How the IQ test started and what it offers
IQ-Test.org has been online since 2010. The site started for a fairly personal reason: I wanted to take an IQ test online myself, but I was not particularly happy with what I found.
Some tests simply did not seem very convincing, others gave little information about how the result was calculated, and with many of them you only discovered at the end that you had to pay to see your score. I already knew that an online IQ test could not replace a professional psychological assessment. Still, I thought there was room for a freely accessible IQ test where you know what to expect before you begin.
That idea gradually developed into IQ-Test.org, and the test has been refined continuously ever since.
Today, it consists of 47 questions with a maximum time limit of 30 minutes. Once you finish, you immediately receive your result along with the solutions to all questions. You can also create a personal certificate free of charge. The test remains completely free to take and does not require registration.
Over time, IQ-Test.org has built up a database of more than 100,000 test completions. This provides a much stronger basis for statistical analyses of the test. For the continuously updated comparison norm, a filtered subset of suitable test completions is used. Following the current data cleaning process, around 43,500 historical test completions meet the criteria for inclusion in the norming sample.
At the same time, IQ-Test.org has grown into a broader knowledge resource where I write about intelligence, IQ testing and findings from the data collected by IQ-Test.org.
In this article, I want to explain how the test is structured, how the result is calculated and what can actually be learned from the available data. Just as importantly, I want to be clear about the limits of an online IQ test.
The IQ test on IQ-Test.org offers:
- The IQ test is completely free.
- The evaluation is based on data from more than 100,000 test sessions.
- No name or postal address is required.
- You receive the full evaluation and the solutions to all questions immediately after the test.
- You can download a personal certificate free of charge after completing the test.
47 questions covering different types of reasoning
The 47 questions place different demands on the participant. Number sequences are mixed with analogies, word relationships, geometric problems, matrices and other forms of logical reasoning.
Broadly speaking, the questions include numerical-logical, verbal, figural and spatial tasks. Some also involve knowledge or semantic classification. These categories simply describe the kinds of demands the questions place on the person taking the test.
One of the oldest and most consistent findings in intelligence research is that performance across different cognitive tasks tends to be related. Charles Spearman described this relationship at the beginning of the 20th century in terms of a general intelligence factor, commonly known as the g factor. Modern models of intelligence are more differentiated and consider both a general component and a range of more specific abilities. [1, 2]
The response data from IQ-Test.org also show that performance on many of the 47 questions is statistically related. This suggests that the overall score captures a shared component of cognitive performance. For that reason, IQ-Test.org reports one overall result rather than presenting individual task areas as separate IQ scores.
An internal data analysis shows how the questions within IQ-Test.org relate to one another statistically. To determine how closely the result corresponds to established professional intelligence tests, such as the Wechsler scales or Raven’s matrices, a separate external validation study would be required.
Test length and time limit
You have a maximum of 30 minutes to complete the test. The time limit gives everyone a comparable timeframe and keeps the overall test manageable. It cannot eliminate fatigue entirely, but it is intended to allow the test to be completed in one focused session.
When designing the test, the aim was to include enough different types of questions without making it unnecessarily long. Tests with very few questions generally provide less information about a person’s performance. Adding more questions only improves a test if those questions also contribute meaningfully to the measurement. Alongside test length, item difficulty, discrimination and the overall quality of the test are therefore important. [3]
How a raw score becomes an IQ result
In psychometrics, IQ scores are generally scaled so that they approximately follow a normal distribution, often illustrated by the familiar bell curve. IQ-Test.org uses the widely established IQ scale with a mean of 100 and a standard deviation of 15.
On a normally distributed IQ scale, around 68% of scores fall between 85 and 115. A mean of 100 and a standard deviation of 15 are widely used in modern intelligence testing.
The calculation begins with the number of questions answered correctly. If you solve 30 of the 47 questions correctly, for example, your raw score is 30.
To interpret that raw score, it is compared with other people who have taken the test. IQ-Test.org uses age-related comparison groups rather than placing all age groups into a single overall norm. This means the result is compared, as far as possible, with participants of a similar age.
The mean and standard deviation of the relevant comparison group are calculated from the available results. This makes it possible to determine how far an individual raw score lies above or below the average for that group. The result is then converted to the familiar IQ scale with a mean of 100 and a standard deviation of 15. A score of 115 is one standard deviation above the mean, while 130 is two standard deviations above it. [4]
The evaluation also shows an approximate percentile rank, calculated from the normal distribution. An IQ score of 130, for example, is around the 98th percentile. In practical terms, this means the result is approximately higher than that of 98% of the relevant comparison group.
Age-related comparison groups
The data show that average raw scores differ between age groups. Different cognitive abilities develop and change at different rates across the lifespan.
Hartshorne and Germine analysed large online samples alongside normative data from established cognitive tests. Depending on the ability being measured, peak performance occurred at very different ages. Some cognitive abilities reach high levels relatively early, while others continue to develop much later in life. [5]
The data collected by IQ-Test.org are not suitable for drawing a general curve of how human intelligence changes with age. Visitors choose for themselves whether to take the test, so they are not a randomly selected sample of the population.
Age differences are still relevant when evaluating results within the test itself. If 15-year-olds achieve different average raw scores from 50-year-olds, comparing someone with participants of a similar age is more meaningful than using one common norm across all age groups.
The analysis uses comparison windows of different sizes depending on age. Narrower age ranges are used for younger participants because average raw scores change more noticeably with age in these groups. In adulthood, several neighboring age groups are combined. For older participants, the comparison windows are broader because fewer suitable test completions are currently available. An age-specific norm is used only when at least 200 suitable test completions are available; otherwise, a broader comparison group is used.
Age is taken into account automatically when the comparison group is selected. This allows the age differences observed in the data to be considered when the raw score is interpreted.
The comparison norm keeps developing
The comparison norm used by IQ-Test.org is updated continuously. New test sessions are checked to determine whether they are suitable for inclusion in the norm. Very short attempts, as well as declared or likely repeat attempts, are among the cases that are not included.
Suitable results expand the comparison data and can therefore contribute to the evaluation of future test participants.
The free certificate
After completing the test, you can create a personal PDF certificate. Your name and result are only entered when you choose to generate the certificate; they are not required to take the IQ test itself.
The certificate documents your result on IQ-Test.org. It is not a psychological report and does not constitute an officially recognised certification of a professionally assessed IQ score.
How meaningful is an online IQ test?
A freely accessible online test is rarely taken under identical conditions. Some participants sit at a quiet desk, while others use a smartphone, get interrupted during the test or start when they are tired. It is also impossible to completely prevent the use of outside help or other resources.
Online cognitive research itself is nothing unusual. In 2012, Germine and colleagues compared self-selected online samples with traditional laboratory samples and found comparable patterns across several cognitive and perceptual tasks. [6]
This does not make a freely accessible online IQ test equivalent to professional psychological assessment. Diagnostic testing takes place under standardised conditions, and an appropriately qualified psychologist can consider not only the score but also how the test was completed and whether anything unusual occurred during the assessment.
My view of the IQ test
The analysis of the test data shows that performance on many of the questions is clearly related to the overall score. The factor analysis also revealed a strong common performance factor. A closer look at the data showed how strongly short-term repeat attempts can affect an unfiltered comparison group. Test sessions that do not meet the criteria for inclusion in the norm are therefore excluded from the comparison data.
If IQ-Test.org gives you a score of 128, the main conclusion is that your performance was clearly above the average of the age-related comparison group used for the evaluation. Without an external validation study, I cannot reliably say how precisely that value would translate to an IQ obtained through professional testing. The result also does not currently include an individual confidence interval.
For someone who is curious about how their performance on the IQ-Test.org questions compares with that of other participants, I consider the result a useful indication. If the result is needed for giftedness assessment, educational decisions or another situation that requires a professionally interpretable psychological result, a standardised test administered by an appropriately qualified psychologist is the better choice.
You can take the IQ test here for free. Your result, answers and certificate are available immediately after completion.
References
- [1] Spearman, C. (1904). “General Intelligence,” Objectively Determined and Measured. The American Journal of Psychology, 15(2), 201–292.
- [2] Savi, A. O., Marsman, M., van der Maas, H. L. J., & Maris, G. K. J. (2019). The Wiring of Intelligence. Perspectives on Psychological Science, 14(6), 1034–1061. DOI: 10.1177/1745691619866447.
- [3] American Educational Research Association, American Psychological Association & National Council on Measurement in Education. (2014). Standards for Educational and Psychological Testing. Washington, DC: AERA.
- [4] American Psychological Association. (2018). Wechsler Adult Intelligence Scale (WAIS). APA Dictionary of Psychology.
- [5] 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, 26(4), 433–443. DOI: 10.1177/0956797614567339.
- [6] Germine, L., Nakayama, K., Duchaine, B. C., Chabris, C. F., Chatterjee, G., & Wilmer, J. B. (2012). Is the Web as Good as the Lab? Comparable Performance from Web and Lab in Cognitive/Perceptual Experiments. Psychonomic Bulletin & Review, 19(5), 847–857. DOI: 10.3758/s13423-012-0296-9.