Why do American Indians vote Democratic?

Jeonghun Min, Daniel Savage

The Social Science Journal · 2013 · 16 citations · 29 references

Concepts

Abstract

AbstractWe examine the relationship between a relatively large population of American Indians in eastern Oklahoma and the slow pace of that region's partisan realignment. With a 2012 exit poll, we empirically examine whether American Indians' cultural ties and their socio-economic characteristics affected their vote choices. Because of their greater poverty, American Indians are more economically progressive and vote more Democratic than their white counterparts.American Indians' socio-economic circumstances play a greater role than their cultural ties in their vote decisions and significantly influence why they are slower to abandon the Democratic Party in eastern Oklahoma.Keywords: American IndiansPartisan realignmentVote choicesSocio-economicCircumstancesPopulism Notes1 Present address: Department of Social Sciences, Northeastern State University, 609 North Grand Avenue, Tahlequah, OK 74464, USA. Tel.: +1 918 444 3506; fax: +1 918 458 2390.2 V.O. Key (1955) originally made the distinction between critical realignments, which occurred quickly in dramatic fashion, perhaps in a single election, and secular realignments, which are gradual and occur over a series of elections.3 While Oklahoma was not a state during the Civil War, it is what might be considered a border state, in that the eastern part of the state—particularly the southeast, which is sometimes referred to as "Little Dixie"—is more "southern" in a cultural sense than the western part. In this sense, Oklahoma is similar to Texas, which is considered a southern state because it was part of the Confederacy during the Civil War. Politically, the state has followed generally the same partisan alignments and realignments as the southern states. According to the Oklahoma State Election Board (http://www.ok.gov/elections/), 83% of voters in Oklahoma were registered as Democrats in 1960. That percentage had dropped to 63% by 1995 and only 47% of the state's voters were registered as Democrats in 2012. Oklahoma has experienced the secular realignment from the Democratic to the Republican Party during the same period that the traditional southern states have.4 Despite the fact that eastern Oklahomans voted Republican in the 2012 presidential and congressional elections, registered Democrats outnumber Republicans in the 2nd Congressional District by a 2–1 margin. As of October 30, 2012, District 2 had the following partisan registration breakdown according to the Oklahoma State Election Board: 260,436 Democrats (63%), 110,170 Republicans (27%) and 39,751 unaffiliated/other (10%). http://www.ok.gov/elections/Voter_Registration/Voter_Registration_Reports/index.html.5 According to the 2010 census, American Indians make up 18.8% of the population of the Second Congressional District, while African- and Hispanic-Americans make up only 7.9%. Another 7.3% of the district's population is mixed race, probably due to the fact that American Indians are more likely to marry into other races—mostly white—than any other racial group. Of those Americans who identified themselves as American Indian, 43.8% reported being of mixed race. Only 3.2% of whites and 7.4% of blacks identify themselves as mixed race. Of the mixed raced American Indians, 62.8% reported that they were mixed American Indian and white. When the percentage of American Indians and those of mixed race are combined with African- and Hispanic-Americans, the combined percentage of minorities in the district is 34%.6 Oklahoma's 2010 Census Population—Demographic Profile. Oklahoma Department of Commerce. http://okcommerce.gov/data-and-research/demographics-and-population-data/.7 These questions were suggested by Corntassel and Witmer (2008). Their book is on the relationship between tribal and state governments. Their survey was of tribal governments while ours is of individual American Indian voters.8 As for the racial composition of the respondents in our exit polls, 64% of respondents were white, 32% identified themselves as American Indian, and the remaining 4% consisted of all other racial or cultural backgrounds. The 32% of respondents who are American Indian is slightly lower than, as stated above, the percentage of American Indians in the two counties.9 It should be noted that American Indian cultural issues did not play a major role in any of the 2012 election campaigns.10 The results of the analysis are available from the authors.11 Our exit poll shows that eastern Oklahoma is an ideologically conservative region where 42% of whites and 39% of American Indians identified themselves as conservative.12 The results of the analysis are available from the authors.13 To measure the influence of the status of the economy on vote choice, we include a variable, Economy, in the presidential and congressional elections. For the presidential election, it is coded 1 if a respondent chose "The status of the economy" as the most important factor for his/her vote for president and 0 if otherwise. For the U.S. House election, it is coded 1 if a respondent answered that "My concern about big government" mattered most for his/her vote and 0 if otherwise. Meanwhile, we include a variable, State Connection, to measure the influence of a candidate's connection to the local community on vote choice in the state legislative elections. It is coded 1 if a respondent responded that "The candidate's connection to the local community" mattered most in deciding his/her vote and 0 if otherwise.14 The results of Likelihood Ratio test, shown as Model χ2 in Table 8, show that the independent variables we employ in the models are jointly significant in explaining the dependent variable, American Indians' Democratic Vote, at .05 level. We also measure whether there is a specification error in the models to see if the models are properly specified and we need more variables in the models (if the model is properly specified, one should not be able to find any additional predictors that are statistically significant except by chance). We found that our models are correctly specified. The results of the analysis are available from authors.

References

29