Of Climate Change and Competing Risks.

Of Climate Change and Competing Risks This Stats, STAT! animated video explores the concept of competing risks - and discusses why it is so important for investigators to consider whether the occurrence of one event can prevent or change the likelihood of the occurrence of another.

Good Intentions to Treat.

Good Intentions to Treat This Stats, STAT! animated video explores common approaches to analyzing data from randomized controlled trials, including intention-to-treat, per-protocol, and as-treated analyses.

How Statistical Power Works.

How Statistical Power WorksThis Stats, STAT! animated video explores the concept of statistical power and explains how clinical investigators determine how many participants to enroll in a randomized trial.

Bayesian Way.

Bayesian WayThis animated video explores two possible approaches to analyzing data in a randomized controlled trial: "Frequentist" versus "Bayesian."

Large Language Models.

Large Language ModelsIn the latest edition of Stats, STAT!, Fralick and colleagues explain the statistics behind large language models - used in chat bots like ChatGPT and Bard. While these new tools may seem remarkably intelligent, at their core they just assemble sentences based on statistics from large amounts of text.

How Censoring Works.

How Censoring WorksA common challenge in clinical research is determining the time to occurrence of a given event. This animated video explores the concept of censoring in survival analysis and how investigators deal with ambiguity in the time of an event's occurrence.

Subgroup Analyses: Subpar or Sublime?

Subgroup Analyses: Subpar or Sublime?This animated video explores some of the potential pitfalls of performing subgroup analyses in randomized controlled trials and explains how to approach potential findings with caution.

The Problem of Multiple Comparisons.

The Problem of Multiple Comparisons This animated video reviews the problem of multiple comparisons in research studies and explains how performing multiple statistical hypothesis tests can produce associations simply by chance.

Threats to Blinding.

Blinding is a critical strategy used to limit certain types of bias in randomized controlled trials. This animated video explores the rationale and examines potential threats to keeping group allocation concealed - from study participants and investigators.