This training session is delivered by Dr Abbey Page for the GUDTP
This two-day quantitative methods workshop provides a practical introduction to four widely used approaches for analysing complex social, behavioural and health data in R. It is assumed you have some familiarity with R and statistical modelling (linear regression – which will be briefly recapped during the session), the workshop will combine conceptual explanation with hands-on coding exercises using applied examples.
The first half-day will introduce multiple imputation for missing data, covering why and when missing data can be problematic, the assumptions behind imputation, and how to implement, diagnose and pool imputed models in R. The second half of the day will cover directed acyclic graphs (DAGs) as a tool for causal thinking, helping participants to clarify assumptions, identify confounding, and decide which variables should or should not be included in statistical models.
The first half of the second day will focus on multilevel modelling after briefly recapping linear regressions in general. It will introducing hierarchical data structures, random intercepts and slopes and interpretation of effects across clustered observations. On the second half of the final day, participants will be introduced to survival analysis, including time-to-event data, censoring, Kaplan-Meier curves, Cox proportional hazards models, and visualising survival estimates in R.
Across the two days, the emphasis will be on building confidence in choosing appropriate methods, understanding the assumptions behind each approach, and translating statistical concepts into reproducible R workflows. Please note that you are expectd to attend both full days if you sign up for this as the workshop topics are designed to build upon each other.
Book now
