Category:Yates's Correction for Continuity

This category contains results about Yates's Correction for Continuity.

Let $C$ be a contingency table with $2$ rows and $2$ columns.

Let a $\chi$-squared test for lack of association be applied to $C$.

Yates's correction is the operation of subtracting $0 \cdotp 5$ from the absolute value of each difference $\size {O_i - E_i}$ before squaring.

Hence this $\chi$-squared test is an approximation to Fisher's exact test.