How IQ tests work
An IQ test is a set of reasoning puzzles with a scoring system designed so that results can be compared between people. Here is what happens behind the number.
Step 1: reasoning puzzles
IQ tests ask questions that need reasoning rather than memorised facts: finding the rule in a pattern, continuing a number series, picturing a shape turned around, working out what must follow from some statements, or choosing the word that fits.
Good tests mix several types, because people tend to be stronger at some of them than others. A mix gives a more balanced picture than any single type.
Step 2: scoring
Your raw score is based on how many questions you answered correctly. Harder questions are usually worth more, because getting a difficult one right says more about your reasoning than getting an easy one right.
Step 3: comparing with a norm group
The raw score is then compared with the scores of a large group of people who took the same test. This group is called the norm group. Your position in it is converted to the IQ scale, where 100 is the group average and 15 points is one standard deviation.
This step is what turns "I got 21 right" into "my IQ is about 112". Without a norm group, a test can only report a raw score, not a real IQ.
Why it is an estimate
For these reasons we show a likely range around the estimate rather than a single exact figure.
- Short tests have few questions, so one lucky guess or one careless slip moves the result.
- Tiredness, noise, distraction and unfamiliar question types change how you perform on the day.
- The norm group for a new online test is small at first, and it improves as more people take it.
Clinical tests are different
Tests such as the Wechsler scales are given one-to-one by a trained psychologist, take one to two hours and have been checked against thousands of people. An online test, including this one, cannot replace that and should not be used for medical, educational or hiring decisions.