The file P10_56.xlsx contains some interesting data on the U.S. presidential elections from 1880 through 2008. The variable definitions are on the Source sheet. The question is whether the Vote variable can be predicted very well from the other variables.
a. Create pivot tables and/or scatter plots to check whether Vote appears to be related to the other variables. Comment on the results.
b. Run a regression of Vote versus the other variables (not including Year). Do the coefficients go in the direction (positive or negative) you would expect? If you were going to use the regression equation to predict Vote for the 2012 election and you had the relevant data for the explanatory variables for 2012, how accurate do you think your prediction would be?