Designing for external validity

External validity is a design decision made before enrolment, not an analysis run afterwards. Who you let in, where you recruit, and what you compare against all fix in advance how far the answer will reach. Shadish, Cook and Campbell's threats work as a checklist on your own design or someone else's paper: interactions between the effect and the people, the setting, the treatment variant, the outcome measure, and the point in history. An efficacy trial under ideal conditions and an effectiveness trial in ordinary ones answer different questions.

Also covered: threats to external validity · efficacy vs effectiveness · eligibility criteria · study setting

Alongside this step

Slides and transcript — the whole video
Read the chapter — External Validity and Generalizability

Check yourself

5 questions, marked as you go. Nothing is recorded as a grade and nothing is locked behind it — if a question catches you out, that is the useful part.

Lesson 3 · 4 steps

1External validity, generalizability, and transportability4:50 2Does the sample stand for the population it came from?10:14 3Do the results apply to a different population entirely?9:46 4Designing for external validity9:57

← Previous step

The Validity Toolkit · lesson 3 of 4