM. The empirical examples and exercises in this course will emphasize Mplus, but equivalent code will also be presented for SAS, Stata and lavaan (a new package for R). Mplus is our preferred SEM package because of its superior capabilities for missing data, multi-level modeling, and ordinal and categorical data. To fully benefit from the course, you should bring your own laptop loaded with a recent version of SAS, Stata, Mplus or R (with the lavaan package installed). Whichever package you choose, you should already be familiar with basic data management operations and the commands/procedures for doing linear regression, logistic regression, etc.


Who should attend? 

This course is designed for researchers with a moderate statistical background who want to apply SEM methods in their own research projects. No previous background in SEM is necessary. But you should have a good working knowledge of basic principles of statistical inference (e.g., standard errors, hypothesis tests, confidence intervals), and should also have a good understanding of the basic theory and practice of linear regression. For the later parts of the course, it is also desirable to be familiar with logistic regression (binary, ordinal, or nominal). To do the hands-on exercises, it is essential that you already be comfortable working with one of the four packages that will be covered in the seminar.


Location and Materials 

The course meets 9 a.m. to 5 p.m. on Monday, July 11 through Friday, July 15 at The Gleacher Center, The University of Chicago Booth School of Business, 450 North Cityfront Plaza Drive, Chicago, IL 60611.  

Participants receive a bound manual containing detailed lecture notes (with equations and graphics), examples of computer printout, and many other useful features. This book frees participants from the distracting task of note taking. 


Registration and lodging

The fee of $1795.00 includes all seminar materials. 

Lodging Reservation Instructions

Guests room blocks have been reserved at the following nearby hotels.

Sheraton Grand Chicago, 301 E North Water St, Chicago, IL 60611. Call 888-627-7106 before 5 pm on June 10 and refer to Statistical Horizons for rates ranging from $229. Approximately 0.1 miles from Gleacher.

Embassy Suites, 511 N. Columbus Drive, Chicago, IL 60611. Call 800-525-2509 by June 10 and refer to Statistical Horizons for a rate of $199 when checking in on Sunday, July 10 or rate of $229 for later check-ins. Approximately 0.2 miles from Gleacher.

Club Quarters, 75 East Wacker Drive, Chicago, IL 60601. Email memberservices@clubquarters.com or call 203-905-2100 before June 9 and refer to the Statistical Horizons Group block SH0710 for the special rate of $192 per night. Approximately 0.4 miles from Gleacher.


Preliminary Seminar outline

1.Introduction to SEM
2. Path analysis
3. Direct and indirect effects
4. Bootstrapping in SEM
5. Nonrecursive models
6. Instrumental variables
7. Reliability and validity
8. Multiple indicators of latent variables
9. Exploratory factor analysis
10. Confirmatory factor analysis
11. Goodness of fit measures
12. Structural relations among latent variables
13. Formative indicators
14. Alternative estimation methods.
15. Missing data
16. Multiple group analysis
17. Interactions with latent variables
18. Models for ordinal and nominal data
19. Models for censored and event-time data
20. Models for longitudinal data


Comments on the 2-day version of this seminar

“This course provided a richly detailed and in-depth introduction to SEM. The course worked through examples in very good detail, and Dr. Allison was very adept at answering questions and providing an exceptionally detailed overview of SEM procedures. Be prepared to be challenged and a bit exhausted but also galvanized to use these procedures with your own data.”
  Jonathan Mattanah, Towson University

“The course is structured well, it’s a good pace for someone with no SEM experience or background, and the information is presented in a very accessible way. I liked the multiple examples with specific data to illustrate more complex ideas. The additional resources and references mentioned during the course will help me to dig into the SEM detail particularly relevant to my work on my own.”
  Lilia Bliznashka, International Food Policy Research Institute 

“This course was accessible for me with limited experience in SEM and no prior experience in MPlus. The pace was excellent with lots of breaks to digest the information. Also very helpful book and codes to make sure we can bring what we learned here in class back home and implement it.”
  Carmen Logie, University of Toronto 

“This course provided a good overview of SEM. I left it feeling more confident than when I arrived.”
  Michael Miner, University of Minnesota 

“Dr. Allison is very knowledgeable and knows how to convey even very complicated concepts such as SEM. He made it look very easy to me. I got many of my questions answered. Thank you.”
  Maryam Ghobadzadeh, University of Minnesota

“Thorough introduction to SEM and lots of practical examples. Helps if you have familiarity with software (MPlus especially), but not necessary. Dr. Allison is very good at explaining concepts in a few different ways to drive home concepts.”
  Valerie Darcey, Georgetown University 

“A very nice overview of SEM. But the course also gave me additional understandings of regression, measurement error and other important topics. I feel more comfortable with methods in general now than I felt before this course.”
  Johan Westerman, Stockholm University

“This course was excellent. Very clear instruction, explanations,and materials. Definitely gives one the background and tools needed to begin exploring SEM.”
  Sydney Martinez, University of Oklahoma Health Sciences Center