1.      A study was conducted to evaluate the effectiveness of a work site health promotion program in reducing the prevalence of cigarette smoking. Thirty-two work sites were randomly assigned either to implement the health program or to make no changes for a period of two years. The promotion program consisted of health education classes combined with a payroll-based incentive system… For each work site, smoking prevalence at the start of the study is saved under the variable name “baseline,” and smoking prevalence at the end of the study is saved under the name “followup.” The variable group contains the value 1 for work sites that implemented the health program and 2 for the sites that did not.  (the program.dta dataset is attached)

a.      For the work sites that implemented the health promotion program, test the null hypothesis that the median difference in smoking prevalence over the two-year period is equal to 0

b.      Test the same null hypothesis for the sites that did not make any changes

c.      Evaluate the null hypothesis that the median difference in smoking prevalence over the two-year period for work sites that implemented the program is equal to the median difference for sites that did not.