Polling for Soup: How Real is the Sanders Surge? [St. Jane Archives]


Last week we saw a barrage of polls and the message from pollsters and pundits had been clear: Bernie had gained ground on Secretary Clinton and could actually have been ahead in both Iowa and New Hampshire. For many Hillary supporters the immediate rebuttal might be the slimness of Bernie’s leads there and her firewall of South Carolina.

However, there is a limited amount of data to point to suggesting that Secretary Clinton will obviously win the nomination. In Bayesian terminology Clinton supporters are relying on their priors and putting less emphasis on the data. They do have one piece of evidence to submit, below.

Secretary Clinton has been the frontrunner for the Democratic nomination since January of 2013 when PPP commissioned its first 2016 poll. The figure above just goes back to July, the time when Senator Sanders solidified his position as the major contender to Secretary Clinton. Although Clinton’s support varies, Sanders has had a slow linear increase in his support for months. However only one poll out of the 140 national polls in this time series ever gives Senator Sanders the outright lead. What’s most likely scaring the Clinton camp, and caused the pointed attacks on Bernie in the last week, is that losses in Iowa and New Hampshire change the entire dynamic of the nomination, and the trajectory of the national polls shown above.

          If you didn’t know anything about the candidates, or their relationship to the national party, you would assume that victories in Iowa and New Hampshire would propel that candidate to the nomination a la Romney in ’08 or Gore in ’00. Obama’s victory in Iowa alone immediately placed him as the alternative to Senator Clinton, and allowed him to withstand a surprising defeat in New Hampshire in 2008. By no means does winning Iowa guarantee you the nomination – just look at the Republican Party for the past few election cycles – but it certainly helps and a win there puts the Clinton camp on the defensive. The two figures below show Hillary with a slight lead in Iowa and Bernie with a slightly stronger lead in New Hampshire.

Digging further than the top line numbers shows the actual depth of Bernie’s support. In a Marist Poll in both Iowa and New Hampshire on the issues Democratic voters care about the most,  Bernie’s owned issues in Iowa and New Hampshire are numbers 1, 2, and 3 =  “Job creation and economic growth”, “healthcare”, and “climate change” respectively.

When I say “owned” issues I refer to Petrocik’s work on issue ownership, the idea that certain candidates or parties are better at dealing with specific issues over others. Democrats are considered to be better at dealing with unemployment, and Republicans are considered to be better at dealing with national security. Thus, they “own” the issue because the public places more trust in their ability to deal with it.

There are of, course, Democrats who say Healthcare is the issue they care about most who are also Hillary supporters and by no means is it absolute. But, in general, those voters are going to be closer to Bernie Sanders, just as we might expect Democrats who say National Security is their number one issue are more likely to support Secretary Clinton.

As you might have guessed, being the candidate who is considered better able to deal with the issues caucus and primary voters care about the most might be a good predictor of support. Hillary owns issues 4 and 5, “National Security” and “The government deficit and spending.” But these are not exactly the issues which drive Democratic voters to the polls. And in the same poll Senator Sanders does better in the head to heads against Trump, Cruz and Rubio. Bernie actually beats all three of them head to head, while Hillary only beats Trump.

So is the surge working for Sanders?

Sorry to revisit 2007, but I think it’s an apt way of describing the rift between the Democratic candidates and their supporters. Does the party want a liberal alternative to the Bush administration a la John Kerry, Barack Obama, and, yes, Hillary Clinton? Or does it want to remake itself in a more liberal image? The polling data, which is of course subject to change, is pointing positively in Bernie’s direction.

The fact that Healthcare is a top issue for Democratic voters after the biggest expansion in coverage since the Great Society, suggests that these primary voters want more than just programmatic change – they want a remaking of the Democratic Party and the country as a whole. And that’s not an exaggeration. I don’t think this nomination process is about minor alterations to the welfare state, or even differences in approaches to interventions in foreign policy. It’s about a bigger transformation of the party, those who lead it, and those who make up its ranks.

You’ve likely read hundreds of stories about the divides within the GOP and the coming revolt against the establishment, but Hillary and Bernie are arousing similar chasms. Not as malicious toward one another, and not targeted against people of different faiths and race as in the GOP, but a similar concern that the country is on the wrong path, and the continuation of incremental change is no longer acceptable. If you believe that the country needs to be remade, the place to start is at the presidency, and both parties are making their case.

The question we’re going to be faced with for anyone following the primary is not about the different issue positions the candidates hold  (which are well-documented at this point), but instead how the party presents itself and the way it builds its coalition. A Bernie victory means a new Democratic Coalition. The party has to decide if it’s ready for that.

Logan Vidal’s writing, including this piece, can be found on his blog, Polling for Soup.