Advanced topics in Structural Equation Modelling using R

Date:

19/09/2016 - 20/09/2016

Organised by:

University of Manchester

Presenter:

Dr Daniel Oberski and Dr Alexandru Cernat

Level:

Advanced (specialised prior knowledge)

Contact:

Claire Spencer
Tel: 0161 275 1980
Email: claire.spencer@manchester.ac.uk

Map:

View in Google Maps  (M13 9PL)

Venue:

Cathie Marsh Institute
Humanities Bridgeford Street
University of Manchester

Description:

Structural Equation Modelling (SEM) is a framework for the formulation, analysis, and evaluation of statistical models that involve series of regression relationships, possibly among unobserved variables. SEM encompasses numerous techniques that are well-known in the social sciences, psychology, economics, epidemiology, and other fields as special cases, including factor analysis, Item Response Theory, multivariate linear or probit regression, latent growth curves, random/mixed effects models, and dynamic models for longitudinal (panel) data.

Owing to this generality, any researcher with an interest in applying these techniques can benefit from a more in-depth knowledge of SEM. This course aims to provide such knowledge to those who have already been introduced to SEM and would like to know more about its background, the way the models can be extended, and some general methodological problems that tend to come up when applying them.

In this course we will use R, mainly with the lavaan package.

This course consists of two parts. In the first part we will introduce some general issues and techniques that can be applied across all SEM models:

  • Multiple group analysis;
  • Categorical data modeling;
  • Dealing with missing data;
  • Local model fit evaluation and sensitivity analysis;
  • Sample size planning, power analysis, and simulation studies.

In the second part we will discuss some more advanced SEM models for longitudinal data to which these techniques can be applied:

  • Latent Growth modelling;
  • Crossed lagged models;
  • Quasi simplex models.

By the end of the course participants will:

  •  Be able to do (categorical) Confirmatory Factor Analysis in R;
  •  Be able to deal with missing data in a number of ways;
  •  Be able to evaluate local model fit and perform a SEM sensitivity analysis
  •  Be able to do power analysis in SEM;
  •  Be able to do longitudinal analysis in SEM in a number of ways.

A basic knowledge of Confirmatory Factor Analysis and/or Structural Equation Modelling and a basic understanding of R is advised.

Cost:

The fee per teaching day is:

• £30 per day for UK registered students
• £60 per day for staff at UK academic institutions, UK Research Councils researchers, UK public sector staff and staff at UK registered charity organisations and recognised UK research institutions.
• £220 per day for all other participants.

All fees include event materials, lunch, morning and afternoon tea. They do not include travel and accommodation costs.

Website and registration:

Region:

North West

Keywords:

Longitudinal Data Analysis, Quantitative Software, R, Structural Equation Modelling , Missing Data , Simulation

Related publications and presentations:

Longitudinal Data Analysis
Quantitative Software
R

Back to archive...