Longitudinal Data Analysis
Date:
04/04/2019
Organised by:
University of Kent
Presenter:
Trainer: Professor George Chryssochoidis.
Level:
Entry (no or almost no prior knowledge)
Contact:
Nikki Gregory, N.C.Gregory@kent.ac.uk
Description:
Description:
Many research topics and aims in social sciences involve a large number of repeated measurements for each individual, a number of (observed) independent variables and one dependent variable (called here longitudinal models). These include subject-specific effects and dynamic models, marginal models and growth-curve models with nested and crossed effects. These type of longitudinal models are intrinsically linked to multilevel/hierarchical linear models given their ultimate nested nature within individuals/single units and these models are the focus of this workshop. These longitudinal models are different to models that involve 3 to 6 data points of repeated measurements and a number of (latent or observed) independent and dependent variables belonging to multiple simultaneous equations (these are usually termed (latent) growth models). These are also different to models that involve many data points of repeated measurements of a number of (latent or observed) independent and dependent variables belonging to multiple simultaneous equations (these are usually called dynamic SEM models). They are also different to topics and aims which involve many more data points (tens/hundreds/thousands) of repeated measurements but for a small number of (observed) independent variables (usually the remit of time series). In this seminar you will learn how to analyze longitudinal data by way of mixed models and topics covered will include:
1. A review of linear regression
2. Longitudinal data structure and time effects and correlated residuals
3. Conventional random intercept models and such models accommodating endogenous covariates
4. Fixed intercept models
5. Random coefficient models
6. Fixed coefficient models
7. Lagged-response (dynamic) models
8. Dropouts
This workshop will use the software package R: participants do not need any prior knowledge of R for the workshop.
Trainer: Professor George Chryssochoidis.
Cost:
Free to Kent postgraduates (and Kent staff) and Free to PGRs in Kent doctoral training partnership institutions,*
£30 for all other PGRs
£60 for other researchers: (non-Kent staff, alumni, charity, not for profit organisations)
£90 for commercial participants.
Website and registration:
https://www.kent.ac.uk/graduateschool/skills/advancedtraining.html
Region:
South East
Keywords:
Quantitative Data Handling and Data Analysis, Regression Methods, Linear regression, Longitudinal Data Analysis
Related publications and presentations:
Quantitative Data Handling and Data Analysis
Regression Methods
Linear regression
Longitudinal Data Analysis