Dates: 4 to 7 November 2019 (Monday to Thursday) Venue: Faculty of Health Sciences Time: 08h30 to 16h00
Course fee is R 7 000 per person (UCT Staff/Students are eligible for a discounted rate. Please enquire on application)
Presented by Division of Epidemiology & Biostatistics at the School of Public Health & Family Medicine
Instructors: Dr Jabulani Ncayiyana, Prof Jonathan Chevrier (McGill University, Canada)
This is a four-day course that aims to introduce basic and intermediate concepts of causal inference for observational studies: Methods based on the counterfactual (potential outcomes) framework will be covered including marginal structural models and mediation analysis.
The use of directed acyclic graphs for the identification and correction of bias will also be introduced with a focus on confounding and selection bias. Time-varying exposure and covariates will be touched upon, but focus will be on situations of point treatment/exposure.
Participants will be required to work on their own personal laptops during the session and must come with a statistical package pre-installed. Participants are free to use the package of their choice however the instructors will be able to provide support primarily using Stata.
Entry requirements: The course is aimed at epidemiologists, statisticians, and other researchers who work with longitudinal observational data. A good theoretical and practical understanding of basic epidemiological and statistical concepts will be assumed including confounding and selection bias as well as standard regression methods (e.g. linear, logistic). Experience coding using a statistical package is also assumed.