Construct validity is the degree to which a test actually measures the theoretical concept (construct) it claims to measure. In AP Psych, it's a core psychometric principle for evaluating intelligence tests, since 'intelligence' is an abstract idea, not something you can observe directly.
Construct validity answers one blunt question. Does this test really measure the thing it says it measures? A "construct" is an abstract psychological concept like intelligence, anxiety, or motivation. You can't put intelligence on a scale or measure it with a ruler, so any test of it is really measuring a theory about what intelligence is. Construct validity is how well the test lines up with that theory.
Here's why this matters so much for intelligence testing specifically. The CED points out that psychologists still don't agree on what intelligence even is. Some argue it's one general ability, others say it's multiple separate abilities. If researchers can't agree on the construct, building a test with strong construct validity gets messy fast. A classic threat shows up when a verbal IQ test is given to someone in their non-native language. At that point the test is partly measuring language fluency, not intelligence, and its construct validity breaks down. Under learning objective 2.8.B, all psychological assessments must follow sound psychometric principles to be useful, and construct validity is one of the big ones.
Construct validity lives in Topic 2.8 (Intelligence and Achievement) in Unit 2: Cognition. It directly supports 2.8.B (explain how intelligence is measured), which states that all psychological assessments must adhere to sound psychometric principles. It also connects to 2.8.A, because the debate over whether intelligence is one general ability or many abilities is literally a debate about the construct itself. And it powers 2.8.C, since personal and sociocultural biases in test items can quietly wreck construct validity, which is how systemic issues like language barriers and educational inequities end up baked into IQ scores. If you can explain construct validity, you can explain why a high or low IQ score doesn't automatically mean what it appears to mean.
Keep studying AP® Psychology Unit 2
Predictive validity (Unit 2)
This is the sibling concept you have to keep straight. Construct validity asks whether the test measures the right concept. Predictive validity asks whether scores forecast a future outcome, like whether an aptitude test predicts college grades. A test can predict outcomes decently and still measure the wrong construct.
Psychometric principles (Unit 2)
Construct validity is one piece of the bigger psychometric toolkit, alongside standardization and reliability. The CED says a test is only useful if it follows these principles. Think of construct validity as the 'are we measuring the right thing?' check within that toolkit.
Test-retest reliability (Unit 2)
Reliability means consistent scores; validity means accurate scores. A test can be perfectly reliable and still have zero construct validity, the way a broken scale that always reads 10 pounds heavy is consistent but wrong. Reliability is necessary for validity, but it's never enough on its own.
Bias in intelligence testing (Unit 2)
Topic 2.8.C's systemic issues are construct validity problems in disguise. When poverty, discrimination, or language barriers affect scores, the test is partly measuring those factors instead of intelligence. That's exactly what it means for construct validity to fail.
Construct validity shows up almost entirely in multiple-choice questions, usually as a scenario you have to diagnose. A typical stem describes a testing situation and asks which psychometric concept it illustrates, or asks you to spot a threat to construct validity. The classic example is immigrant students scoring low on verbal IQ sections in their non-native language while scoring high on non-verbal sections, which signals the test is measuring language ability, not intelligence. Watch for distractor traps too. A question about identical proctor training and scripts is testing standardization, not validity, and a question about getting the same score twice is testing reliability. Your job is to match the scenario to the right psychometric principle. No released FRQ has used the term verbatim, but it fits naturally into questions that ask you to evaluate research design or explain why a measurement might be flawed.
Both are types of validity, so they blur together easily. Construct validity is about the present concept. Does the test actually capture intelligence as theorized? Predictive validity is about the future. Do scores forecast a later outcome, like an aptitude test predicting GPA? Quick check for MCQs: if the question mentions whether the test reflects the underlying concept or theory, it's construct validity. If it mentions forecasting future performance, it's predictive validity.
Construct validity is the degree to which a test measures the abstract theoretical concept it claims to measure, such as intelligence.
It matters most for intelligence tests because psychologists still disagree on what intelligence is, which makes the construct itself contested.
A test can be highly reliable (consistent) and still lack construct validity (accuracy), so never treat reliability and validity as the same thing.
Language barriers and cultural bias are classic threats to construct validity because the test starts measuring those factors instead of intelligence.
Construct validity asks whether the test measures the right concept now, while predictive validity asks whether scores forecast future outcomes.
Under the CED, all psychological assessments must follow sound psychometric principles, and construct validity is one of them, to be considered useful.
Construct validity is how well a test measures the abstract psychological concept (construct) it claims to measure, like intelligence or anxiety. It's tested in Topic 2.8 as one of the psychometric principles every useful assessment needs.
No. Reliability means the test gives consistent scores, but a test can consistently measure the wrong thing. A verbal IQ test given in someone's non-native language might produce the same low score every time (reliable) while measuring language fluency instead of intelligence (not valid).
Construct validity checks whether the test measures the right concept, like whether an IQ test really captures intelligence. Predictive validity checks whether scores forecast a future outcome, like an aptitude test predicting college performance. On MCQs, 'measures the concept' means construct and 'predicts the future' means predictive.
Giving a verbal intelligence test to someone in their non-native language. If they score high on non-verbal sections but low on verbal ones, the test is partly measuring language skill rather than intelligence, so its construct validity is compromised.
Because the construct itself is disputed. The CED notes that researchers still debate whether intelligence is one general ability or multiple abilities, and definitions can be subject to bias. If experts can't agree on what intelligence is, agreeing on whether a test measures it accurately is even harder.
Connect this key term to the AP exam workflow: review the course, practice questions, and check related study tools.
Review units, study guides, and course resources.
Check this vocabulary in multiple-choice context.
Apply key concepts in written AP responses.
Estimate the exam score you are working toward.
Review the highest-yield facts before practice.
Put the full course together before test day.