Generalizability in AP Psychology
Generalizability is the extent to which results from a study's sample can be applied to the larger population the researchers care about. It depends mainly on how representative the sample is, which comes from who was selected (ideally through random sampling) and how varied those participants are.
This definition is part of Fiveable's AP Psychology review, which also has guided practice questions.
Random sampling builds generalizability; random assignment builds causation
Researchers almost never test a whole population. They test a sample, a smaller group drawn from it. If the sample resembles the population in age, culture, background, and other relevant traits, it's called a representative sample, and the findings can be extended to that population with more confidence.
Random sampling is the main tool for getting there. Every member of the population has an equal chance of being picked, which reduces selection bias. Some groups can still end up over- or underrepresented by chance, especially in small samples. This step happens before the study starts, when researchers decide who gets in.
Random assignment is a different step that happens after participants are already in. It puts each person into the experimental or control group by chance, which evens out differences between groups and lets researchers claim the independent variable caused changes in the dependent variable. That's internal validity, a separate question from generalizability. Keep the two linked to their jobs. Sampling answers "who do these results apply to?" Assignment answers "did the IV cause the effect?"
A tight experiment that still doesn't travel far
Say a researcher recruits 80 psychology majors at one university. She randomly assigns half to sleep 4 hours and half to sleep 8 hours, then gives everyone the same memory test. The 4-hour group scores lower.
The causal claim here is solid for this sample, because random assignment and a control group are in place. Generalizability is weak, though. Every participant is a college student of similar age at one school, picked because they were convenient. The results may not hold for older adults, children, shift workers, or people in other countries.
To improve generalizability, she'd randomly sample from a broader population, like adults of different ages across several regions. Adding another control group or re-running the random assignment wouldn't change who the results apply to. A bigger sample from the same campus wouldn't help much either. More psychology majors are still psychology majors.
Designs and samples that limit how far results reach
Generalizability isn't tied to one numbered unit or topic. It belongs to the course-wide research methods skills, especially evaluating design elements in experiments (Skill 2.B) and in non-experimental studies like surveys, correlational studies, and case studies (Skill 2.C). So it can be attached to a study about memory, development, personality, or social behavior.
Some situations cap generalizability almost automatically:
- Case studies. One person or a very small group gives rich detail but can't stand in for a population.
- Convenience samples. Volunteers or students who happened to be available often differ from the general population.
- Culturally narrow samples. Much psychology research has drawn on participants from Western, educated, industrialized, rich, and democratic (WEIRD) societies. Cultural norms shape behavior (Skill 1.B), so findings from one culture may not hold in another.
- Artificial settings. A lab task may not reflect how people behave in everyday life, which limits how well results extend to real situations.
A correlational study with a large random sample can generalize well even though it can't show causation. Generalizability and causation are judged separately, so one strength doesn't make up for the other's weakness.
Judging generalizability in the Article Analysis Question
The Article Analysis Question (free-response Question 1) gives you a summarized research article and explicitly asks you to judge whether its findings can be generalized. A strong answer names a specific detail from the summary, like who the participants were, how many there were, or how they were selected. Then it explains why that detail widens or narrows the population the results can apply to. "The sample was small" alone is thin. "All participants were female undergraduates from one university, so the results may not apply to men or older adults" earns the point.
In the Evidence-Based Question, sample limits can strengthen your reasoning (Skill 4.B). If one of the three studies used a narrow sample, you can say its support for your claim is more limited than a study with a broader, randomly selected sample.
Multiple-choice items often describe a study and ask which change would improve its generalizability. The right answer involves random sampling or a more diverse, representative sample. The usual trap answers are random assignment, a control group, or a placebo condition, which strengthen causal conclusions but don't change who the findings apply to.
Keep studying AP® Psychology Unit 0
Official unit cheatsheet
open one-pagerKeep studying AP Psychology
Connect this key term to the AP exam workflow: review the course, practice questions, and check related study tools.
AP Psychology hub
Review units, study guides, and course resources.
AP-style practice
Check this vocabulary in multiple-choice context.
FRQ practice
Apply key concepts in written AP responses.
Score calculator
Estimate the exam score you are working toward.
Cheatsheets
Review the highest-yield facts before practice.
Practice exam
Put the full course together before test day.