Why one number is not exact
Performance varies with item sampling, day-to-day conditions, and ordinary measurement error. A reported composite therefore estimates a standing rather than revealing a perfectly fixed quantity.
Professional score reports commonly pair the point estimate with a confidence interval derived from a standard error of measurement or estimation.
How to read the interval
If a report gives a score of 104 with a 95 percent interval of 99–109, the interval should be considered when comparing the result with a cutoff or another score. It is not a prediction that 95 percent of future scores must fall there.
Wider intervals indicate less precision. Different composites can have different standard errors even within the same assessment.
Decisions near thresholds
Small differences should not be overinterpreted, especially when confidence intervals overlap. A rigid decision based on a single point can imply more precision than the test supports.
Qualified interpretation also considers subtest patterns, referral question, history, behavior, language, and the consequences of error.
Common questions
Frequently asked questions
Does the confidence interval mean my IQ changes within that range?
It mainly expresses uncertainty in the estimate; real performance can also vary across occasions and conditions.
Why do online tests often show no interval?
Some provide simplified feedback or lack enough published evidence to estimate precision responsibly.
References