Search

In 2001 Republican Latino candidate for mayor Orlando Sanchez won 73% of the heavily Democratic Latino vote in Houston while in Miami, Cuban Democrat Manny Diaz won 70% of the staunch Republican Latino vote. In 2003, Green Party candidate for mayor Matt Gonzalez captured 66% of the Latino vote in San Francisco while in Colorado Springs a majority of Latinos voted for Republican Lionel Rivera. These elections renew the debate over Latino vote preference, and call into question whether Latinos follow partisan cues or ethnic cues when casting a ballot. I argue that while party is still an important predictor of vote choice, for Latinos with a high degree of shared ethnicity, party ties are less significant than ethnic attachment. Using a unique survey fielded by the Tomás Rivera Policy Institute in 2002, I model crossover vote preference for Latino candidates in California and New York using probit and ordered probit regression. While strong partisans were significantly less likely to crossover and vote for the ethnic candidate of the opposing party, Latino voters who scored high on ethnic attachment were significantly more likely to ditch their party for the Latino candidate. By incorporating more flexible measures of ethnicity (as opposed to a dichotomous measure), this paper finds that, in addition to SES and party, ethnicity can be an important predictor of candidate preference. [ABSTRACT FROM AUTHOR]

The effect of ethnicity upon political participation presents an interesting paradox to political scientists. No consensus exists on the extent to which the interaction between contextual factors and individual level characteristics allow individuals to overcome barriers to participation, or how this relates to patterns of mobilization. While it is well documented that contacts and mobilization are effective in turning out the vote, few scholars have asked the question of who gets mobilized, and why? In this paper, we use a modified model of the effects of group size and group empowerment for Latinos in the United States to predict which Latinos reported being mobilized prior to the 2000 Presidential election. The modifications include a departure from an emphasis on the presence of Black or Latino mayors as a measurement of empowerment. Instead, we propose using a multiple-level Latino representation model of empowerment. Using a five state survey of registered Latino voters collected after the 2000 elections, we test this alternative model of empowerment to determine if Latino registered voters, living in districts with Latino representatives, are more likely to be contacted and asked to vote. Further modifications include making a distinction between type of contact and the addition of other characteristics that have been found to significantly impact Latino participation. The strategic nature of party contacting leads organizations to contact those individuals who are already the individuals most predisposed to participate. Probit regression analysis finds that Latinos represented by co-ethnics are more likely to be mobilized than Latinos represented by White or Black elected officials, providing additional support for the notion that descriptive representation empowers minority communities. However, the relationship is not linear. Latinos with only one Latino representative are the most likely to receive mobilization while those with two or three representatives are less likely. [ABSTRACT FROM AUTHOR]

Since the 2000 Presidential election voter education and mobilization has witnessed a renaissance in targeted contact and segmented messaging. Candidates, political parties, and interest groups have taken advantage of advances in electronic databases to divide and subdivide the electorate into different groups, and have different messages and messengers for each subgroup of voters. For decades, scholars and practitioners have known that personalized campaign appeals are more effective. Labor unions, religious leaders, military veterans, and racial and ethnic community leaders have typically been used by campaigns to make a stronger, more personalized, and perhaps more trusting connection with voters. This paper takes up the question of whether or not personalized or segmented contact during a campaign is more successful at convincing voters than "generic" contact, or no contact at all.Using data from a national survey of Latino registered voters in 2004, we examine the impact of being contacted by a co-ethnic messenger on support for the Republican and Democratic parties. While some previous studies have examined voter turnout or vote choice, this paper examines the deeper implications of co-ethnic contact, including support for public policy and candidate favorability. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Abstract: Despite limited experience courting the Latino community, both political partiesmade a point to campaign aggressively for Latino "swing" votes through advertisements andget-out-the-vote (GOTV) drives in 2006. While there is a new commitment to Spanishlanguageadvertising and direct contact with Latino voters, no evidence exists that thistargeted outreach has worked. The question remains, does targeted advertising influence thevote choice of Latinos, and, how does this advertising affect the more general population?To address these questions, we implemented an experiment during the 2006 Californiaelection. Latino and non-Latino subjects were randomly assigned to a control or treatmentgroup. Those in the treated groups received a flyer endorsing Arnold Schwarzenegger forgovernor, containing an endorsement from a Latino or an Anglo political figure. Thetreatment ads were further divided into English only or in both English/Spanish. Thus, thegroups were as follows: Anglo endorser English, Latino endorser English, Anglo endorserbilingual, and Latino endorser bilingual. We expect to find that targeted messages shouldhave a stronger influence on vote preference than non-targeted messages among Latinos.Further, the combined effect of a Latino endorser with a message delivered in bothlanguages might have the strongest effects. Finally, non-Latinos should be less receptive tothe Latino-targeted ads, perhaps witnessing a demobilizing effect. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]