Average IQ by State and Region: What the Data Can—and Cannot—Show

“Average IQ by state” looks like a simple ranking question. It is not. No agency gives one representative IQ battery to a random sample of every state, and the most-cited numbers were estimated from National Assessment of Educational Progress results rather than collected as IQ scores.

This page consolidates fifty state pages and Washington, DC into one auditable resource. It provides the complete historical table, explains exactly what it measures, and points readers to current NAEP results when the real question is how students are performing now.

To understand your own score rather than a geographic aggregate, use the free IQ test. For the methodological risks in group comparisons, read IQ test bias and heritability versus environment.

Average IQ by state: complete historical table

The values below reproduce the estimates reported by McDaniel (2006). They are shown to one decimal because that is how the paper reported them—not because state intelligence is known to one-tenth of a point. The rank is calculated from that historical table.

Historical regional comparison

Grouping the same estimates into U.S. Census regions reduces the false drama of a 1-to-50 ranking, but it does not solve the underlying measurement limitations.

The Northeast and Midwest are higher in this historical calculation; the South and West are lower. That pattern overlaps with differences in schooling, poverty, health, language background, urbanization and migration. It is not evidence that a region possesses a fixed biological level of intelligence.

How the state estimates were made

McDaniel combined state-level NAEP reading and mathematics performance and transformed it onto an IQ-like scale. The paper reported a strong relationship between the estimates and other state indicators, but the approach inherits every limitation of using school achievement as a proxy for general intelligence.

• NAEP tests students, not the whole adult population. The sampling frame and age group differ from a state-wide IQ survey.
• Achievement and intelligence overlap without being identical. Reading and mathematics reflect instruction, curriculum and opportunity as well as general reasoning.
• The values are cohort- and era-specific. They do not describe a permanent state trait.
• Small rank gaps are not robust. Measurement and modeling uncertainty can reverse close positions.
• Between-state patterns do not identify causes. A correlation cannot separate schooling, health, income, migration, test conditions and other influences.

The Flynn effect is a useful reminder that population test performance changes across generations. The g-factor explainer also shows why a proxy built from achievement tests should not be treated as a direct measurement of general intelligence.

What current NAEP results can tell us

For current educational achievement, the National Center for Education Statistics publishes the Nation’s Report Card with state and district trends. The 2024 results report reading and mathematics performance by grade and jurisdiction, including statistical comparisons with earlier assessment years.

Those results are the right source for questions such as “How did fourth-grade reading change in my state?” They still should not be relabeled as current IQ. Achievement data can guide education policy; an IQ-style transformation adds assumptions that readers need to see.

A better way to read a state difference

Suppose two historical estimates differ by three points. That is one-fifth of the standard deviation used on conventional IQ scales. It is smaller than the uncertainty attached to many individual scores and far smaller than the spread of people inside either state. The responsible interpretation is “the proxy differed modestly in this dataset,” not “people in one state are smarter.”

State-level outcomes are also responsive to conditions. Childhood health, lead exposure, school funding, instructional time, family resources, nutrition, language access and selective migration can all affect measured performance. These are mechanisms and hypotheses to investigate—not excuses to convert a ranking into an innate hierarchy.

Frequently asked questions

Related reading

← Browse all articles on the blog

References

1. McDaniel, M. A. (2006). Estimating state IQ: Measurement challenges and preliminary correlates. Intelligence, 34(6), 607–619. doi:10.1016/j.intell.2006.08.007.
2. National Center for Education Statistics. (2025). NAEP 2024 Reading: Performance Trends for States and Districts. The Nation’s Report Card.
3. National Center for Education Statistics. (2025). NAEP 2024 Mathematics Results. The Nation’s Report Card.
4. American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for Educational and Psychological Testing. AERA.
5. Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer, E. (2012). Intelligence: New findings and theoretical developments. American Psychologist, 67(2), 130–159. doi:10.1037/a0026699.