Skip to content
AP Psychology · Mental and Physical Health

Experimental design

Experimental design appears in 25 reviewed AP Psychology questions and sits in Mental and Physical Health. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

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.

Experimental design in real questions

Example 1Medium
In a study on sleep deprivation and reaction time, participants complete a reaction time test after 8 hours of sleep, then again after 24 hours without sleep. Which design is this, and what is the primary concern?
  1. ALongitudinal design; the primary concern is attrition, as participants may drop out over the extended time period
  2. BBetween-subjects design; the primary concern is that the two groups may differ in initial reaction time
  3. CWithin-subjects design; the primary concern is order effects, as being tested twice may affect performanceCorrect
  4. DCross-sectional design; the primary concern is cohort differences between participants tested at different times

Why C is right

A within-subjects (repeated measures) design uses the same participants in all conditions. This controls for individual differences but introduces order effects: participants may perform differently on the second test due to practice (improving performance) or fatigue/carryover (worsening performance), independent of the sleep manipulation.
Why the other 3 choices miss
A
Incorrect because a longitudinal design tracks participants over months or years; this study measures the same participants across two conditions in a shorter timeframe, which is within-subjects.
B
Incorrect because a between-subjects design uses different participants for each condition; here the same participants complete both conditions, making it within-subjects.
D
Incorrect because a cross-sectional design compares different age groups at one point in time; this study uses the same participants across two conditions.
Example 2Harder
A researcher studying implicit racial bias uses a reaction time paradigm where participants respond to positive or negative words paired with White or Black faces. Which type of research does this BEST represent, and what is its primary advantage over self-report surveys?
  1. ACase study research; it focuses on detailed, individual-level data about each participant's response patterns
  2. BCorrelational research; it examines the relationship between racial attitudes and response time without manipulation
  3. CExperimental research; it can measure implicit attitudes that participants may not be aware of or willing to reportCorrect
  4. DNaturalistic observation; it provides data in a real-world context rather than an artificial lab setting

Why C is right

Reaction time paradigms (like the Implicit Association Test) are experimental measures that use millisecond differences in response time to infer implicit (automatic, unconscious) attitudes. The key advantage over self-report surveys is that participants cannot easily control or fake reaction times, bypassing social desirability bias.
Why the other 3 choices miss
A
Incorrect because case study research focuses intensively on individuals; this experiment collects data from many participants and analyzes group patterns, not individual case histories.
B
Incorrect because correlational research examines naturally occurring relationships without manipulation; this study manipulates stimulus pairings (race × valence) — it is experimental.
D
Incorrect because this study is conducted in a controlled lab environment; naturalistic observation occurs in real-world settings without experimental manipulation.
Example 3Medium
In a clinical trial for a new antidepressant, neither the patients nor the physicians administering the drug know whether a patient is receiving the drug or a sugar pill. Patients in both groups show symptom improvement. Which factor MOST likely explains why the placebo group improved?
  1. ADemand characteristics, because patients are guessing the study's hypothesis and behaving accordingly
  2. BRegression to the mean, because patients naturally improve over time regardless of treatment
  3. CExperimenter bias, because the physicians' expectations are influencing patient outcomes
  4. DThe placebo effect, because patients' belief in the treatment produces real reliefCorrect

Why D is right

The placebo effect occurs when participants experience real improvements in symptoms because they believe they are receiving an effective treatment, even when they are not. It operates through expectancy and psychological mechanisms that can produce measurable physiological and psychological changes.
Why the other 3 choices miss
A
Incorrect because demand characteristics occur when participants guess the study's purpose and change behavior accordingly; the placebo effect is about belief in treatment effectiveness, not about figuring out the study's hypothesis.
B
Incorrect because regression to the mean is a statistical phenomenon where extreme scores return toward average over time; while it may contribute, the primary psychological explanation for placebo group improvement is the placebo effect.
C
Incorrect because experimenter bias involves the researcher's expectations influencing results; in a double-blind study, physicians do not know who receives the drug, so this bias is controlled.

Studied alongside

Other concepts from Mental and Physical Health.