EPM5008 · Longitudinal and correlated data analysis

Overview

This unit will develop statistical models for longitudinal and correlated data in medical research. The concept of hierarchical data structures will be developed, together with simple numerical and analytical demonstrations of the inadequacy of standard statistical methods. Normal-theory model and statistical procedures i.e. mixed linear models are explored using SAS or Stata statistical software packages. Extension to non-normal outcomes emphasising clinical research question. Case studies contrast generalised estimating equations and generalised linear mixed models. Limitations of traditional repeated measures analysis of variance and non-exchangeable models.

Offerings

Assessment

The Handbook does not list a final examination among the assessment items. That is not a guarantee there is none.

Assessment details may change. Please refer to the assessment information in Moodle closer to the start of the teaching period.

Requisites

Learning outcomes

1. Recognise the existence of correlated or hierarchical data structures, and describe the limitations of standard methods in these settings.
2. Develop and analytically describe an appropriate model for longitudinal or correlated data based on unit matter considerations.
3. Be proficient at using a statistical software package (e.g. Strata or SAS) to properly model and perform computations for longitudinal data analyses, and to correctly interpret results.
4. Express the results of statistical analyses of longitudinal data in language suitable for communication to medical investigators or publication in biomedical or epidemiological journal articles.

Workload

No workload detail published.

Student experience, not official rules. Nothing here changes what the Handbook says.