In testing a method of gender selection, 40 couples are given a treatment designed to increase the likelihood of a girl, and each couple has one baby.
a. If the 40 babies include exactly 22 girls, would you consider this result statistically significant or would you attribute it to random fluctuations?
b. If the 40 babies include exactly 35 girls, would you consider this result statistically significant or would you attribute it to random fluctuations?
c. The P-value of 0.0000007 is obtained for the results of either part a or b. Which is it? Why?