Hypothesis Testing-Is it hard?

Question

"A newspaper article claims that there are many more newborn females than males, although you had assumed that there were a 50-50 chance of being born male or female. Through the local hospital, you determined 628 baby boys and 718 baby girls were born over the past 2 month period."

What are Ho and Ha and how do I compute the appropiate statistic?

Answer 1

Hi Latifah! Hypothesis testing can be tricky, but if you're careful to follow the right test you can figure it out no problem.

The null hypothesis (H_O) generally states that there is no significant difference between our populations. In this instance, it would be that there is no difference between the number of males being born vs. females, and there is a 50% chance of either being born. Thus, H_O: p = 0.5

For a two-tailed hypothesis test, the alternative hypothesis is simply ≠the opposite of the null hypothesis. Thus, Ha: p≠0.5. We know this is a two-tailed test because there are no "direction" words in the post, i.e. no "fewer boys than girls", "fewer girls than boys", etc, just "is there actually a 50-50 chance of boy or girl.

This is a 1-proportion z test, so to calculate the test statistic use the equation:

z = (phat - p) / ( sqrt(p*q/n)). Here, phat would be the sample proportion (628/1346), p is the population proportion (0.5), q is the complement of p(1-0.5 = 0.5), and n is the sample size (1346).