Four calculators for making sense of a psychometric result you already have. No sign-up, no email, and every one of them shows its working.
These free IQ and personality tools convert a score you already hold into something you can actually read: a percentile, a rarity figure, an equivalent on a different test, or a high-IQ society threshold. Every calculation runs in your browser using the normal distribution, and none of them ask for an email address.
If you do not have a score yet, start with the free IQ test or the free personality test. Everything else on this page needs a number to work from.
Turns a raw score into a percentile, a rarity figure and a classification band, with a bell curve showing where you sit.
AnswersWhat percentile is my IQ score?Converts a score between the Wechsler, Stanford-Binet and Cattell scales, which all report very different numbers for identical ability.
AnswersWhat is my Cattell score on the Wechsler scale?Checks a score against the entry threshold for Mensa and six other high-IQ societies, using the correct cutoff for your test.
AnswersDoes my score qualify me for Mensa? Take the testDo not have a score yet? Get one across five cognitive domains in about 15 minutes, with your percentile shown before any payment.
AnswersWhat is my IQ in the first place?Compares any two of the 16 types across all four preference axes and shows which differences balance a pair and which create friction.
AnswersHow do my type and theirs actually fit? Take the testNot sure of your four letters? Self-typing from descriptions tends to pick the type you admire rather than the one you use.
AnswersWhich of the 16 types am I?All three IQ tools take the same input, a score plus the scale it came from, and answer different questions with it. This is the quickest way to pick.
| If you want to know | Use |
|---|---|
| How many people I scored above | IQ Percentile Calculator |
| What my score equals on a different test | IQ Scale Converter |
| Whether a supervised test is worth booking | Mensa Eligibility Checker |
| How two personality types fit together | MBTI Compatibility Checker |
| What my score means in plain terms | How to interpret your IQ score |
All three IQ tools run the same two steps, and neither involves a judgment call.
First the score becomes a z-score, which is how many standard deviations it sits from the average of 100. Subtract 100, then divide by the standard deviation of the test that produced it. Second, that z-score is read against the normal distribution to give a percentile, or rebuilt on another scale by multiplying by that scale's standard deviation and adding 100 back.
The z-score is the part that actually travels between tests. It is why the percentile stays identical across every scale while the raw numbers do not, and why 130 on a Wechsler, 132 on an older Stanford-Binet and 148 on a Cattell all describe exactly the same standing.
People trade raw IQ numbers as though they mean the same thing everywhere. They do not. A raw score is a position on one particular scale, and that scale was a design decision made by the test publisher.
The percentile is different. It is a statement about people rather than about the test: it says what share of the comparison group you outperformed. That statement survives any change of instrument, edition or country. When somebody quotes a number without naming the test, the percentile is the question that makes the number usable again.
This is why every tool on this page reports a percentile alongside whatever else it calculates, and why our guide on how to interpret an IQ score leads with percentile rather than the headline figure.
Every modern intelligence test sets the population average at 100. That much is shared. What differs is how widely each test spreads scores around that average, a figure called the standard deviation.
On the Wechsler scales, which include the WAIS for adults and the WISC for children, one standard deviation is 15 points. The modern Stanford-Binet Fifth Edition also uses 15. Older Stanford-Binet forms use 16. The Cattell III B, still used by British Mensa, uses 24.
Someone who performs exactly two standard deviations above average scores 130 on a Wechsler, 132 on an older Stanford-Binet and 148 on a Cattell. Same person, same day, same ability, three numbers spread 18 points apart. Nothing has changed except the ruler.
Whether the number came from a school record, a clinical report or an online test, the same short checklist turns it into something meaningful.
Most arguments about IQ scores online come from skipping one of those four steps. Run through them and the disagreement usually disappears.
Where these tools stop being reliable. The arithmetic is exact, but the norms behind it are not. Standardisation samples contain very few people above roughly 145, so percentiles out there rest on a handful of cases. Test norms also drift over decades, which means converting between tests standardised years apart hides a real difference in the comparison group.
A converted or calculated figure is a good estimate of where you stand. It is not a score you can say you earned on that other test.
Four situations account for almost everyone who lands here, and each one maps to a different tool.
The number on its own tells you very little. Run it through the percentile calculator to find out what share of people you actually outscored, and read the classification band it falls into. That takes about ten seconds and is the single most useful thing you can do with a fresh result.
Older records frequently come from Stanford-Binet forms using a standard deviation of 16, which reads slightly differently from a modern Wechsler result. The scale converter puts the two on the same footing. Check the date as well, because norms from decades ago were set against a different reference group.
A proctored sitting costs money and usually means travelling to a testing centre. Knowing roughly where you stand beforehand is a reasonable way to decide whether that is worth doing now or later. The eligibility checker shows the cutoff for your specific test rather than quoting one universal figure.
The compatibility checker is built for the conversation that starts when two people notice they process things differently. It names which of the four preference axes you actually differ on and what that difference tends to look like day to day.
Every tool here works fully without an email address, an account or a payment. There is no step where a result gets held back, and nothing you type is sent to a server.
That is a deliberate choice. A calculator that demands an email before showing a percentile is not really a calculator, it is a lead form wearing one. The tools are free because they are useful on their own, and because people who find them useful tend to want a properly scored result afterwards.
What none of them can do is qualify you for anything. Mensa and every comparable society accept only supervised, proctored tests administered by an approved professional. Our eligibility checker tells you whether booking one is likely to be worth the fee, which is a genuinely useful thing to know beforehand and a completely different thing from an application.
The same honesty applies on the personality side. No research shows that a type pairing predicts whether a relationship lasts. What type describes reasonably well is communication style, so the compatibility checker is built on that and says so plainly rather than dressing a preference profile up as a forecast.
A percentile is a starting point, not a conclusion. Three things are worth doing with it.
Read the domain breakdown rather than the composite. A full-scale score is an average of several distinct abilities, and averaging is exactly the operation that hides the interesting part. Two people can both score 110 while one is strong verbally and slow on processing speed and the other is the reverse. Those two should make different decisions about how they study and what kind of work suits them.
Treat the figure as a band. Differences of a few points are measurement noise rather than signal. If two results sit within about ten points of each other, they are not really disagreeing.
Keep it in proportion. General cognitive ability correlates with academic achievement at roughly 0.5, and with job performance at around 0.3 following a major 2022 re-analysis by Sackett and colleagues that corrected earlier and higher estimates. A correlation of 0.3 leaves the large majority of variation in performance unexplained by ability alone. Conscientiousness, accumulated domain knowledge, interest and opportunity all carry real weight.
If you want the long version of that argument, how to interpret your IQ score and are online IQ tests accurate cover it properly.
Each tool answers one narrow question. These go into the reasoning.
The same handful of errors come up constantly, and each one is easy to avoid.
Quoting a number without the scale. "I got 132" is not a complete statement. On an older Stanford-Binet that is the Mensa cutoff exactly. On a Cattell it is well short. Always record which test produced a score, ideally alongside the date.
Treating a percentage as a percentile. These get confused constantly. A percentage is the share of questions you answered correctly. A percentile is the share of people you outperformed. Answering 60 percent of a hard test correctly can easily put you in the 90th percentile.
Retaking the same test to chase a higher number. Repeated exposure to similar items produces a practice effect of roughly three to five points. That is familiarity with the format, not a change in ability, and it fades on a genuinely unfamiliar test.
Reading a tail figure as precise. Standardisation samples contain very few people above about 145, so percentiles out there rest on a handful of cases and most tests run out of hard enough items anyway. A difference of five points near 100 is meaningful. The same five points near 150 is mostly noise.
Both tests are free to take and show your result before any payment. Pick whichever question you came here with.
