The Study
Forty-three people were polled from random message boards on the internet. 9 different message boards were used, and each member was selected at random. Each person received a private message either via the message board's own mail system or their listed AOL Instant Messenger screen-name. They were asked two questions: whether or not they favored legalizing marijuana and whether or not they favored legalizing prostitution.
1. Let N1 represent the proportion of people who favor legalization of marijuana, and N2 represent the proportion of people who favor legalization of prostitution.
2. Ho = N1 is equal to N2
3. Ha = The proportions are different
4. Significance level a=.05
5. Minitab output for two-prop t-test:
Estimate for p(Prostitution) - p(Marijuana): -0.193548
95% CI for p(Prostitution) - p(Marijuana): (-0.434435, 0.0473377)
Test for p(Prostitution) - p(Marijuana) = 0 (vs not = 0):
Z = -1.57 P-Value = 0.115
6. Since the P-value is so high, we cannot reject the null hypothesis at any reasonable level of significance.
7. There is insufficient evidence to make a claim that either proportion is greater than the other.