Question 3 (6 points) There are four different clinical sites in a small city, patients in each site are enrolled and randomly receive either the experimental medication designed to lower blood pressure or placebo, then the patients blood pressure measured after 5 hours. Data is given in the following table: Treatment Site 1 Site 2 Site 3 Site 4 Hypertensive 27 56 45 62 No Hypertensive 58 69 53 19 a. [2] Test if there is a difference in the proportions of hypertensive patients across clinical sites. Run the Chi-Squared test at a 5% level of significance and clearly state hypotheses, test statistic and conclusion. b. [4] Fit independent and saturated loglinear models for these data. Is there sufficient evidence to indicate that there is a relationship between treatment and site? Compare two models by "anova"