2
$\begingroup$

Question:

18 men and 12 women have agreed to participate in a randomized controlled experiment. A simple random sample of 15 of these 30 people is chosen as the treatment group; the others form the control group.

a) Find a decimal value for the chance that the treatment group contains more women than men.

b) Find the chance that all the women are in the same group.

For a) I chose the hypergeometric distribution as it is sampling without replacement. I set it up in the following way:

$\frac{\binom{12}{8} \cdot \binom{15}{7}}{\binom{30}{15}} + \frac{\binom{12}{9} \cdot \binom{15}{6}}{\binom{30}{15}} + ... + \frac{\binom{12}{12} \cdot \binom{15}{3}}{\binom{30}{15}}$

Is this a correct approach to the problem?

for b) I am confused why it can't be $\frac{\binom{12}{12} \cdot \binom{15}{3}}{\binom{30}{15}}$. I think the $\frac{1}{2}$ should also be taken into account for choosing between either the Treatment Group or Control group, but I don't know how to use this information.

$\endgroup$

1 Answer 1

0
$\begingroup$

Your approach for a) is correct.

As for part b),you must multiply by 2 instead of $\frac{1}{2}$. Another way to calculate this is $2\cdot\frac{\binom{18}{15}}{\binom{30}{15}}$, i.e. by looking at the 15 men in the other group.

$\endgroup$
2
  • $\begingroup$ does the 2 only come from the fact that it could be in one group OR the other? $\endgroup$
    – Jayant.M
    Feb 7, 2017 at 6:24
  • $\begingroup$ Right, this is the reason! $\endgroup$
    – sdd
    Feb 7, 2017 at 7:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.