Average IQ by Age Group
Average IQ by age group based on our large online dataset. Scores are percentile-based (mean 100, SD 15) and shown as age bands from 12 to 100.
In our dataset (N=78,251 validated), average IQ is highest in the Young Adult (19-35) (19-35) group and changes slowly across adulthood.
This page shows observed performance by age band in our dataset, which can differ due to sampling, device familiarity, pacing, and other real-world factors.
Method (30-second summary)
- Each result is converted from its percentile rank in the selected dataset view (Validated / All results) to an IQ-style scale (mean 100, SD 15). We then summarize those scores by age band.
- Validated results pass internal quality checks; "All results" includes unvalidated submissions and may be noisier.
- We show 95% confidence intervals for the mean. Smaller groups have wider intervals, especially the Senior band (N=1,200).
Method details (confidence intervals)
CI formula: mean +/- 1.96 * SD / sqrt(N).
SD is approximated from the interquartile range (p25-p75).
Data notes: Results exclude seed data and use percentile-based IQ scoring. Age groups are self-reported ranges from signup. For methods and caveats, see Methodology & limits.
- Mean shows the average; distribution shows the spread.
- Groups overlap a lot; differences are modest.
Mean IQ by age group
IQ distribution by age group (p10-p90)
"p" means percentile. The box spans the middle half of scores (25th to 75th). The whiskers show most scores (10th to 90th). The line in the box is the median (50th).
Groups overlap a lot - this chart shows the spread, not just the mean.
Validation rate by age group
Shows validated vs all results, plus validation rate.
Average time taken by age group
Older groups tend to take longer on the test, which can reflect pacing differences and strategy.
How to interpret age differences
- The mean chart shows the average; the distribution chart shows the spread (middle 80% of scores).
- The 95% CI column shows uncertainty around the mean; smaller groups have wider intervals.
- Scores are percentile-based and mapped to an IQ-style scale (mean 100, SD 15).
- Validated results pass quality checks; all results include unvalidated submissions.
- For details on scoring, sampling, and limitations, see Methodology & limits.
What does average IQ by age group mean?
This page reports observed averages by age band using a percentile-based IQ scale.
- The IQ scale is standardized (mean 100, SD 15); age bands are for reporting, not clinical norming.
- Observed performance in this dataset can differ by age because of sampling, device familiarity, pacing, and other real-world factors.
IQ across the lifespan (observed)
In our dataset, the highest average appears in the Young Adult (19-35) (19-35) band. Differences are modest and groups overlap heavily.
Online test performance reflects a mix of reasoning ability, familiarity with digital tests, and pacing. Age bands can differ on all three, so averages shift slightly.
Practice effects and sampling bias
The dataset is large but self-selected. Repeat test-takers and motivated participants can influence averages, so treat results as descriptive statistics rather than population norms.
Research context: cognition and age
Fluid abilities (processing speed, novel problem solving) tend to peak earlier, while crystallized knowledge can hold steady longer or peak later (Horn & Cattell, 1967).
Large-scale evidence shows different skills peak at different ages, so there is no single universal peak for cognition (Hartshorne & Germine, 2015).
Age-related change varies by domain and test format; normative IQ tests use age bands to keep the scale centered around 100 (WAIS-IV Manual, 2008).
For scoring details, see Methodology & limits. For representativeness and validation, see Dataset overview & sources.
References
- Wechsler, D. (2008). Wechsler Adult Intelligence Scale - Fourth Edition (WAIS-IV) Technical and Interpretive Manual. Pearson.Publisher page (age norming).
- Hartshorne, J.K., & Germine, L.T. (2015). When does cognitive functioning peak? The asynchronous rise and fall of different cognitive abilities across the life span. Psychological Science.DOI.
- Horn, J.L., & Cattell, R.B. (1967). Age differences in fluid and crystallized intelligence. Acta Psychologica.DOI.
- Salthouse, T.A. (2009). When does age-related cognitive decline begin? Neurobiology of Aging.PubMed.
Data table & download
Universaliqtest. Average IQ by age group (accessed 2026-01-17). https://www.universaliqtest.com/en/statistics/average-iq-by-age
<iframe src="https://www.universaliqtest.com/en/statistics/average-iq-by-age" width="100%" height="720" loading="lazy"></iframe>
| Age group | Age range | N | Average IQ | 95% CI | Median (p50) | p10 | p90 | Avg time | Last updated |
|---|---|---|---|---|---|---|---|---|---|
| Youth (12-18) | 12-18 | 21,530 | 98.8 | 98.6 to 99 | 98.5 | 79.7 | 116.4 | 32:50 | 2026-01-17 |
| Young Adult (19-35) | 19-35 | 32,750 | 102 | 101.8 to 102.2 | 102.8 | 83.5 | 122 | 41:08 | 2026-01-17 |
| Adult (36-65) | 36-65 | 22,771 | 98.5 | 98.3 to 98.7 | 98 | 79.3 | 116.9 | 42:52 | 2026-01-17 |
| Senior (66-100) | 66-100 | 1,200 | 95.1 | 94.1 to 96.1 | 92.7 | 72.1 | 113.2 | 55:05 | 2026-01-17 |
FAQ
Our dataset groups ages into bands (12-18, 19-35, 36-65, 66-100). We do not yet report single-year averages.
Ages 12-18 are grouped together as Youth. See the Youth row in the table for the current average.
Ages 19-35 are grouped together as Young Adult. See that row for the current average.
Ages 19-35 are grouped together as Young Adult in this dataset.
Age-normed IQ scores are designed to average around 100 at every age, so a "peak" here refers to observed performance in this dataset, not a clinical claim about peak ability. In our data, the highest average appears in one of the adult bands, but differences are modest and groups overlap.
We observe small differences across age bands, but this is not a clinical study and many factors influence test performance.
This page does not measure intervention effects. It summarizes test performance in a large online dataset.
No. These are online test results summarized for dataset statistics.
The dataset is large but self-selected; results should be interpreted as online test statistics, not population norms.
Yes. Education and other factors may differ across age bands, which is why we show distributions and caveats.
Time taken can reflect test strategy; we include average time by age to provide additional context.