IIF Certified Assessment
Start IQ Test
IIF Certified Assessment
Surveys going back to the 1970s tie vocabulary-test scores to party identification. The gap is small, mostly explained by education and income, concentrated in verbal ability specifically, and has partly reversed since the 1980s.
The claim that one side of the political aisle is measurably smarter than the other gets forwarded every election cycle, usually stripped of every qualifier the original research carried. The actual literature — running from the 1970s to a 2024 twin study — tells a far more specific and far less flattering-to-anyone story: the gap is small, it depends heavily on which kind of intelligence and which kind of political belief you measure, and part of it has reversed direction since the Cold War ended.
Almost none of this research uses a formal IQ test. The workhorse instrument is Wordsum, a ten-item vocabulary test built into the General Social Survey, a US survey run regularly since 1972. Wordsum is not an IQ test — it asks respondents to pick the correct synonym for an uncommon word out of five options — but vocabulary size correlates strongly enough with general cognitive ability that researchers use it as a practical stand-in when a full test is not available. That substitution matters: it means most of the headline findings here are really about verbal ability specifically, not intelligence as a whole, a distinction that turns out to matter a great deal.
Political scientist Noah Carl published the most widely cited version of this analysis in the journal Intelligence in 2014, using decades of GSS data. Republicans scored 2 to 5 points higher on average than Democrats on the raw vocabulary measure. Once Carl adjusted for respondents' education and income, the advantage "almost entirely disappeared or, in some cases, even reversed." A follow-up using a broader set of cognitive measures — not just vocabulary — found the Republican advantage persisted at 1 to 4 points even after adjustment, with more than half of the original relationship still explained by socioeconomic position.
That is the central complication in this entire field: education, income and vocabulary-test performance are all correlated with each other for reasons that have nothing to do with which party someone votes for, so a raw comparison between party identifiers is measuring several things at once. Any figure quoted without saying whether it is adjusted for socioeconomic status is not comparable to any other figure in this space.
A 2015 meta-analysis in the European Journal of Personality by Emma Onraet and colleagues pooled 67 separate studies covering more than 84,000 participants and found a moderate, reliable association between lower cognitive ability and more conservative social attitudes — views on traditional values, social change and out-groups. The relationship with economic conservatism — tax policy, the size of government, redistribution — is much weaker and considerably less consistent from one study to the next. A 2021 analysis in Personality and Social Psychology Bulletin by Alexander Jedinger and Alexander Burger described two "countervailing" pathways operating at once: self-interest tends to pull higher-cognition, typically higher-income respondents toward economically conservative positions, while a stronger drive toward novel or complex ideas pulls the same people toward liberal ones. The two pathways partly cancel out, which is a large part of why the overall economic-conservatism link comes out so weak and so inconsistent across studies.
A 2017 study by Steven Ludeke, using two American community samples with separate verbal and non-verbal test scores, found that the liberal association tracked verbal ability specifically — non-verbal ability showed no independent relationship once verbal scores were accounted for. A larger 2024 study in the journal Intelligence, led by Tobias Edwards and drawing on thousands of twins and siblings from the Minnesota Center for Twin and Family Research, confirmed the pattern at scale: verbal IQ predicted voter turnout, civic participation and liberal political attitudes roughly twice as strongly as performance IQ, and the association held up even after controlling for education and family background.
Verbal IQ predicted political engagement about twice as strongly as non-verbal IQ — and part of the relationship has flipped since the 1980s.
The clearest reason to distrust any confident claim in this space is that the relationship has changed sign over time. Carl's GSS analysis noted that the Republican vocabulary-score advantage began narrowing in the early 1980s and had flipped by around 2010. Edwards's 2024 twin study, working with a different generation of data, found higher intelligence now associated with less economic conservatism — the opposite of what older mid-twentieth-century research reported. Edwards's own comment on the finding is worth keeping in mind whenever this topic comes up again: there is no law saying intelligent people must always land on one particular side of a debate. A correlation measured in one country in one decade describes that country in that decade — not a timeless property of intelligence itself.
The honest summary is unglamorous: the measured gaps are small — a handful of points on a vocabulary test, not a chasm — concentrated specifically in verbal ability rather than intelligence broadly, substantially explained by education and income, different in size and even direction for social versus economic attitudes, and partly reversed from what older research found. That is a long way from the confident one-line claims this topic usually gets reduced to on social media. Our explainer on what twin studies show about genetics and environment in test scores covers the same causal-inference problem from a different angle, and our piece on how much IQ actually predicts income is useful context for why education and income keep showing up as the variables that explain away most of the raw gap here.
If you want to see where your own score sits against a reference population — independent of any political framing — our IQ percentile calculator converts a raw score into a percentile the way the studies above do, and a full IQ test is the only way to get a score that is not a ten-word vocabulary proxy.
Every number in this article came from a specific, named study — not a talking point. If a claim about intelligence and politics reaches you without a study, a sample size and a year attached, treat it as folklore until it has all three.
Find your IQ score now! →None of this settles an argument that was never really about data to begin with. What the research supports is much narrower than either side wants: a small, shifting, mostly-explained-by-other-factors association between one kind of vocabulary test and one kind of political label — not a verdict on which half of the country is smarter.
Neither, reliably. Studies using the General Social Survey's vocabulary test have found Republicans scoring 2 to 5 points higher on average before adjusting for education and income — an adjustment that, according to the researcher who ran it, makes the gap "almost entirely disappear or, in some cases, even reverse." The relationship has also changed direction since the 1980s, so a figure from older research does not describe today.
It depends which kind of conservatism. A 2015 meta-analysis of 67 studies found a moderate link between lower cognitive ability and social conservatism — views on tradition and social change. The link with economic conservatism, such as views on taxation and government size, is much weaker and less consistent across studies, because two opposing effects appear to cancel out.
Mostly not. The largest body of evidence uses Wordsum, a ten-word vocabulary test embedded in the General Social Survey, as a practical proxy for general cognitive ability rather than a full IQ test. Some newer studies, including a 2024 twin study, do use separate verbal and performance IQ measures — and find the political association concentrated specifically in verbal ability.
The research cannot show that. Every study referenced here is observational — survey or twin data showing a correlation, not a controlled experiment. Education, income, genetics and regional culture could all produce similar patterns without intelligence causing political belief in either direction.
Corrections: spotted an error? Email corrections@iqmetrics.org and we will update this story and note the change here.
Identical twins raised in separate homes end up more alike on IQ than fraternal twins raised in the same one. That single, replicated pattern is the whole logic behind how heritability gets estimated — and behind why the number is so often misread.
The meta-analytic correlation between measured intelligence and earnings is around 0.23, which leaves roughly ninety-five per cent of the differences in pay to everything else. A 2026 analysis of two large US cohorts adds a second finding: your specific strengths carry a third to a half of the weight that general ability does.
No state has a measured average IQ. The popular tables convert school test scores from 2005 into IQ points, the spread between typical states is about two points on that scale, and the poverty gap inside one state is bigger than the whole map.
Our IIF-certified assessment reports your score with its scale, percentile and confidence range — and a breakdown of the cognitive domains behind it.
Start IQ Test →
