Skip to content
AP Psychology · Mental and Physical Health

Confounding

Confounding appears in 3 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.

A confounding variable is a factor other than the independent variable that differs between groups. Because it could also explain the results, it threatens internal validity.

Why it matters

A confound means you cannot be sure the treatment caused the effect. That is why experiments use random assignment instead of letting people pick their group.

How the exam tests it

Stems describe a therapy study and ask which flaw weakens it. If participants chose CBT themselves, they may have been more motivated to get better. That self-selection confound could explain the improvement, even if the p-value is very small.

Don’t confuse it with

Statistical significance only says a result is unlikely to be due to chance, but it cannot rule out a confound or prove the treatment was the cause.

Confounding in real questions

Example 1Medium
A social psychologist studying aggression reviews one city's records for a single summer. She finds that on days when stores sell more cold drinks, police receive more reports of physical fights. She concludes that drinking cold beverages makes people more aggressive. Which concept best explains the error in her conclusion?
  1. AIllusory correlation, because the link between drink sales and fights in the records does not actually exist statistically
  2. BThe directionality problem, because fights in the city may lead residents to buy more cold drinks afterward
  3. CObserver bias, because the researcher's expectations about aggression led her to see a spurious association in the data
  4. DThe third variable problem, because hot weather may raise both cold-drink sales and irritability that fuels aggressionCorrect

Why D is right

This is the third variable problem: a correlation alone cannot show causation, because an unmeasured variable may cause both measured variables. Here that variable is hot weather. Heat increases cold-drink sales. In social psychology, heat is also a known situational factor that raises irritability and aggression. The correlation between drink sales and fights is real, but concluding that one causes the other ignores the temperature that drives both.
Why the other 3 choices miss
A
Incorrect because an illusory correlation means perceiving a relationship that is not in the data. Here the sales and fight records really do vary together; the mistake is the causal conclusion, not the existence of the correlation.
B
Incorrect because the directionality problem asks whether B caused A instead of A causing B. Fights causing drink purchases is implausible, and the real issue is an outside variable driving both.
C
Incorrect because observer bias means a researcher's expectations distort how data are recorded. Here the data come from store sales and police reports, so the error lies in interpreting a correlation, not in collecting the data.

Studied alongside

Other concepts from Mental and Physical Health.