A new report weighs in on the performance of last year’s polls. Few Americans have faith in the government. And trust in scientists’ accuracy is deeply polarized. This is HuffPollster for Friday, May 5, 2017.
WHAT HAPPENED WITH 2016 ELECTION POLLING? - An American Association for Public Opinion Research (AAPOR) report, released Thursday, digs into the performance of last year’s surveys. HuffPollster: “National polls, which largely predicted a modest win for Hillary Clinton in the popular vote, weren’t too far off. But state polling missed the mark significantly, and often uniformly, leaving much of the public feeling utterly blindsided by Donald Trump’s victory...A late shift in key swing states and a failure to correct for the underrepresentation of less-educated voters played out against the backdrop of a close race that saw different winners in the Electoral College and the popular vote. The resulting errors were fundamentally worsened by pollsters, forecasters, and aggregators who were either overconfident in their results, or unable to convey a proper level of uncertainty to the public.” [HuffPost, full report, additional coverage from WashPost, Politico, AP]
What went wrong, according to the report: More from HuffPollster: “In Florida, Michigan, Pennsylvania, and Wisconsin, between 11 percent and 15 percent of voters said they made up their minds in the final week of the campaign. Late deciders in those states, several of which saw a last-minute surge in attention from the campaigns, broke heavily for Trump….[B]etter-educated voters were likely to support Clinton. That presented a problem for some pollsters, because highly educated voters are also more likely to answer polls….While many pollsters weight their surveys to correctly reflect Americans’ educational backgrounds, some, especially at the state level, did not.”
What we still don’t know: “The impact of other potential factors is less clear, according to the report. The report sheds no light, for example, on the effect of FBI Director James Comey’s letter to Congress….Similarly, while likely voter models ― pollsters’ efforts to determine which of the people they talk to will actually show up on Election Day ― presumably played a role, much of the data that would help measure the exact effect isn’t yet available.”
What didn’t go wrong: “The study found no evidence of a consistent bias toward one party in recent polling. While Trump was underestimated last year, Democrats Barack Obama and Al Gore also saw their standing underestimated in election polling. ‘The trend lines for both national polls and state-level polls show that ― for any given election ― whether the polls tend to miss in the Republican direction or the Democratic direction is tantamount to a coin flip,’ the report’s authors write.”
What to know for next time: “Pollsters who didn’t already account for educational levels can take more care to do so. But there’s no reason why future elections won’t also see last-minute shifts in swing states, or an Electoral College result that doesn’t reflect that popular vote…. panelists on Thursday suggested that pollsters, aggregators and journalists should spend more time emphasizing that surveys represent only a snapshot of public opinion at the time that they’re taken, and that the uncertainty surrounding polling goes far beyond the stated margin of error.”
HOUSEHOLDS WITH LANDLINES ARE NOW THE MINORITY - Pollsters who only call landline phones have an ever-shrinking universe to work with. Stephen J. Blumberg and Julian V. Luke: “The second 6 months of 2016 was the first time that a majority of American homes had only wireless telephones. Preliminary results from the July– December 2016 National Health Interview Survey (NHIS) indicate that 50.8% of American homes did not have a landline telephone but did have at least one wireless telephone (also known as cellular telephones, cell phones, or mobile phones) —an increase of 2.5 percentage points since the second 6 months of 2015... Despite operational challenges, most major survey research organizations include wireless telephone numbers when conducting RDD surveys. If they did not, the exclusion of households with only wireless telephones (along with the small proportion of households that have no telephone service) could bias results. This bias—known as coverage bias—could exist if there are differences between persons with and without landline telephones for the substantive variables of interest.” [CDC]
TRUST IN THE GOVERNMENT IS AT A NEAR-HISTORIC LOW - Pew Research: “The changes in the dynamics of power in Washington have registered with members of both political parties. Somewhat more Republicans express trust in government today than did so prior to the election, while views among Democrats have moved in the opposite direction. For the first time since George W. Bush’s presidency, Republicans (28%) are more likely than Democrats (15%) to say they can trust the government in Washington to do the right thing just about always or most of the time. The share of Democrats expressing trust in government is among the lowest levels for members of the party dating back nearly six decades. The national survey by Pew Research Center, conducted April 5-11 among 1,501 adults, finds that the overall level of trust in government remains near historic lows; just 20% say they trust the government to do what’s right always or most of the time. Far more say they trust the government only some of the time (68%); 11% volunteer that they never trust the government to do what’s right.” [Pew]
TRUMP’S APPROVAL RATING DIPPED SLIGHTLY BUT RECOVERED THIS SPRING - Mark Blumenthal, on SurveyMonkey’s data: “After a brief dip in March, President Donald Trump’s job approval rating rebounded slightly in April….That slight dip in March represented the lowest ebb so far in Trump’s presidency. Prior to that, our tracking of Trump’s ratings varied between 44 and 48 percent – essentially where they are now….[W]hile it is difficult to attribute recent changes to any one issue, it is likely that the shift in news coverage from a high-profile defeat of an unpopular proposal to military action and other news helps explain the glacial regression to the mean in Trump’s ratings. Independents who lean Republican have also been more variable in their assessment of Trump (partly because that subgroup is smaller, typically about 10 percent of all adults), but do show a slightly different pattern: Well over 80 percent approve of Trump, and while the numbers dipped slightly in late March, hitting a low of 84 percent in the week ending on April 3, they only rebounded in the past week (to 89 percent).” [HuffPost]
TRUST IN SCIENTISTS SPLITS ALONG PARTISAN LINES - HuffPollster: “Americans’ trust in scientists is deeply split along partisan lines, a new HuffPost/YouGov survey finds. 54 percent majority of Democrats, compared with just 13 percent of Republicans, say they have ‘a lot’ of trust that what scientists say is accurate and reliable…. In a HuffPost/YouGov poll taken four years ago, Democrats were 28 percentage points likelier than Republicans to express a lot of trust in scientists’ findings; in 2015, the gap was a relatively similar 25 points. But by last spring, that divide was 31 points, and the most recent poll finds it widened to 41 points, due both to increasing levels of trust among Democrats and a corresponding decline among Republicans.” [HuffPost]
HUFFPOLLSTER VIA EMAIL! - You can receive this update every Tuesday and 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).
FRIDAY’S ‘OUTLIERS’ - Links to the best of news at the intersection of polling, politics and political data:
-Nate Cohn sees risks for the GOP in the current health care debate. [NYT]
-Harry Enten hypothesizes that the health care bill could hurt President Trump among working-class voters. [538]
-Philip Bump charts out the expected impact of the Republican health care bill. [WashPost]
-Polls find little support among either the public or the medical community for weakening protections of pre-existing conditions. [Politico, WashPost]
-Quoctrung Bui, Claire Cain Miller and Kevin Quealy explore whether Trump’s presidency is becoming more normal. [NYT]
-Amy Walter asks whether GOP voters will stick with the party in the 2018 midterms. [Cook Political]
-Scott Bland notes Jon Ossoff’s surprisingly high level of Republican support in GA-06. [Politico]
-Eshe Nelson contends that the French presidential election won’t have a shock outcome. [Quartz]
-Most Americans say that LGBT athletes face discrimination from their teammates. [Marist]
-AEI (R) rounds up polling on the relationship between Trump and the press. [AEI]