test statistic and confidence intervals

This is the question i am having issues with, if anyone is keen to help, i'd really appreciate it.

Cheers

Two groups, A and B, consist of people with a disease. Each group was randomly selected from a large number of patients. A serum is given to Group A, while Group B receives a placebo. Otherwise the two groups are treated identically. In group A, 74 out of 108 people recover from the disease, while in group B, 61 out of 113 recover.

(a) Using a test statistic which has a normal distribution, test the hypothesis that the serum helps to cure the disease. Assume a 5% significance level.

(b) Calculate a 95% confidence interval for the difference in recovery rates between serum and placebo.

for a) the test stat i would use is:http://upload.wikimedia.org/math/d/f...58f83ac27b.png

orhttp://upload.wikimedia.org/math/f/1...e12c18127d.png

and for b) i would use the 95% which is 1.96

and the formula for that would be

CI=(ρ ̂_1-ρ ̂_2)±k√((ρ ̂_1 (1-ρ ̂_1))/n_1 +(ρ ̂_2 (1-ρ ̂_2))/n_2 )

Anyone?

Re: test statistic and confidence intervals

A Statistician / psychologist can help you with your statistics in spss, amos, statistica, Econometrics, minitab, stata, r, gpower. I can do factorial analysis, correlation, time series, path analysis, anova, ancova, glm, non parametric reliability, standardization, validity, mediation and moderation analyses can be done. We can also evaluate questionnaires. Private lessons are offered also using skype. estatistics. email: info@estatistics.eu.