Skip to main content

Designing an Experiment

This video presents the same text shown beside it, spoken and on screen. It adds nothing the text does not say.

State

Designing an experiment means naming the hypothesis, operationalizing both variables, assigning randomly, specifying the control condition, and closing each confound in turn.

Show

One question — does a ten-minute nap improve afternoon recall? — carried through every design decision aloud, including two confounds found and closed, and one that cannot be closed and is reported instead.