Wilson Interval
A method for estimating confidence intervals for binomial proportions. Compared with the normal approximation method, it performs more robustly with small samples and proportions near the boundaries (0 or 1), and is often used to judge statistical significance in A/B testing experiments.
Core Facts
Timeline (last 90 days)
Qwen 3 235B achieved a perfect score of 112/112, causing the normal approximation interval to collapse to [1.00, 1.00]; switching to the Wilson interval yields a true range of approximately 0.97 to 1.00.
The normal approximation interval (Wald interval) has its variance term approach zero when the sample proportion is close to 0 or 1, degenerating into a point and conveying a false sense of complete certainty; the Wilson interval still provides a reasonable uncertainty range at extreme proportions.
The 95% Wilson confidence interval is more accurate than the normal approximation (Wald interval) when sample sizes are small or when proportions are close to 0 or 1.
All Facts (3)
The 95% Wilson confidence interval is more accurate than the normal approximation (Wald interval) when sample sizes are small or when proportions are close to 0 or 1.
75%UnverifiedQwen 3 235B achieved a perfect score of 112/112, causing the normal approximation interval to collapse to [1.00, 1.00]; switching to the Wilson interval yields a true range of approximately 0.97 to 1.00.
50%UnverifiedThe normal approximation interval (Wald interval) has its variance term approach zero when the sample proportion is close to 0 or 1, degenerating into a point and conveying a false sense of complete certainty; the Wilson interval still provides a reasonable uncertainty range at extreme proportions.
50%