Reading a Longitudinal Design with Attrition
This video presents the same text shown beside it, spoken and on screen. It adds nothing the text does not say.
State
A longitudinal design with heavy attrition supports conclusions only about the participants who remained.
Show
A sixty-year study ends with the healthiest quarter of its original sample still enrolled.
Watch for
Attrition is rarely random. The survivors differ from the dropouts in ways that shape the result.