American Indian party identification: why American Indians tend to be democrats

ABSTRACT

Surprisingly little research has examined the political attitudes and behaviors of American Indians. This paper helps fill the void by examining their party identification. Specifically, we expect that American Indians who are close to their American Indian heritage will be strong Democrats because the Democratic Party has an image and policy positions that are thought to be better for racial and ethnic minorities. To test this, we survey 300 American Indians using an opt-in Internet panel. We measure closeness to American Indian heritage with identity (race), attachment (linked fate), primary identity with tribe/U.S./American Indians, group consciousness (perceptions of discrimination and need for collective action), growing up on the reservation, language and lineage (number of American Indian grandparents). We find evidence that those who identify with another minority race, primarily identify with their tribe, believe American Indians are held back, and grew up on a reservation were more likely to be Democrats. We also found that those who were more proficient in their tribal language and had more American Indian grandparents were more likely to be Republicans. We argue that this offers general support for our expectation.

KEYWORDS:

• Racial politics
• social identity
• American Indians
• indigenous people
• party
• identification
• group identification
• group consciousness

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Although race is often defined by physical traits shared by a group of people and ethnicity by the culture of a people, the terms can be murky. This may be particularly true when considering American Indians since there is much variation in both culture and appearance from tribe to tribe. The assimilationist policies of the U.S. government also contributed to the phenomena that some people who may appear white can have strong cultural bonds and people who do not appear white have few cultural bonds. On top of this, there are many African Americans who are members of tribes due to the freedman policy following the civil war. As a consequence, we use the term race broadly to encompass both race and ethnicity.

2 Their measure of cultural ties was limited. They only examined responses to a direct question on whether the tribe or its leaders directly affected their vote. This question likely suffers from social desirability (people do not want to say someone influenced their vote) and does not truly measure individuals’ identity.

3 The Pew Research Center (2015) estimates that 50% of the mixed race Americans are a mix of White and American Indian.

4 According to Sturm (2010) most race shifters are true believers, whether or not they have any Indian blood, who find comfort in the communal and spiritual practices, while most Cherokee, the tribe they are most likely to claim ancestry in, believe many are seeking Indian services such as health care and find the appropriation of their culture offensive.

5 Originally from Deyhle and Swisher (1997, 166).

6 It is difficult to generate a sample of American Indians since there is no list of American Indians. In addition there are about 560 tribes across the country, many people who identify as American Indian are not members of tribes and American Indians make up less than 2% of the population.

7 The exact questions have been modified from questions used to measure these concepts by other researchers (Chong and Roger 2005; Masuoka 2006; Sanchez 2006; Wright Austin, Middleton, and Yon 2012). Some of these questions came from the 1994 NBES, a 1999 survey of Latinos conducted by the Washington Post, Harvard University and the Kaiser Family Foundation, and the 2000 Pilot National Asian American Political Survey.

8 Caution is needed in interpreting pseudo-R²s, as they cannot be interpreted along the lines of R². These, while ranging from 0 to 1, are much lower and an acceptable range can be as low as .10–.40. Our model that reports a lower R² has a large number of possible scores on the dependent variable and a small number of cases.

9 We would like to thank one of our anonymous reviewers for offering this suggestion for the findings.