Hypothesis testing for proportion in dependent samples. 

1. You are comparing whether a junior high anti-drug program has an effect on the rates of drug use in junior high students such that students completing anti-drug program will have lower rates of drug usage. You random sample 100 seventh grade students before they enter the program and after they complete the program.  Which test do we use to see if the difference in our sample is statistically significant?  Why?  Should it be one or two-tailed?

2.A researcher is studying the difference between gender and support of the death penalty.  The researcher feels that men are more likely to support the death penalty than women.  Both IV and DV are defined as nominal dichotomies and compared in a cross-tab.  Which measure of association would the researcher use to compare the relationship between the IV and DV?  Why?