Advanced Structural Equation Modelling using Mplus

*****************************************

Note re laptops:

This course will take place mianly in a computer lab with Mplus installed, so there is no need to bring a laptop. On the final day this course will take place in a room conducive to laptops, so any course participant with Mplus installed on their laptop may wish to bring it along, however this is not crucial.

*****************************************

This course is designed as an advanced course in Structural Equation Modeling (SEM) for existing Mplus users or for existing AMOS and LISREL users who wish to convert to Mplus. Introductory courses typically cover path analysis amongst observed variables, confirmatory factor analysis, and full SEM models with latent variables. Furthermore, introductory courses usually deal only with continuous variables. However, many research questions involve independent and/or dependent non-continuous variables including ordered categorical (ordinal) variables; unordered categorical (nominal) variables that may be binary (Yes/No, Male/Female, Patient survives or dies, etc.) or polynomial; and count variables. This course covers a large number of complex models that are not covered in an introductory course and covers the uses of both continuous and non-continuous variables in such models. Detailed notes with worked examples and references will be provided as a basis for both the lecture and hands-on computing aspect of the course. The course is divided into six parts.

Part 1: Introduction to Mplus through revision of basic models. This part of the course introduces the Mplus programming language by revising basic models and also introduces the treatment of non-continuous data. Topics include linear regression with continuous dependent variables; Probit, Logistic, and Multinomial logistic regression for categorical dependent variables; Poisson regression for count dependent variables; Path analysis with continuous and/or categorical dependent variables; CFA with continuous and/or categorical factor indicators; Second-order factor analysis; and full Structural Equation Models.

Part II: Advanced single-level models. Topics include the testing of model and parameter invariance across groups (multi-group analysis); analysis of interactions with categorical moderator variables or continuous variables; non-linear modeling; and mean structure analysis approaches to the Analysis of Covariance.

Part III: Latent Growth-Curve Modeling (Repeated Measure designs). Topics include Linear Growth Modeling (LGM) for a continuous outcome; LGM for a categorical outcome; Quadratic growth modeling; and LGM for a continuous outcome with time invariant and time-varying covariates.

Part IV: Mixture Modeling (including Latent Class Analysis). Cross-sectional models include Mixture regression analysis; Latent Class Analysis (LCA) with binary, ordinal, nominal or continuous latent class indicators; CFA mixture modeling; and Structural equation mixture modeling. Longitudinal models include Growth Mixture Modeling (GMM) for a continuous or categorical outcome; GMM with known classes (multi-group analysis); and Latent Class Growth Analysis (LCGA) for a binary, ordinal or count outcome.

Part V: Introduction to Multi-level Models. This part of the course gives a very basic introduction to multi-level modeling including two-level regression analysis; two-level CFA; two-level SEM; two-level growth models; and multi-level mixture modeling.

Part VI: Personal Research. Finally, the course provides an opportunity for participants to work on their own research problems with the instructor’s assistance. Therefore participants are encouraged to bring a data set and/or research problem with them.