Conceptsign test
Sign Test
The sign test is a non-parametric statistical method used to assess whether a binary outcome occurs at a frequency that significantly deviates from a random baseline probability of 0.5. In NLP evaluation, it is commonly applied to verify the statistical significance of a model's directional judgment.
Timeline (last 90 days)
Sep 23
The authors apply a sign test to directional consistency, obtaining a significance level of p=0.00011.
Unverified50%