I have to come up with a plan for my statistics project. In this project, I must use a confidence interval and at least two significance tests. My question is "There is a perception among Americans that lawmaking activity increases in election years. Is there evidence supporting this perception?"
Since election years fall every second year, I am comparing the difference between the number of bills passed in the first and second years of each Congressional Session. I will be comparing the means of two groups using SRS’s. The total number of observations (62) will be separated into two groups of 31, representing year 1 and year 2 (an election year), and will cover the years 1947 to 2008. I will be taking an SRS of 15 from each group (total 30 SRS).
I will produce a 95% Confidence Interval and then conduct a Test of Significance for the difference between the two groups’ means using a two-tailed significance test for alpha = 0.5. Ho: u2 – u1 = 0, Ha: u2 – u1 ≠ 0.
I'm trying to come up with a second significance test that I might apply to this problem. I thought of correlation/regression but I can't think of how it might apply. I'm wondering now if I should somehow expand my question to include a comparison of Presidential election years and test that question as well.
Any advice would be appreciated.