# hypothesis test ,test statistic wrong!

• Mar 27th 2009, 08:02 PM
oxrigby
hypothesis test ,test statistic wrong!
Hi,q is: A margarine firm has invited 200 men and 200 women to see if they can distinguish margarine from butter. It is found that 120 of the women, but only 108 of the men can. Investigate whether there is any evidence of a sex difference in taste discrimination. I did x-b(200,120/200) converting to normal gaves me a test statistic 1.13,,,,i was just wandering if anyone knew how to get the test statistic they had in the back of the book of 1.21 in order to confirm there is no sex discrimination supposedly at the standard 5%? thanx!
• Mar 27th 2009, 08:17 PM
matheagle
Show me your work. I don't understand what x-b means.
This seems to be a straightforward test of the difference between two proportions.

$\displaystyle H_0: p_1=p_2$ vs. $\displaystyle H_a: p_1\ne p_2$

The test stat is $\displaystyle {\hat p_1-\hat p_2\over \sqrt{\hat p\hat q\biggl({1\over n_1}+{1\over n_2}\biggr)}}$

where $\displaystyle \hat p$ is the pooled sample estimate of the common p.

Use the two sided standard normal distribution for your rejection region.

I can't believe how nice that Tex came out.
• Mar 27th 2009, 08:33 PM
oxrigby
oh ok thanks for that the chapters are all messed up thats in the next chapter thanks and yeh it came out pretty gud:P ill let the post know if i get stuck on it when i get to it.
• Mar 28th 2009, 03:59 AM
oxrigby
x-b denoted binomially distributed sorry. I understand this now thanks.