Bernie Sanders, and How Indian Food Can Predict Vote Choice - The New York Times

There’s something about going to an Indian restaurant that correlates with candidate vote choice. — Credit...Caitlin Ochs for The New York Times

When was the last time you had vindaloo or tandoori chicken? Chances are if you’re a Democrat in Iowa supporting Joe Biden, it has been a while.

The latest New York Times/Siena College poll asked 584 possible Iowa Democratic caucusgoers lots of typical political questions, like whether they were Democrats or Republicans, and whether they planned to vote. But it also asked a few less obviously political ones, like if they’d been out for Indian food or how important it is to buy organic food.

Sometimes seemingly nonpolitical topics can shed light on people’s political choices, even after accounting for things like partisanship, education, geography and ideology. It’s as if the answers to these questions help account for some of what traditional political measures leave unexplained.

I’ve written before about how this might work, drawing on work done in sociology during the 1950s.

Questions about food, travel and the kinds of sports people engage in can be used as an index of someone’s local versus cosmopolitan orientation. In polling during the 2008 Democratic primary, such questions helped differentiate voters who chose Barack Obama in the 2008 Democratic primary from those who chose Hillary Clinton. The more likely that people were to experience other cultures probably unfamiliar to them — through travel or food — the more likely they were to vote for Mr. Obama, even controlling for things like income, education, personality, racial attitudes and city living.

This orientation toward the world also helped differentiate people who supported Donald J. Trump from those who supported any of the 16 other candidates in the Republican primary in 2016. Voters who had been to Europe, Australia, Canada or Mexico or had eaten at an Indian restaurant were less likely to choose Mr. Trump by 10 to 12 percentage points beyond the differences explained by other factors like the ones mentioned above.

The 1950s sociology researchers were looking not at voting behavior but at workplace dynamics. They used the distinction in trying to sort out whether employees were likely to solve problems locally (by working it out within their unit) or globally (by leveraging ideas outside their group).

In a recent paper, David Broockman, Gregory Ferenstein and Neil Malhotra returned to a workplace setting and showed that these orientations also separated technology entrepreneurs from other economic elites in terms of their attitudes toward economic and social policy. (Tech entrepreneurs are more cosmopolitan than other economic elites, and the authors think the rise of tech could therefore help to reduce economic inequality and other social and political inequalities, as these cosmopolitans start influencing politics and policy.)

In Iowa this year, a similar theme is emerging among possible Democratic caucusgoers. The Times/Siena poll revealed the same descriptive differences across the candidates’ supporters on basic demographics like age, education and race, and on political characteristics like whether they describe themselves as ideologically moderate or very liberal. For example, Bernie Sanders’s supporters tend to be younger and more liberal; Mr. Biden’s are older and more likely to be nonwhite.

It also showed policies on which everyone’s supporters mostly agree. Supporters of all the candidates give the president a low approval rating, disapprove of the military strike that killed the Iranian general Qasem Soleimani, and believe the president should be removed from office. One policy area where there was less agreement was on whether the United States should withdraw its military forces from the Middle East. Iowans who supported Mr. Sanders were the most in favor of the withdrawal, while those of Amy Klobuchar were the most critical.

There were no discernible differences on most of the nonpolitical questions across the candidates’ supporters in Iowa, such as on buying organic foods (most supporters of all the candidates think it’s important), using Twitter to read political news (most don’t) or watching television shows on premium outlets (also uncommon). Accounting for things like age and education soaked up most of the differences that appeared at first glance.

But, as has also been true in past contests, Indian food was a distinguishing characteristic. In Iowa, supporters of Mr. Sanders are its biggest fans: 71 percent of them report going to an Indian restaurant sometime in the last 10 years. Mr. Biden’s supporters are less likely to have done so by about 30 points. This makes sense. Mr. Sanders’s supporters are younger and perhaps more likely to live in the college towns or in major metropolitan areas. Still, this relationship persists even after accounting for age, race, gender, education, ideology, being an independent, or where a person lives in the state.

Mr. Biden loses 14 points of vote share among those who have been out for Indian food relative to those who have not, and Elizabeth Warren loses three. Mr. Sanders gains eight points, Pete Buttigieg gains five, and Ms. Klobuchar gains four.

Of course, it’s not that eating Indian food leads a person to support one Democratic candidate over another — that’s silly. (And there are voters for whom Indian food is the taste of home.) But a voter’s orientation toward the world is related to candidate choice, and it turns out that eating in restaurants that celebrate less familiar cultures is one way to measure where people think they are more connected: to those around them locally or to people farther afield.

Which Democrats will prevail this primary season — the cosmopolitans or the local-focused? Something to consider the next time you eat out.

Lynn Vavreck, the Marvin Hoffenberg Professor of American Politics and Public Policy at U.C.L.A., is a co-author of “Identity Crisis: The 2016 Presidential Campaign and the Battle for the Meaning of America.” Follow her on Twitter at @vavreck.

2020-01-30 10:00:00Z
https://news.google.com/__i/rss/rd/articles/CBMiSWh0dHBzOi8vd3d3Lm55dGltZXMuY29tLzIwMjAvMDEvMzAvdXBzaG90L2Jlcm5pZS1zYW5kZXJzLWluZGlhbi1mb29kLmh0bWzSAU1odHRwczovL3d3dy5ueXRpbWVzLmNvbS8yMDIwLzAxLzMwL3Vwc2hvdC9iZXJuaWUtc2FuZGVycy1pbmRpYW4tZm9vZC5hbXAuaHRtbA?oc=5