Statistics Problem, how to do this?

You already asked this question. Like I did before, I'll copy and paste what I already answered. You can't use binomial coefficient for hypothesis testing.

"Okay, for this I am going to do a 2 sample proportion test because we are trying to find out if there was a significant change in the proportion of people who died in each test group. Proportion 1 (p1) is the proportion of people who died in the treated group, or .1. Proportion 2 (p2) is the proportion of people who died in the control group, or .5. Our hypotheses are:

Ho:p1=p2 (this is our control hypothesis)

Ha:p1<p2 (this hypothesis says the treatment works)

We are going to give ourselves an a (alpha) of .01 because, since this is a medical test, we want to be pretty sure about our results. Usually people use either .01 or .05 for a.

If you have a graphing calculator you should be able to do these next steps on it, but I'll type out the formula just in case. For the formula, p* is the difference between our p1 and p2, or -.4, n1 and n2 are both 10 because each group tested had 10 people.

Z=p*/√(p*(1-p*)((1/n1)+(1/n2)))

I did my calculations on my calculator, so you might have gotten a number slightly different than me, but the value the calculator gives is -1.95. when we find this number in a p value table, it gives us a p value of .025

Because the p value is greater than a (alpha), I fail to reject the null (Ho) and conclude that p1 could equal p2. Therefore, the results of the treatment do not appear to be statistically significant, and the treatment is not statistically effective."