How long until something happens? How long until a patient dies, a machine breaks down, a customer cancels their subscription, or a piece of equipment fails. These "time until X" questions seem simple, but they trip up standard statistical methods in ways that are easy to miss and hard to undo. This issue is one of the more common and consequential mistakes in applied statistics, and it has its own dedicated toolkit for a reason.

This workshop offers a practical, code first introduction to survival analysis using R. We will build up from first principles to applied modeling, covering:

Foundations: what makes time to event data different, and why it needs specialised methods Non parametric estimation: Kaplan Meier curves and log rank tests, including how to interpret and visualise them Semi parametric modeling: the Cox proportional hazards model, hazard ratios, and checking the proportional hazards assumption Extensions (time permitting): parametric survival models and a brief look at competing risks

Participants will work through live coded examples and short guided exercises in R, using realistic datasets, and will leave with a working script they can adapt to their own time to event data. Some familiarity with R and basic regression (for example linear or logistic regression models) is assumed. No prior survival analysis experience is necessary.

Format: 3 hours, mix of live coding, guided exercises, and Q&A. Requires laptop with a recent R installation and RStudio or similar IDE installed.