Two-tailed test

we want to know if the difference in the population proportions $p_1-p_2$ is significantly different from 0 , in either direction.

classic method: If the $z$ -statistic (cell B22) is less than $-z_{\alpha/2}$ (cell B23) or greater than $+z_{\alpha/2}$ (cell B21), reject the null hypothesis. Otherwise, accept it (the appropriate conclusion appears in cell B24).