Overview
This advanced course develops statistical modelling strategies for applied research in health, medicine, public health, and the life sciences. Topics include multiple linear and generalized linear models; logistic and Poisson regression; count and rate outcomes; link functions; interactions; nonlinear effects; model specification; diagnostics; calibration; discrimination; and predictive performance.
Students examine confounding control, causal diagrams, adjusted analysis, clustered and repeated-measures data, mixed-effects models, generalized estimating equations, survival and time-to-event analysis, censoring, hazard models, competing risks, missing data, multiple imputation, internal validation, and sensitivity analysis. Selected applications may include longitudinal, multistate, Bayesian, or penalised models.
Practical work uses authentic health datasets and statistical software. Students formulate estimands, develop analysis plans, fit and compare models, assess assumptions, conduct diagnostic and sensitivity analyses, distinguish prediction from explanation, and communicate results through reproducible code and publication-style reporting that clearly addresses uncertainty and limitations.
Learning Outcomes
- Formulate clinically and scientifically meaningful estimands for applied health research questions.
- Select regression, survival, longitudinal, clustered-data, or predictive modelling strategies that reflect the outcome structure and study design.
- Fit and interpret linear, generalized linear, mixed-effects, generalized estimating equation, and time-to-event models using appropriate link functions and parameterizations.
- Evaluate confounding structures using causal diagrams and justify adjusted analyses in relation to the target estimand.
- Diagnose model inadequacy through residual analysis, calibration assessment, discrimination measures, influence diagnostics, and assumption checking.
- Compare explanatory and predictive models using uncertainty estimates, validation procedures, and clinically relevant performance measures.
- Implement principled approaches to censoring, competing risks, missing data, multiple imputation, and sensitivity analysis.
- Synthesize statistical results into transparent, publication-quality reports with reproducible code, appropriate clinical or population interpretation, and explicit limitations.
Timetable
| Type | Length | Frequency | Period |
|---|---|---|---|
| Lecture | 2 hours | Weekly | All semester |
| Lab | 2 hours | Weekly | All semester |
| Tutorial | 1 hour | Fortnightly | All semester |
| Workshop | 1 hour | Fortnightly | All semester |
Assessment Schedule
| Type | Description | Weighting |
|---|---|---|
| Assignment | Analysis plan and estimand specification | 15.00% |
| Assignment | Model development exercises (4 × 5%) | 20.00% |
| Test | Practical modelling test | 15.00% |
| Capstone | Publication-style applied analysis report with reproducible code | 35.00% |
| Exam | Final examination | 15.00% |
Teaching Staff & Programs
This course is delivered jointly by faculty from the participating programs listed below. In line with the Douchewater Way, the University of Sexology tailors core instruction directly to each cohort's specific discipline — adapting curriculum to program needs rather than forcing students into a one-size-fits-all model. Learn more about our approach at The Douchewater Way.