Guide
What IQ predicts, and what it does not
IQ scores are linked to school results, job performance, education, job level and income. The links are real and some are strong. They are also correlations across large groups, which means they describe tendencies, not destinies. This guide gives the main numbers and shows how to read them without overreaching.
Want your own number? The test is 25 visual puzzles, about 12 minutes, no sign-up.
First, what a correlation means
Every finding in this guide is a correlation: a number from -1 to 1 that describes how closely two measures move together across a group of people. At 0 there is no straight-line link. At 1 the link is perfect: rank people on one measure and you have ranked them on the other. Real findings sit in between.
A rough guide to reading the strength:
- Around 0.2: a modest link. People differ hugely, with lots of overlap between groups.
- Around 0.5: a clear, moderate link, with many exceptions.
- Around 0.8: a strong link, though still not a rule for individuals.
Two cautions apply to every number below. A correlation describes a group, not a person. And it does not by itself show cause: two measures can move together because one affects the other, because both depend on something else, or both.
School results
The strongest finding comes from school. In a study of about 70,000 English children, general intelligence measured at age 11 correlated 0.81 with total GCSE exam points at age 16 (Deary, Strand, Smith & Fernandes, 2007). That is a strong link, and it makes sense: school exams reward learning new material, reasoning with it and recalling it under time pressure, which overlap with what intelligence tests sample.
Even at 0.81, the link is not destiny. Children with the same score at 11 went on to a range of exam results at 16. Teaching, effort, attendance, health and circumstances all left room for different outcomes.
Try one · Pattern Recognition
Which tile comes next?
Tap the option you think is right, or show the answer.
Show the answer
The missing tile shows four hatched crosses (option 4).
How to solve it
- Compare neighboring tiles and name what changes each time.
- From tile to tile, one more cross is added.
- The fill cycles hatched, outline and solid, then starts again.
- Everything else stays the same from tile to tile.
- The last tile shows three solid crosses. Apply each change once more.
- So the missing tile shows four hatched crosses.
That was 1 puzzle. The full test has 25.
Job performance
A landmark review of 85 years of personnel research estimated the validity of general mental ability for predicting job performance at 0.51 (Schmidt & Hunter, 1998). In plain terms: across many jobs and studies, people who scored higher on general ability tests tended, on average, to be rated as better performers. A validity of 0.51 is a moderate link with plenty of exceptions in both directions.
This finding concerns properly administered, validated tests used in research. It says nothing about whether a short online test is fit for hiring. IQ Octopus is not, and cannot be used for employment screening. If you face a cognitive test as part of an application, our page on IQ tests vs aptitude tests explains how the two differ.
Education, job level and income
A 2007 meta-analysis pulled together many studies of intelligence and later life outcomes (Strenze, 2007). It reported correlations of about:
| Outcome | Correlation with intelligence |
|---|---|
| Education (how far people went) | about 0.56 |
| Occupational level | about 0.45 |
| Income | about 0.23 |
Notice how the strength falls as the outcome gets further from the classroom. Education is moderately linked. Income, at about 0.23, is a modest link: it is real across large groups, but knowing someone's score tells you little about what they will earn. Income depends on field, location, family, health, choices, timing and luck, among much else.
What these numbers cannot tell you
- Your future. Group correlations cannot be turned into a forecast for one person. Plenty of people sit far from the trend line in both directions.
- Why the links exist. Correlations alone do not separate cause from shared background. For example, the heritability of intelligence rises from about 20% in infancy to perhaps 80% in later adulthood (Plomin & Deary, 2015), so family background is entangled with both scores and outcomes in ways a correlation cannot untangle.
- What matters to you. None of these studies measures kindness, creativity, judgment, happiness or a good life.
- Anything about a short online estimate. The research used established tests. It does not transfer automatically to any other test, including ours.
Why the research still matters
It is fair to say that general reasoning ability, as measured by good tests, is one of the better-studied predictors of how groups of people fare in school and at work. It is also fair to say that it leaves most of the variation between individuals unexplained. Both statements are true at once, and a careful reading keeps both.
That is the spirit in which IQ Octopus reports a result: an estimated IQ, with a confidence range, describing where you sat compared with other people on one set of puzzles on one day. For how to read such a result, see how to read an IQ score. For the short answer to a common question, see does IQ predict success? And for what tests sample in the first place, read what an IQ test measures.
Sources
- Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence, 35(1), 13-21. doi.org/10.1016/j.intell.2006.02.001
- Schmidt, F. L., & Hunter, J. E. (1998). Psychological Bulletin, 124(2), 262-274. doi.org/10.1037/0033-2909.124.2.262
- Strenze, T. (2007). Intelligence, 35(5), 401-426. doi.org/10.1016/j.intell.2006.09.004
- Plomin, R., & Deary, I. J. (2015). Molecular Psychiatry, 20, 98-108. doi.org/10.1038/mp.2014.105