Structural equation modeling offers vast flexibility for estimating many types of statistical models. Multilevel modeling allows treating data that reflect nested sources of variation. Both allow treating cross-sectional and longitudinal/panel data of many types. Mplus is an ideal software package for this purpose.

Participants will be able to use structural equation and multilevel models in a program of research with both cross-sectional and longitudinal/panel data. This will allow using existing data to test new theories and hypotheses. Researchers will also gain insight into research design to allow more comprehensive theorizing and hypothesis testing.

Program Overview & Requirements

This intensive week-long course will provide attendees with knowledge of many issues surrounding structural equation and multilevel modeling, including practical application in Mplus. Models incorporating both cross-sectional and longitudinal/panel data will be treated. Econometric methods for reducing endogeneity and heterogeneity in the form of models estimated with instrumental variables and longitudinal/panel data will be discussed.

Topics include:

• Probability and Estimation
• Path analysis, including model identification and fit, moderation & mediation, and instrumental variables
• Structural equation models, including latent variables, confirmatory factor analysis, multi-group models and invariance tests
• Multilevel models, including multilevel data, random intercepts, random slopes and multilevel SEM
• Longitudinal/Panel data, including static (econometric) models, dynamic (econometric) models, and latent growth models

Participants are encouraged to bring a laptop computer along to the course. All models will be estimated and discussed in class, and program code will be provided to participants, so a computer with Mplus is not required. However, it is recommended. For those without Mplus, a demo version allowing the estimation of most in-class models can be downloaded here: http://statmodel.com/demo.shtml

Who should attend?

This course is targeted at researchers and PhD students from social and health sciences with basic regression knowledge.

Presenter

Dr Michael Zyphur is a member of the Department of Management and Marketing at the University of Melbourne.

Program

Download a map of the University

To register

PhD student registrations are now closed. Limited places still available for full registrants at $1100 AUD (inclusive of GST). Registration includes morning and afternoon tea and lunch each day. For information and to register please contact Brooke Young, +61(0)3 8344 2166 or byoung@unimelb.edu.au