Glossary
Confidence interval
Definition
A confidence interval is a range of scores that is likely to contain your true score, given the error in the test. A 95% interval of 110 to 120 means that, if the test were repeated many times, ranges built this way would capture the true score about 95% of the time.
Want your own number? The test is 25 visual puzzles, about 12 minutes, no sign-up.
Why it matters for IQ scores
A single number hides the uncertainty in any measurement. Clinical full-scale IQ scores carry a standard error of a few points, so test manuals report a confidence interval, often around plus or minus 5 points at 95% (Wechsler, 2008). A measured 112 is better read as “somewhere around 107 to 117”.
An example
Two people score 118 and 121. Their intervals overlap heavily, so the test cannot say with confidence that the second person is more able. The same goes for scores either side of a label boundary such as 120, where the WAIS-IV Superior band starts. Read the range, not just the point.
Common confusions
A 95% interval does not mean there is a 95% chance your true score is inside this particular range in a strict sense; it describes how often the method captures the truth. It also does not cover every source of error, such as being tired or distracted on the day. On IQ Octopus the range comes from the Rasch model's standard error around your theta. See standard error of measurement, reliability and how to read an IQ score.
Try one · Quantitative Reasoning
Which number comes next?
5, 60, 10, 63, 15, ?
Tap the option you think is right, or show the answer.
Show the answer
The next number is 66 (option 2).
How to solve it
- The gaps jump around, so look at every other number.
- The 1st, 3rd, 5th numbers (5, 10, 15) go up by 5 each time.
- The 2nd, 4th, 6th numbers (60, 63) go up by 3 each time.
- The missing number continues the second series: 63 + 3 = 66.
That was 1 puzzle. The full test has 25.
Sources
- Wechsler, D. (2008). WAIS-IV Technical and Interpretive Manual. Pearson.