HUFFPOLLSTER: Is Bernie Sanders Really Leading In New Hampshire?

Pundits weigh in on a New Hampshire survey showing Sanders ahead. Primary polls are valuable but not very reliable. And a majority of Americans agree on stricter gun access laws. This is HuffPollster for Friday, August 14, 2015.

ABOUT THAT NEW HAMPSHIRE POLL - A new Boston Herald-Franklin Pierce University poll released earlier this week found Sen. Bernie Sanders (I-Vt.) with a not-quite statistically significant 44 percent to 37 percent lead in New Hampshire over former Secretary of State Hillary Clinton. Support for Sanders has been growing in New Hampshire polling, although every previous survey, including a previous Herald-Franklin Pierce poll in March, had shown Clinton ahead. The results got pundits buzzing and questioning just how well Sanders is doing.

-Matthew Dickinson: "[M]y last post cautioning readers to be wary of drawing conclusions based on one poll has been drawing its fair share of criticism in light of a more recent Boston Herald/Franklin Pierce poll that has Bernie leading Clinton 44%-37% in New Hampshire...Since I received a few emails after my last post asking me to clarify the difference between a “statistical tie” and what the New York Times mistakenly (in my view) called a “dead heat”, I thought it might be useful to present the latest poll results visually, using a nifty app developed by Nicholas Neuteufel that graphs the polls results, including the margin of error."

"Once again, as the graph suggests, we can’t discount the possibility, given the margin of error, that Clinton and Bernie are tied, or that Clinton might even be slightly ahead. At the same time, however, based on this one poll, the odds are greater that Bernie is now ahead of Clinton in New Hampshire. But, of course, as I reminded readers in my last poll, we shouldn’t rely on only one poll. Not surprisingly – and my caution notwithstanding – Bernie supporters seem convinced that this latest poll is an accurate barometer of the current state of the Democratic primary race New Hampshire. Note, however, that both the RealClearPolitics and Pollster.com aggregate polls continue to have Hillary holding a slim lead over Bernie in the Granite state." [PresidentialPower]

-Philip Bump: "[The Herald-Franklin Pierce poll is] good news for Sanders, after a poll from WMUR-TV last week appeared to show his gains flattening. But it's not necessarily great news….Sanders's growth is in part a function of his increased name recognition. What's more, Clinton's 'very favorable' support has dropped, and Sanders's has risen. All of which is good news for Sanders — assuming that this poll reflects both reality and a continued trend. It's too early to tell that, of course. And also assuming that the sample used in the poll reflects the demographics of the primary electorate. There are big differences in how younger and older voters view the candidates, for example." [WashPost]

-Sean Trende: "Should we buy [the Herald-Franklin Pierce poll?] The “yes/no” dichotomy rarely serves us well in election analysis. All data have some value to them, and we should generally be reluctant to dismiss data outright simply because they do not conform to our previously held assumptions. It’s simply, or perhaps not so simply, a matter of figuring what that value is, and then how much we should adjust our prior beliefs on the basis of those data. So as with most things, the reality here is complicated. On the one hand, this is the only poll that has ever shown Clinton down, and it shows her down, suddenly, by a fairly impressive margin. It also shows her support dropping below 40 percent for the first time ever, after having being fairly stable since March. And while Franklin Pierce is a well-respected pollster, it isn’t among the more active ones in the Granite State." [RCP]

More on methodology: The Herald/Franklin Pierce survey used a methodology that is now common in national media polls. It called both landline and cellular telephone numbers selected by a process known as random digit dial (RDD) sampling. The pollsters interviewed all the adults they contacted and weighted separate landline and mobile phone samples together using Census and other government statistics for gender, age and the telephone usage. The "likely primary voters" were those respondents culled from the larger sample who said they were both registered to vote and would "definitely" or "probably" vote in the presidential primary.  

As typically used in media polling, this procedure allows the composition of the likely primary voters to vary depending on respondents' self-reported likelihood of voting. The pollsters generally apply no weighting or controls specific to likely primary voters, an omission that usually raises hackles from pollsters hired by campaigns, who are much more willing to use statistics on past turnout to weight their likely voter samples.

IS CLINTON 'REALLY IN DANGER?' - Nate Cohn: "You might not know it if you’re reading the various articles about Mr. Sanders’s rise in the polls, but Mrs. Clinton still holds as strong a position as any primary candidate in history....The Sanders surge has slowed over the last month. Yes, a poll out of New Hampshire has him leading Mrs. Clinton. But Iowa and New Hampshire were always going to be his strong spots — just as liberal havens like Seattle and Boulder, Colo., are favorable terrain. A closer look at the polls shows that he is simply not within striking distance of winning the nomination. [Sanders'] support has run into a wall: women, blacks and Hispanics continue to support Mrs. Clinton by a wide margin, as do white moderate and conservative Democrats. Mr. Sanders has become the favorite of one of the Democratic Party’s most important factions: the overwhelmingly white, progressive left....This voting support is enough for him to compete in Iowa; New Hampshire and elsewhere in New England; the Northwest; and many Western caucuses. But it is not a viable electoral coalition in a Democratic Party that is far more moderate and diverse than his supporters seem to recognize." [NYT]

'ARE THE POLLS SKEWED?' - Steven Shepard: "Donald Trump is on top in Iowa, Ben Carson is hot on his heels, and Bernie Sanders has flown past Hillary Clinton in New Hampshire. That’s the snapshot polls captured this week. It’s a far cry from the Bush-Clinton horse race that many had predicted going into the 2016 campaign season — and it belies conventional wisdom that predicted Trump’s demise after he attacked Fox News host Megyn Kelly after last week’s GOP presidential debate. The results have skeptics asking: Are these numbers the real deal?" Some Republicans have insisted for months that Trump’s meteoric rise in the polls vastly overstates his standing in the race for the GOP nomination because the polls are surveying people who — despite what they tell pollsters — won’t actually cast ballots in their state’s primary or caucus. That argument reached a new crescendo on Wednesday, after a CNN/ORC poll showed the real estate magnate with a significant lead in Iowa, trailed by neurosurgeon Ben Carson. The two first-time candidates were the only ones to register double-digit support." [Politico]

THE PROBLEM WITH PRIMARY POLLS - Ryan Grim and Natalie Jackson: "[T]here are big problems with how much we rely on polls: They have almost no ability to predict the results of elections that are more than a year away. And it's getting worse: the increasing use of cell phones and the decline of people willing to answer surveys has put the election polling industry in a state of "near crisis," as one prominent expert recently put it. The problems are myriad." Among them, "nobody's paying attention." Here's one chart that illustrates how closely people are paying attention over time leading up to the election. [HuffPost]  

SURVEY RESEARCH ORGANIZATIONS JOIN LAWSUIT AGAINST AUTODIALER RULES - Howard Fienberg: "America’s two national research associations representing the profession and industry of survey, opinion and marketing research have filed a “motion to intervene” in a court case against new telephone rules from the Federal Communications Commission (FCC). In their motion, the Council of American Survey Research Organizations (CASRO) and the Marketing Research Association (MRA) contend that 'the definition of an autodialer' in the FCC’s new Telephone Consumer Protection Act (TCPA) rules that restrict the use of autodialers to call cell phones 'must be clarified to focus on the current capacity to generate and dial random or sequential numbers, and/or clarified to exclude calls that involve human intervention in the dialing.' 'If the court rules in our favor, we could walk away with a more constrained autodialer definition and an applicable human intervention test — both of which could be major points of relief for the research industry,' said Diane Bowers, CASRO president. 'We also seek relief from class action litigation over reassigned cell phone numbers. The FCC’s new rules create an unnecessary level of risk for researchers,' added David W. Almy, MRA CEO. [MRA]

AP PILOT AIMS TO REINVENT THE EXIT POLL - Scott Bland: "Exit pollsters do their work in a few short days, but the information they glean dominates the political conversation for years. That's because the surveys conducted outside voting precincts are the single-best tool for understanding American politics. They generate a treasure trove of data—about who votes, who they voted for, and why—that tell the story of how each election was won and lost. But now, as more voters cast ballots early or by mail and polling those voters becomes more costly and potentially less reliable, the exit polls are getting more expensive, too—and might even be losing some of their predictive power...In response, the Associated Press is trying to build the exit poll of the future....Last month, the media organization won a $250,000 Knight News Challenge grant to conduct experiments aimed at building a poll that's both reliable and cost-effective in 21st-century conditions. Over the next 15 months, the news organization and its polling partners will use the money to test online survey methods during real campaigns, from the 2015 Kentucky governor's race through next year's presidential elections. By the time it closes the books on the 2016 elections, AP hopes to have identified survey methods that it can recommend to the NEP, which includes the main broadcast and cable news networks, for use in the official exit poll in future elections." [National Journal]

A MAJORITY OF AMERICANS SUPPORT STRICTER GUN ACCESS LAWS - Pew Research: "Two years after the failure of Senate legislation to expand background checks on gun purchases, the public continues to overwhelmingly support making private gun sales and sales at gun shows subject to background checks. Currently, 85% of Americans – including large majorities of Democrats (88%) and Republicans (79%) – favor expanded background checks, little changed from May 2013 (81%)....Nearly eight-in-ten (79%) favor laws to prevent people with mental illness from purchasing guns, 70% back the creation of a federal database to track all gun sales, while a smaller majority (57%) supports a ban on assault-style weapons." [Pew Research] 

This Week's Polls

-Ben Carson trails GOP frontrunner Donald Trump in Iowa, while Hillary Clinton holds strong in the Democratic primary. [CNN, Bloomberg]  

-Carly Fiorina gets a bump, while Rand Paul takes a drop in the Iowa caucus. [PPP]

-Donald Trump remains the favorite in Iowa post-debate. [NYT, Suffolk University]

-John Kasich and Fiorina gain ground in New Hampshire while Trump falters. [Boston Herald]

-Clinton loses in every match-up against a Republican candidate in Missouri. [PPP]

-Alabama Democrats overwhelmingly choose Clinton over Sanders. [WKRG]

-Republicans say Fiorina won the GOP Fox debate. [NBC]

-Fioirina, Rubio and Carson get the biggest image boost post-debate. [HuffPost]

-Democrats are split on wanting Biden to run for president. [Gallup]

-Passing time hasn't eased -- or much increased -- Americans' concerns about Hillary Clinton's emails. [HuffPost]

HUFFPOLLSTER VIA EMAIL!- You can receive this weekly update every Friday morning via email! Just click here, enter your email address, and click "sign up." That's all there is to it (and you can unsubscribe anytime).

THIS WEEK'S 'OUTLIERS' - Links to the best of news at the intersection of polling, politics and political data:

-Clinton has a huge advantage in favorability among black voters. [Gallup]

-Harry Enten looks at candidates' poll numbers before and after the debate. [538]  

-Iowa voters were left impressed with Rubio and Carson. [WashPost]

-Philip Bump finds that Republicans prefer Trump over any other candidate on almost every major issue. [WashPost]

-Nate Silver says Trump is winning in the polls, but losing the nomination. [538]

-Trump says Frank Luntz is biased against him “because I didn’t want to hire him commercially.” [Business Insider]

-Christine Matthews (R) shares what she knows about why Trump is where he is. [Bellwether Research]

-The Upshot unveils a new chart ranking the Republican candidates on polling, endorsements, fundraising and the rankings of prediction markets. [NYT]

-Jon Ralston explains how internal campaign polls inform his reporting. [RalstonReports]

-Jesse Singal argues that aVanity Fair article on Tinder emphasized anecdotes and ignored contradictory data. [NYMag]

-Ben Casselman reports on the dire consequences of Canada's shift to a voluntary survey to measure household data. [538]

-A growing majority of Facebook and Twitter users say they use the platforms as their main source of news. [Pew Research]

-Despite improvements in the economy, millennials are still living with their parents. [Pew Research]

-Facebook's researchers find "haha" beating "lol." [Facebook via Flowing Data]