This unit covers modern statistical methods that are now available for assessing the causal effect of a treatment or exposure from a randomised or observational study. The unit begins by explaining the fundamental concept of counterfactual or potential outcomes and introduces causal diagrams, known commonly as directed acyclic graphs (DAGs). DAGs enable us to visualise causal pathways to identify confounding, selection and other biases that prevent unbiased estimation of causal effects. Throughout the unit students are introduced to statistical methods for analysing data from observational studies that generate estimates with a causal interpretation. You will learn propensity score methods and how to assess whether the effects of an exposure on an outcome is mediated by one or more intermediate variables, suggesting potential mechanisms and pathways. Comparisons will be made with 'conventional' statistical methods. Emphasis will be placed on interpretation of results and understanding the assumptions required to allow inferences to be called 'causal'. Stata and R software will be used to real datasets from cohort studies.
the expected workload for this unit is 8-12 hours per week on average for 13 wks, consisting of guided readings, discussion posts, independent study and completion of assessment tasks.
two major assignments worth 30% each. these consist of statistical analysis of data from a cohort study and the preparing a report presenting the results in tables and figures with a written description of the research question, methods, results and discussion/conclusion equivalent to 1, 500 word essay. five online quizzes each worth 8% (for a total of 40%) are each equivalent to a series of short answers of about 400 words, so 2, 000 words in total.
There is no single prescribed text for the subject, but a number of reference books are suggested as background material.
(PUBH5010 or BSTA5011) and BSTA5023 and (BSTA5007 or PUBH5017)