Yate s' continuity correction: When the testing for independence in contingency table, a continuous probability distribution, known as chi-squared distribution, is used as the approximation to the discrete probability of all the observed frequencies, called as the multinomial distribution. To improve the approximation Yates suggested a correction which involves subtracting 0.5from positive discrepancies (observed - expected) and adding 0.5 to the negative discrepancies prior to these values are squared in the calculation of the usual chi-square statistic. If the sample size is large correction will have less effect on the value of test statistic.