Share this:

Loren Collingwood 80x108In 2012, Latino turnout increased by more than a quarter, despite a fall in overall turnout that election year. The vast majority of this increased vote went to Obama. Why was Obama so successful with Latino voters? In new research, Loren Collingwood finds that a model that goes beyond voter demographics, and takes into account the Obama campaign’s outreach to Latinos and policy stances such as deferred action for undocumented immigrants helps to explain this success. He writes that when taking into account Obama’s ability to tap into the shared racial and ethnic identities of Latinos, voters were up to 70 percent more likely to vote for him because of this engagement.

One of the main storylines that came out of the 2012 presidential election was the role the Latino vote played in Obama’s victory. Among Latino voters, Barack Obama outpaced Mitt Romney by a margin of seventy-five to twenty-three—the highest rate of support for any Democratic candidate among Latinos. While turnout declined nationally from 2008 to 2012 by 2 percent, among Latinos there was a 28 percent increase in votes cast in 2012 (from 9.7 to 12.5 million) and Obama further increased his vote share among Latinos in 2012 compared to 2008. However, this was not a foregone conclusion, and many theories circulated since 2009 suggested the Latino vote might be underwhelming in 2012.  Indeed, as late as September 2012, a common headline in the popular press was “Latinos’ enthusiasm gap worries Dems” and that the Latino “seemed to be fading” .

Post-election media accounts of the 2012 Latino vote have suggested that Obama performed so well among Latino voters precisely because of their unique demographic characteristics: Latino voters are younger than average voters (younger voters tend to vote Democratic), have lower income (historically, poorer voters side with Democrats), and, perhaps as a result, tend to identify as Democrats. Still, others have suggested that Obama did so well among Latinos because he supported the Dream Act and initiated an executive order—“deferred action”—for undocumented Latino youth.

We put these accounts to the test by evaluating whether traditional vote-choice models adequately explain Latino voting behavior, or whether an enhanced model—which takes into account the candidates’ specific Latino outreach and policy stances, as well as voters’ perceptions of the candidates—better explains the 2012 Latino presidential vote. We find that when taking into account the Obama campaign’s ability to tap into the shared racial and ethnic identities of Latinos, voters were up to 70 percent more likely to vote for Obama because of this enagement.

We analyze Latino vote choice in the context of a theoretical framework—which we call cross-racial mobilization—that accounts for candidates’ ability to tap into shared racial and ethnic identities, as well as voters’ perceptions of outreach aimed at them. We do not put forward this framework as an alternative to standard vote-choice models, per se, but rather it seeks to improve standard vote-choice models by including candidate behavior and voter response in the multiracial context. Cross-racial mobilization is designed to explain both variation in candidate positioning on minority-specific issues and variation in candidate outreach aimed at members of different racial groups. Incorporating this type of candidate behavior—we argue—can improve our analysis of minority vote-choice models by tapping into how candidates exploit shared ethnic and racial identities.

We use an election eve survey conducted by Latino Decisions to evaluate Latino voting behavior, where our dependent variable is presidential vote choice. Our statistical analysis takes the traditional model as a starting point, and then we compare it against models accounting for the perceived outreach of cross-racial mobilization and the policy of cross-racial mobilization. Looking at just Obama’s cross-racial mobilization, our first measure of cross-racial mobilization asks the following question:

“Thinking about the 2012 campaign for President, would you say that Barack Obama is someone who truly cares about the Hispanic/Latino community, that he didn’t care too much about Hispanic/Latinos, or that Obama was hostile towards Hispanic/Latinos?”

Second, to evaluate the effect of policy cross-racial mobilization, we embed Obama’s most publicly discussed position on immigration policy. Obama’s immigration position is referred to as deferred action. We ask respondents the following question:

“In June President Obama announced a new Department of Homeland Security policy to stop the deportation of any undocumented immigrant youth who attends college or serves in the military and to provide them with a legal work permit that is renewable. Did this announcement make you feel more enthusiastic about Obama, less enthusiastic about Obama, or did it have no effect on how you feel about Obama?”

Looking at Figure 1 (the data are split sampled), the variable “Obama Cares” – our measure of perceived cross-racial mobilization outreach – and “Obama Direct CRM” – our measure of policy cross-racial mobilization — are both statistically significant. Compared with someone who said Obama’s deferred action had no effect on them, someone who said they are more enthusiastic about Obama is now effectively 20 percent more likely to vote for Obama. Likewise, a voter who thinks Obama cares about the Hispanic community compared to one who does not think so is over 70 percent more likely to vote for Obama. Both variables are either competitive with party identification or much more impactful on overall vote choice. This suggests that the traditional vote-choice model is under-specified with regards to the Latino vote. Indeed, including cross-racial mobilization variables improves the overall accuracy as well.

Figure 1 – Predicted change in probability of voting for Obama and cross-racial mobilization (click to enlarge)

Collingwood Fig 1

While traditional vote-choice models continue to play an important role in Latino voting behavior, our ability to explain this behavior is greatly enhanced by incorporating cross-variable mobilization variables that measure a candidates’ ability to tap into shared racial/ethnic identities. As America continues to diversify, we suspect that these cross-racial mobilization/shared identity variables will become a mainstay in the voting literature.

This article is based on the paper, ‘Revisiting Latino Voting Cross-Racial Mobilization in the 2012 Election’, in Political Research Quartery. 

Featured image credit: myJon (CC-BY-NC-SA-2.0)

Please read our comments policy before commenting.

Note:  This article gives the views of the author, and not the position of USApp– American Politics and Policy, nor of the London School of Economics.

Shortened URL for this post:


About the author

Loren Collingwood 80x108Loren Collingwood – University of California, Riverside
Loren Collingwood is an Assistant Professor at the Department of Political Science at the University of California, Riverside. His research and teaching focuses specifically on campaigns and elections, political behavior, race and ethnicity, and quantitative methodology.