Experimental design is the plan for testing cause and effect. The researcher manipulates an independent variable, measures a dependent variable and uses random assignment so the groups start out equal. In a within-subjects design, the same people experience every condition instead.
Why it matters
Only an experiment can show that one thing causes another, because random assignment and control remove confounding variables. Correlational and survey studies can show links but not causes.
How the exam tests it
Stems ask you to name the design and its main weakness. Testing the same people rested and then sleep-deprived is within-subjects, and its risk is order effects like practice or fatigue, which counterbalancing fixes. Others ask which design best isolates a cause, like Asch's lab study of a unanimous wrong majority.
Don’t confuse it with
Students confuse random assignment with random sampling, but random sampling picks who is in the study, while random assignment decides which group each person joins.