Dr Jon Heron (course organiser), Prof Kate Tilling, Dr. Laura Howe, Dr. Liam Mahedy.
22-24 March 2016
This course provides an introduction to methods for analysing longitudinal/repeated continuous data, with a focus on methods for analysing individual trajectories over time. Multilevel modelling and structural equation modelling (SEM) approaches will be described and compared. Throughout the course there will be an emphasis on what kinds of research question different types of method can be used to explore and on the interpretation of results.
The course is intended for researchers who are, or will be, involved in analysing longitudinal data. Participants should be familiar with the algebra of standard regression models for continuous outcomes (to beyond the standard of the short course " Introduction to Linear and Logistic Regression Models", or to the level implied by module 3 of the Centre for Multilevel Modelling's online multilevel modelling course, which can be found at www.bristol.ac.uk/cmm/learning/course.html).
The course will comprise a total of seventeen hours teaching, including introductory talks, lectures, worked examples and practical sessions using both Stata and Mplus software package. Approximately 40% of the course will be practical work.
We also recommend that participants refresh their knowledge of single-level models for continuous responses in advance of the workshop by reading the ‘concepts’ parts of Module 3 of the Centre for Multilevel Modelling's online multilevel modelling course (www.bristol.ac.uk/cmm/learning/course.html). In particular, participants should ensure they are familiar with algebraic regression equations, and partitioning of variation (e.g. ANOVA).
Please note: This course is concerned with analysis of repeated measures. If you wish to learn about rates or survival analysis, please see the short course Rates and Survival Analysis. The course is also not about time-series models nor does it cover repeated binary/ordinal data.
For further information: please contact email@example.com.