The best available estimate puts the average IQ of programmers at about 111 — around the 77th percentile — based on 181 programmers and software developers in the UK Understanding Society survey (Wolfram, 2023). Developers vary widely, and plenty of good programmers score in the average range.
The data on software developers
In Wolfram’s (2023) analysis of the UK Understanding Society survey, programmers and software development professionals (N = 181) had an estimated mean IQ of 111.2. With 181 people the estimate is reasonably precise (95% uncertainty of the mean about ±1.9 points). It places developers close to accountants (110.3), teachers in higher education (111.4) and medical practitioners (112.7).
The survey was conducted between 2011 and 2013, before the latest wave of bootcamps and AI-assisted coding broadened who works in software. The group also mixes very different roles — web developers, embedded engineers, data engineers, testers — and we could not find reliable, representative IQ data broken down by these specialisations.
Programmer IQ range
Assuming the typical spread inside occupations (SD about 13), the middle two-thirds of developers would score between about 98 and 124:
| IQ range | Programmers (model) | General population |
|---|---|---|
| below 90 | ~5% | 25% |
| 90–99 | ~14% | 25% |
| 100–109 | ~27% | 25% |
| 110–119 | ~29% | 16% |
| 120–129 | ~18% | 7% |
| 130 or above | ~7% | 2% |
Model: normal distribution around the estimated mean with SD 13, the average within-occupation spread implied by Wolfram (2023), where roughly three-quarters of IQ variance lay within rather than between occupations. Real distributions can be skewed.
Is coding ability the same as IQ?
Not quite. Programming draws on abstract reasoning, working memory and the ability to decompose problems — all related to general cognitive ability. But the UK survey battery used to estimate these figures (word recall, number series, everyday maths, serial subtraction, verbal fluency) does not test spatial reasoning or formal logic in depth, so it may not capture the abilities developers use most. Day-to-day software work also depends on accumulated knowledge of tools and codebases, persistence while debugging, reading other people’s code, and collaborating through reviews.
Across occupations, general mental ability predicts job performance with a validity of roughly .31 in the most recent meta-analytic estimates (Sackett et al., 2022), down from the .51 long quoted from Schmidt and Hunter (1998). Work-sample tests — for programmers, realistic coding exercises — are themselves good predictors, which is partly why technical interviews focus on them rather than on IQ.
Do you need a high IQ to learn programming?
No. Learning to program is a skill that improves with practice, feedback and good teaching. People at the average range regularly become competent developers; the data above imply that roughly a fifth of working developers score below 100. If you enjoy breaking problems into steps and are willing to stay stuck on a bug for an hour, those traits matter more than a test score.
How to read these numbers
Every number on this page comes with three caveats. First, the UK estimates are based on a short survey battery — word recall, number series, everyday numeric problems, serial subtraction and verbal fluency — rather than a full clinical IQ test, so they are less precise for individuals and tend to compress group differences. Second, they describe people working in an occupation at one time (2011–2013), not what the job requires. Third, they are averages: Hauser’s analysis of the Wisconsin Longitudinal Study found that the correlation between occupation and measured ability was only moderate (about .39–.44), and that the spread of ability was almost as wide in high-status occupations as in others.
Sources
- Wolfram, T. (2023). (Not just) intelligence stratifies the occupational hierarchy: Ranking 360 professions by IQ and non-cognitive traits. Intelligence, 98, 101755. Occupation figures as tabulated in J. Thompson, “Minds and professions” (2023).
- Hauser, R. M. (2002). Meritocracy, cognitive ability, and the sources of occupational success. CDE Working Paper 98-07, University of Wisconsin–Madison (Wisconsin Longitudinal Study).
- Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology. Psychological Bulletin, 124(2), 262–274.
- Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection. Journal of Applied Psychology, 107(11), 2040–2068.
Frequently asked questions
What is the average IQ of programmers?
About 111 for UK programmers and software development professionals (Wolfram, 2023, N = 181), roughly the 77th percentile.
Do software engineers have a higher IQ than other engineers?
The verified data cannot separate them reliably; electrical engineers scored higher in a tiny sample of 9, so differences are uncertain.
What IQ do you need to be a programmer?
None is required. Many competent developers score in the average range; practice and persistence matter a great deal.
Do tech companies use IQ tests?
Most rely on work-sample coding exercises and structured interviews rather than IQ tests, partly because work samples predict job performance well.