# hypothesis testing #2

• Nov 12th 2010, 02:25 PM
wik_chick88
hypothesis testing #2
people with chronic kidney disease require hemodialysis, where their blood is passed through a machine that does the work of their kidneys. unfortunately blood clots can form around the connection to the machine and cause problems if they then lodge elsewhere in the body. a study of the effect of aspiring in helping reduce the formation of blood clots involved 44 subjects undergoing dialysis. of the 19 people receiving aspirin, 6 developed blood clots; of the 25 people receiving placebo 18 developed blood clots.

perform a hypothesis test to determine if there is a reduction in the rate of blood clots with aspirin. include a statement of \$\displaystyle H_{0}\$ and \$\displaystyle H_{1}\$, the test statistic and its distribution under \$\displaystyle H_{0}\$.
• Nov 12th 2010, 10:02 PM
CaptainBlack
Quote:

Originally Posted by wik_chick88
people with chronic kidney disease require hemodialysis, where their blood is passed through a machine that does the work of their kidneys. unfortunately blood clots can form around the connection to the machine and cause problems if they then lodge elsewhere in the body. a study of the effect of aspiring in helping reduce the formation of blood clots involved 44 subjects undergoing dialysis. of the 19 people receiving aspirin, 6 developed blood clots; of the 25 people receiving placebo 18 developed blood clots.

perform a hypothesis test to determine if there is a reduction in the rate of blood clots with aspirin. include a statement of \$\displaystyle H_{0}\$ and \$\displaystyle H_{1}\$, the test statistic and its distribution under \$\displaystyle H_{0}\$.

Please post what you have done tried.

CB
• Dec 15th 2010, 06:15 AM
wik_chick88
all i have so far is \$\displaystyle H_0: p = 0.72\$ (which is 18/25) and \$\displaystyle H_1: p < 0.72\$
where do i go from here?
• Dec 20th 2010, 04:33 PM
ANDS!
This is from awhile ago, but presumably you'll still be working on this - your hypothesis isn't correct. You aren't performing a test on what a population proportion is - your interested in whether or not there is a significant different between those receiving the placebo and those taking aspirin. Thus your hypothesis should be to assume there is NO difference (Placebo-Proportion = Aspirin-Proportion), and find the probability of observing a difference as large as the one observed (Placebo-Proportion > Aspirin-Proportion).

I would review notes on testing differences between population proportions, and the applicable tests involved.