Your result depends not only on the number of correct answers but also on how other participants performed on the tasks. The following sections explain how this comparison works and how your IQ score is calculated.
What the test measures
The test primarily assesses fluid intelligence — the ability to identify patterns, recognise logical relationships and solve new problems. It uses visual tasks that require you to identify the rule governing an image and select the missing piece.
This format depends less on language and vocabulary than verbal tests do. However, education, cultural experience and familiarity with similar tasks can also affect the result. The test does not provide separate assessments of general knowledge, memory, creativity, emotional abilities or social skills.
How points are counted
The test consists of 60 tasks with an overall time limit of 20 minutes. Each correct answer earns one point. Incorrect and skipped answers neither earn nor deduct points.
All tasks carry equal weight. There are no bonus points for finishing early: your score is the number of correct answers given within the time limit.
Why the number of correct answers is not yet an IQ score
The number of tasks solved does not, by itself, show how common your score is. For example, 40 correct answers could be a high score on a difficult test and a typical score on an easier one.
We therefore first determine where your score falls within the comparison sample — the results of other participants eligible for comparison. This position is then converted to the IQ scale.
Why we count all lower scores and half of the matching scores
If another participant scored fewer points, your score ranks above theirs. If your scores are equal, there is no basis for ranking one participant above the other. All participants with the same score are therefore assigned a position at the midpoint of their group.
Suppose you answered 40 tasks correctly and the comparison sample contains 1,000 results:
| Score | Number of results |
|---|---|
| Below 40 points | 600 |
| Exactly 40 points | 200 |
| Above 40 points | 200 |
The results below your group account for 60% of the sample. Including your group brings this to 80%. The midpoint between these boundaries is 70%. This is the position used in the calculation: the 70th percentile.
Counting only lower scores would place everyone with the same score at the lower boundary of their group. Adding all matching scores would place them at the upper boundary. Adding half places them in the middle.
Calculation formulas
First, we determine the position within the sample:
p = (L + 0.5 × E) / N
Here:
Lis the number of eligible results with lower scores;Eis the number of eligible results with the same score;Nis the total number of results in the comparison sample;pis the position within the sample, expressed as a proportion from 0 to 1. Multiplying it by 100 gives the percentile rank.
In the example above:
p = (600 + 0.5 × 200) / 1,000 = 0.7
We then convert this to the IQ scale:
IQ = 100 + 15 × Φ⁻¹(p)
The function Φ⁻¹ converts the position within the sample to the corresponding point on the standard normal distribution. The value 100 sets the centre of the scale, while 15 sets its standard deviation.
For example, the 50th percentile corresponds to an IQ of 100, and the 70th percentile to an IQ of approximately 108. However, the 70th percentile does not mean 70% correct answers: these are different measures.
Before conversion to IQ, the value of p is limited to the range 0.000032–0.999968, corresponding to approximately IQ 40–160. These are the limits of the displayed scale, not a guarantee of equal measurement accuracy throughout the range.
Who your result is compared with
The comparison sample consists of test attempts completed on the website. Selection takes technical criteria into account, including completion time and signs of repeat attempts.
Filtering reduces the influence of unsuitable records but cannot detect every repeat attempt: some users may bypass the restrictions.
If an age range is specified for the calculation and it contains enough eligible results, the corresponding age group is used. Otherwise, the overall sample is used. This comparison takes participants’ age group into account, but does not in itself establish standardised age norms.
Visitors choose for themselves whether to take the test. Countries, age groups and educational groups may therefore be unevenly represented. A large dataset and filtering do not guarantee representativeness: comparison with the website’s participants cannot automatically be equated with comparison with the general population.
Why the same number of points can produce different IQ scores
As more results are added to the sample, the position of a given score may change. The same number of correct answers may therefore correspond to a different IQ score when recalculated.
The task version, presentation order and time limit also matter for the comparison. A change in the estimate after recalculation does not necessarily mean that a person’s abilities have changed.
How to use your result
The result shows how you performed on nonverbal reasoning tasks relative to the comparison sample. Fatigue, attention, stress and testing conditions can affect performance.
A professional assessment of intelligence involves an in-person evaluation by a psychologist using an appropriate standardised assessment method. The specialist controls the testing conditions, checks that tasks are completed independently, takes the participant’s condition into account and assesses other cognitive abilities when necessary.
Sources and further reading
ITC: assessing linguistically and culturally diverse populations
These documents explain general principles of assessment. Linking to them does not imply certification or endorsement of IQ-meter.com by the respective organisations.