Install this application on your home screen for quick and easy access when you’re on the go.
Just tap then “Add to Home Screen”
Install this application on your home screen for quick and easy access when you’re on the go.
Just tap then “Add to Home Screen”
Monday 17 – Friday 21 February 2020, 09:00–12:30
15 hours over five days
The course gives an introduction to theory and practice of Structural Equation Modelling (SEM).
Why is it more beneficial to use SEM, compared to classical path models? This course shows how the theoretical latent constructs (e.g. social and political attitudes, and values) can be measured and explained with their relationships with other variables.
The course highlights the theoretically and empirically important aspects of model comparison between different groups, such as countries or different years in a longitudinal survey.
Please bring your laptop. PC/Windows and Mac are OK with R Studio – see under 'Software requirements', below.
Tasks for ECTS Credits
2 credits (pass/fail grade) Attend at least 90% of course hours, participate fully in in-class activities, and carry out the necessary reading and/or other work prior to, and after, class.
3 credits (to be graded) As above, plus complete one task (tbc).
4 credits (to be graded) As above, plus complete two tasks (tbc).
Julia Koltai is an assistant professor at the Faculty of Social Sciences, Eötvös Loránd University. She is also a research fellow at the Centre for Social Sciences, Hungarian Academy of Sciences. She gained her PhD in sociology in 2013.
Julia has led several domestic research programs and has taken part in international research projects and groups, including EU FP6-funded programs.
Her main scientific focus is on statistics and social research methodology, so her research has ranged widely, from minority research through political participation to social justice and integration.
In recent years, Julia's interest has turned to computational social science, especially network analysis and big data processing.
Structural Equation Modelling (SEM) is a powerful tool to analyse latent variable models common in social sciences, e.g. the analysis of social and political attitudes or social values.
SEM combines factor analysis and path analysis by simultaneously estimating relations between latent constructs and/or manifest variables, and also relations of latent constructs and their corresponding manifest indicators.
SEM allows the estimation and control for random and systematic measurement errors. Thus, SEM methodology allows an adequate modelling and empirical testing of measurement models and complex theoretical assumptions. SEM can also compare these models between different groups, like countries or waves of a longitudinal study, or also social groups.
The course introduces theory and practice of SEM on a general level. We will use R software.
Basic modelling techniques of SEM are explained and applied by exercises using free access social science data, though you can also use your own data for analyses. Daily assignments allow you to apply and transfer of SEM methodology to your own research interests.
We go through the statistical and methodological basics of SEM, such as regression analysis and classical path model analysis. We show the advantages of SEM and present its fitting principles. Basic (visual and statistical) notations and statistical tests and indices will be mentioned.
We focus on the first step of building a SEM model, namely confirmatory factor analysis (CFA) for the creation of latent constructs, in comparison with other factor analysis methods. We go through different model-building (parametrisation) and model improving techniques, and implement them during the lab session.
I expand on the latent variable model along with other, explanatory variables to get a more complex and interpretable model. These models are better for answering a scientific research question because they give more space for explanation. Again, you can apply these models in the lab, with emphasis on the interpretation and practical questions.
All about multiple group comparison, which is – taking the different levels of testing into account – one of the most useful parts of SEM. We begin with the theoretical problems of multi-group comparison and connect these problems with SEM tools that can help to decide the depth of the comparison. I will provide a step-by-step guide. In the lab, I'll give technical advice on the measurement and realisation of these models, especially on the interpretation of the different results.
A recap of what we've learned during the week, with the help of a concrete example in which we apply all the techniques. After going through a complex example, we will generalise the consequences and draw other conclusions. To conclude, I'll present some suggestions for publishing papers that include SEM methods.
You should understand basic principles of regression analysis and the meaning of regression results.
A basic understanding of principal component analysis (explorative factor analysis) would be helpful.
You should also have some familiarity with software R to manage data.
Day | Topic | Details |
---|---|---|
1.1 | Basic regression and path model analysis |
|
1.2 | Using and testing overidentified models |
|
2.1 | The measurement model: CFA |
|
2.2 | Creating a measurement model in the practice |
|
3.1 | The structural model |
- moving towards an explanatory model
- codes in R |
3.2 | Instructor supported self-working lab session |
From research question to results: using complex structural models |
4.1 | Theory of multiple group comparison (MGCFA) |
|
4.2 | Lab session about a multigroup comparison |
|
5.1 | Instructor supported self-working lab session about multigroup comparison |
Empirical analysis of a research question, which includes several latent concepts and compare more than one group |
5.2 | Summary |
|
Day | Readings |
---|---|
1 |
Rex B. Kline (2011) Chapter 2 Fundamental Concepts/Multiple Regression |
2 |
Timothy A. Brown (2006) Chapter 3 Introduction to CFA Rosseel, Y. (2012) |
3 |
Rex B. Kline (2011) Chapter 10 Structural Regression Models/from 'Analyzing SR Models' to 'Detailed Example' |
4 |
Timothy A. Brown (2006) Chapter 7 CFA with Equality Constraints, Multiple Groups, and Mean Structures/CFA in Multiple Groups Holger Steinmetz et al. (2009) |
5 |
No reading set for our final day |
R version 3.5.2 or higher
RStudio Desktop version 1.1.463 or higher
R packages
Please bring your laptop: PC/Windows and Mac are OK with R Studio using the abovementioned software version and packages.
You will need to user privileges to install R and R packages. If you have limited access – because, for example, it is a work laptop – speak to your IT department.
Summer School
Introduction to Inferential Statistics: What you need to know before you take regression
Multiple Regression Analysis: Estimation, Diagnostics, and Modelling
Summer School
Regression Refresher (before you take a more advanced stats course)
Linear Regression with R/Stata: Estimation, Interpretation and Presentation
Summer School
Multi-Level Structural Equation Modelling