Test Method and Score Calculation

This page explains how answers are converted into a score and what limitations to consider when interpreting it.

What the tasks assess

Visual matrices require you to identify relationships between elements and apply the rules you discover. This format primarily draws on nonverbal reasoning and the ability to solve new problems.

The language demands are lower than in vocabulary tests, but experience with abstract images, education, and familiarity with similar tasks can also affect performance. The absence of verbal questions does not make the test entirely independent of culture.

From correct answers to IQ

The calculation consists of several steps:

  1. Raw score. One point is awarded for each correctly solved task. Incorrect or skipped answers earn no points and carry no penalty.
  2. Position within the sample. The number of eligible results below your score is added to half the number of results with the same score. This is used to calculate your percentile.
  3. Conversion to the IQ scale. The percentile is converted using the inverse standard normal distribution function into an IQ score on a scale with a mean of 100 and a standard deviation of 15.

p = (L + 0.5 × E) / N

IQ = 100 + 15 × Φ⁻¹(p)

Here, L is the number of results with a lower score, E is the number with the same score, N is the size of the comparison sample, and Φ⁻¹ is the inverse standard normal distribution function. In the formula, p is expressed as a proportion between 0 and 1.

For example, the 50th percentile corresponds to an IQ of 100. This is a mathematical example, not a description of the distribution of visitors’ results. The proportion of correct answers is not the same as the percentile: correctly solving 40 tasks can place you at different percentiles in different samples.

Before conversion, the percentile is limited to the range of 0.0032–99.9968%, corresponding to approximately IQ 40–160. The limits of the displayed scale do not mean that measurement is equally accurate throughout this range.

What data are used for comparison

The comparison dataset is drawn from test attempts on the website that meet technical eligibility criteria. These criteria take into account factors such as the duration of each attempt and indications of repeat participation. Such checks reduce the influence of some unwanted records, but they cannot distinguish between all participants or identify all repeat attempts without error.

If an age range is specified and enough eligible results are available for that range, the calculation may use the corresponding age group. Otherwise, the overall sample is used. Comparison with an age group does not in itself demonstrate that validated age norms are available.

Why the score may change

As the comparison dataset is updated, the position of the same raw score within it may change. Changes to the images, solution rules, task order, or time limit also affect how comparable test attempts are.

When comparing results, it is important to consider the task version and calculation rules in effect when the score was calculated. A change in the score after recalculation does not necessarily mean that the person’s abilities have changed.

What the test establishes — and what it does not

The IQ-meter.com test is primarily intended to assess fluid intelligence — the ability to identify patterns, make logical connections, and solve new problems. Its result reflects performance on nonverbal reasoning tasks. The test is not designed to provide a separate assessment of crystallized intelligence — vocabulary, general knowledge, and accumulated knowledge — and does not provide a standalone assessment of creative, emotional, or social abilities.

When interpreting the result, the characteristics of the comparison sample must be taken into account. In open online testing, there is no guarantee that the sample accurately reflects the composition of the population: users choose whether to take the test, and some countries, age groups, and educational groups may be overrepresented or underrepresented. A large number of test attempts and the exclusion of repeat attempts do not, by themselves, eliminate this bias. Comparison with online test participants should therefore not automatically be treated as comparison with the population as a whole.

Taking the test online does not replace a specialized in-person assessment by a psychologist. During such an assessment, the psychologist selects a standardized assessment method suited to the purpose, checks that instructions are followed and tasks are completed independently, controls the conditions in which tasks are presented, and takes the participant’s condition and individual characteristics into account. This helps reduce the influence of extraneous factors and provides a professional interpretation of the result, supplemented, where necessary, by an assessment of other cognitive abilities.

Sources and further reading

ITC: Guidelines on Test Use

ITC: Assessment of Linguistically and Culturally Diverse Populations

These documents explain general principles of assessment. Links to them do not imply that IQ-meter.com has been certified or endorsed by the respective organizations.