Homeschool IQ: Are Homeschooled Kids Smarter?

There is no good evidence that homeschooling makes children smarter. The widely quoted studies showing homeschoolers at the 70th to 80th percentile on achievement tests used volunteer samples from families with above-average education and income, and studies that control for family background find small or no differences.

No study we could find has compared the IQ scores of homeschooled and conventionally schooled children with a representative sample. This article explains what the achievement research does show, why its best-known numbers overstate the case, what the evidence on schooling and IQ implies for teaching at home, and how homeschooled children can be assessed.

Homeschooling is a minority choice in the United States. The National Center for Education Statistics (NCES) estimated that 2.8% of children aged 5 to 17, about 1,457,000 students, were homeschooled in 2019. The rate was highest, 6.6%, in two-parent households where one parent was in the labor force.

The most cited achievement study is Rudner (1999). In spring 1998, 20,760 homeschooled students from 11,930 families took the Iowa Tests of Basic Skills or the Tests of Achievement and Proficiency through a testing service run by Bob Jones University Press. Their median scores were typically in the 70th to 80th percentile. For a sense of scale, the 70th to 80th percentile sits about 0.5 to 0.85 standard deviations above the norm group's average (converting percentiles to z-scores on a normal curve), which would be roughly 8 to 13 points if expressed on an IQ scale. These were achievement tests, though, and the conversion is only a guide.

Rudner was clear about the limits. The parents in his sample had more formal education than the general population, and their median income was significantly higher than that of all US families with children. He wrote that because "this was not a controlled experiment, the study does not demonstrate that home schooling is superior to public or private schools," and he noted that he could not evaluate whether the sample represented all homeschoolers.

In their review of more than 2,000 texts on homeschooling, Kunzman and Gaither (2020) add context. The study was conceived and commissioned by the Home School Legal Defense Association (HSLDA), an advocacy group, and parents mostly administered the tests themselves. A series of HSLDA-sponsored studies by Brian Ray between 1990 and 2010 recruited volunteer families, who submitted their own children's test scores, and consistently reported averages at or above the 80th percentile. Kunzman and Gaither repeat Kunzman's earlier description of Rudner's report as "perhaps the most misrepresented research in the homeschooling universe," noting that it continues to be cited as proof that homeschoolers outperform schooled children despite its author's warnings.

Studies that compare homeschooled students with similar schooled students tell a quieter story. Kunzman and Gaither summarize the pattern this way: homeschooling "does not have that much of an effect on student achievement once family background variables are controlled."

• SAT takers: an analysis of 6,033 self-identified homeschooled SAT takers in 2001, summarized by Kunzman and Gaither, found no large gap once family background was controlled, with homeschoolers slightly ahead on the verbal section and slightly behind on mathematics.

• Verbal versus math: that verbal-high, math-low pattern recurs across studies from several US states over three decades, which suggests home instruction may favor reading and language more than mathematics.

• Structured versus unstructured: Martin-Chang, Gould and Meuse (2011) tested homeschooled and demographically similar public-school children under the same conditions, with the same researchers giving every test. Children taught from organized lesson plans outscored the public-school group, while those in unstructured "unschooling" families had the lowest scores of the three groups. The sample was small.

• A wider spread: in national drug-use survey data from 2002 to 2011, one subgroup of homeschooled adolescents was about three times more likely than schooled peers to report being behind their expected grade level (Green-Hennessy, 2014). Homeschooling covers very different practices, and averages hide that range.

College outcomes look similar. Yu, Sackett and Kuncel (2016) matched 732 homeschooled students at 140 colleges with traditional students on demographics, high school grades and SAT scores, and found no difference in first-year college grades or retention. Cogan (2010) found that homeschooled students at one Midwestern university earned somewhat higher grades, with no difference in retention or graduation. Both studies include only homeschoolers who went to college, which is itself a selected group.

The strongest evidence on education and intelligence comes from studies that are not about homeschooling at all. Ritchie and Tucker-Drob (2018) meta-analyzed 142 effect sizes from 42 data sets, covering more than 600,000 people, using designs that get closer to cause and effect: controlling for earlier IQ, exploiting changes in compulsory schooling laws, and comparing children just either side of school-entry age cutoffs. They found that an additional year of education raised cognitive test scores by roughly 1 to 5 IQ points, and that the gains lasted across the life span.

Those studies measure extra years of education, not where the education happens. What they suggest is that sustained, structured instruction is one of the few reliable ways to raise measured intelligence. A homeschool that delivers a full, organized curriculum is plausibly getting the same benefit, and the Martin-Chang results fit that reading. A home setting where formal instruction is thin may not, but no study has measured IQ directly in either situation, so this remains an inference. For more on how far schools can move outcomes, see our article on whether effective schools can make every child academically proficient.

Children in school are sometimes referred for an intelligence test when teachers notice they are struggling or far ahead; our article on whether schools test for IQ explains how that usually works. Homeschooled children are not observed by classroom teachers in the same way, so parents who suspect giftedness or a learning difficulty usually have to start the process themselves, either privately with a licensed psychologist or by asking what their local school district offers. A psychoeducational evaluation combines an IQ test with achievement testing and is the usual way to sort out whether a child's difficulties reflect ability, instruction or a specific learning problem.

When reading any claim about homeschoolers' test scores, the two most useful questions are who chose to be tested and what comparison group was used.

There is no representative study showing that. Homeschoolers in volunteer samples score above average on achievement tests, but their families also have above-average education and income, and controlled studies find small or no differences.

On average it appears roughly neutral once family background is taken into account. Structured homeschooling has been linked to higher scores and unstructured approaches to lower ones, with a tendency toward stronger verbal than math results.

The best-known figures come from families who volunteered and paid for testing, often with parents administering the tests, and they were compared with national norms rather than with similar families.

No study has measured this directly. Education raises IQ scores by roughly 1 to 5 points per year, so a home setting with little sustained instruction could plausibly miss some of that benefit.

Parents can arrange an evaluation with a licensed or school psychologist, who will use an individually administered, normed test and can combine it with achievement testing.

Homeschooled students are not, on the current evidence, smarter because of homeschooling. The high scores in famous studies mostly reflect who chose to be tested and which families homeschool, and controlled comparisons find a small, uneven effect that depends heavily on how structured the teaching is. Since years of solid instruction reliably raise measured intelligence, the evidence points to the quality and amount of teaching mattering more than the setting. If you want a reliable reading on your own reasoning ability, an online IQ test built by psychometricians is a sensible place to start.

1. National Center for Education Statistics. (2022). Homeschooled children and reasons for homeschooling. Condition of Education. U.S. Department of Education, Institute of Education Sciences. [nces.ed.gov](. nces.ed.gov

2. Rudner, L. M. (1999). Achievement and demographics of home school students: 1998. Education Policy Analysis Archives, 7(8). [doi.org/10.14507/epaa.v7n8.1999](. doi.org

3. Kunzman, R., & Gaither, M. (2020). Homeschooling: An updated comprehensive survey of the research. Other Education: The Journal of Educational Alternatives, 9(1), 253-336. [web.archive.org](. web.archive.org

4. Martin-Chang, S., Gould, O. N., & Meuse, R. E. (2011). The impact of schooling on academic achievement: Evidence from homeschooled and traditionally schooled students. Canadian Journal of Behavioural Science, 43(3), 195-202. [doi.org/10.1037/a0022697](. doi.org

5. Green-Hennessy, S. (2014). Homeschooled adolescents in the United States: Developmental outcomes. Journal of Adolescence, 37(4), 441-449. [doi.org/10.1016/j.adolescence.2014.03.007](. doi.org

6. Yu, M. C., Sackett, P. R., & Kuncel, N. R. (2016). Predicting college performance of homeschooled versus traditional students. Educational Measurement: Issues and Practice, 35(4), 31-39. [doi.org/10.1111/emip.12133](. doi.org

7. Cogan, M. F. (2010). Exploring academic outcomes of homeschooled students. Journal of College Admission, 208, 18-25. [eric.ed.gov](. eric.ed.gov

8. Ritchie, S. J., & Tucker-Drob, E. M. (2018). How much does education improve intelligence? A meta-analysis. Psychological Science, 29(8), 1358-1369. [doi.org/10.1177/0956797618774253](. doi.org

Hero image: Globe and desk - needlework on display at Stevens County Museum, by Unhindered by Talent, CC BY-SA 2.0. Via Wikimedia Commons.