Explaining state outcomes in presidential elections: Does nominee campaigning affect the vote?

This research explores whether campaigning by presidential nominees affects state outcomes in presidential elections. It integrates variables of presidential campaigning into general models of explaining state outcomes. Besides Democratic and Republican campaigning, the research focuses on the effects of a number of state-level variables: past voting history, state ideology, state partisanship, home state and regional advantage, and growth in personal income. The national variables are incumbency, growth in GNP, and tenure of the party controlling the White House. New data are gathered for creating new measures of these state-level variables. In the case of partisanship and ideology, dynamic variables are created in which values for each state change for each election. New theoretical frameworks are created for home state and regional advantage. The presidential campaign variables are based on campaign appearances. There is an extensive analysis of candidate appearances and their relationship to election outcomes and electoral votes. Ranked variables are used to separate out election-specific effects that skew variance among the states. Correlation analysis and OLS regression are used to test the variables. Among the many findings are: (1) Democratic campaigning has an effect on state outcomes, while Republican campaigning does not; (2) state ideology is the single most important statelevel variable in explaining state outcomes; (3) growth in GNP and party tenure appear to be the driving national forces in presidential elections; (4) presidential incumbency does not appear to have an independent influence in elections. Trying to build highly explanatory models of state outcomes is difficult without including election-specific ad hoc variables. However, the relatively parsimonious models here are still able to explain 60 to 65% of state outcomes in presidential elections. There is strong support for theories of voting that stress economics, ideology, and issues. The campaign theory of voting is partly confirmed. There is only a little evidence for the party-voting theory of elections. More research is needed to confirm the findings that Democratic campaigning affects votes and Republican campaigning does not. The reasons for these effects should also be explored.