AP® Psychology · Unit 2 · Topic 2.8
The Flynn Effect and Bias in Testing
The Flynn effect is the steady rise in raw intelligence test performance across generations worldwide, driven by environment, not genetics. It's a key reason test-score differences between groups can't be read as proof of innate ability gaps.
James Flynn noticed something strange: give people today the exact same intelligence test their grandparents took, unaltered, and today's average comes out way higher. This has happened in country after country, decade after decade. The usual suspects are environmental: better nutrition, more schooling, smaller families, and a world that constantly demands abstract problem-solving, from reading maps to using apps. None of this is genetic change, which is far too slow to explain a shift this fast. Test makers respond by re-norming: they redefine the scale so the mean always lands at 100. That re-norming hides the climb. The raw performance of the whole distribution is sliding right generation after generation, even while the reported average stays fixed.
The Flynn effect matters beyond its own trivia because it's a clean demonstration that intelligence test performance is sensitive to environment, not just fixed biology. That sets up a bigger caution: when two groups show different average scores, the gap between those averages is almost always tiny compared to the spread of scores within each group. Picture two overlapping bell curves. Even if one curve sits a bit left of the other, huge stretches of both curves cover the same territory, so plenty of people in the 'lower' group outscore plenty of people in the 'higher' one. A between-group average difference, especially one that small relative to within-group variation, tells you nothing about innate capacity.
History shows the cost of ignoring this. In the early twentieth century, intelligence tests, often biased by language and culture, were used to label immigrant groups as less intelligent, feeding restrictive immigration quotas and discriminatory hiring. The scores reflected unfamiliarity with test language and culture, and economic disadvantage, not fixed ability, yet they were treated as biological fact.
Key terms
- Flynn effect
- The documented rise in raw intelligence test performance across successive generations in many countries.
- Re-norming
- Periodically adjusting a test's scale so the average score stays fixed at a reference point like 100.
- Within-group variation
- How much individual scores differ from each other inside a single group.
- Between-group variation
- How much the average scores differ from one group to another.
- Overlapping distributions
- A pattern where two groups' score ranges cover much of the same territory despite differing averages.
Watch it explained
- How it worksHow does the whole distribution slide right while the mean is pinned back to 100?
- How it worksWalk me through variation within a group dwarfing variation between groups.
- Common mistakeWhy isn't a group difference in test scores evidence of a group difference in ability?
- Going furtherWhat happened when intelligence scores were used to restrict immigration and jobs?
Check yourself
End of topic.