My third dependent variable is State party strength. This measure combines the vote for governor, and those for the state house and state senate that are controlled by the two parties. This dependent variable along with the previous two were explained in “A New Measure of Party Strength.” These sorts of elections are often low information elections. Voters in these elections tend to be motivated to participate by friends, family, and partisanship. Relationships with friends and family does not tend to change much so most change in these elections tends to be driven by changes in partisanship and the demographics that support partisanship. With relationships playing a larger role in elections to state house and state senate offices, election…show more content… One of the novel aspects of this study is that I explore the impact of a variable that has previously been ignored by many studies: voter turnout. There are two basic concepts of partisan realignment. The first is that realignment is a sudden enduring shift of voters based on some event. The principle example of this from history was the realignment of the northeast in 1896 (Burnham 1970), though there is recent speculation that this occurred over a greater period time than previously thought (Stonecash and Silina 2005). The other is that the electorate is relatively stable, and that realignment is a process in which new voters shift the electorate because they vote heavily in one direction. The second notion has significantly stronger support, particularly from research on the realignment of the South. We know that there was a dramatic increase in voter turnout among white () and non-white voters in the South during the period of study (). Thus, in my models I examine the effects of voter turnout on party…show more content… For decades, black voter turnout was insignificant in the south. If African-Americans turned out to vote before World War II, they voted for the party of Lincoln. Overall, a very small percentage of black people were registered at the start of the period of this study. The percentage of registered voters was selected for inclusion in models for analysis since measures of the turnout of black voters do not start early enough for this study. One’s likelihood of voting is strongly correlated with whether or not one is registered to vote. Moreover, the dramatic increase in voter registration among black people appears to follow the same pattern of voting among the black population. The data shows dramatic increases in black voter registration and turnout between the beginning of the period of study and 2000, with 94 percent of black people in Arkansas being registered to vote in 1976. Recent work has shown that the movement of black voters to the Democratic Party led to white conservatives moving to the Republican Party (Hood, Morris, Kidd 2006). This includes conservatives that may not previously have voted. This variable should have a positive impact on party