QRIOUSLY’S POLL FOR KEEPIN’ IT 1600 ON THE 2ND PRESIDENTIAL DEBATE
Qriously’s latest political poll comes in the form of a debate performance analysis: we ran a pre-debate poll on the weekend of 8-9 October (after the release of the tape involving Trump’s comments on groping women), and a second poll just following the debate on Monday / Tuesday 10-11 October (see methodology for more details).
The idea was to measure the impact of this debate on likely voters in eight key battleground states (Florida, Pennsylvania, Ohio, North Carolina, Colorado, Iowa, Nevada and New Hampshire). We also looked at how well each candidate performed during the debate according to viewers.
Overall, we found that Clinton comprehensively won the debate among our bipartisan sample of likely voters in battleground states (including among Independents); but her overall support didn’t increase significantly after the debate, and Trump’s support didn’t collapse. We don’t expect the next series of polls to show a significant change after the debate (unless they are in fact measuring the impact of the tape release).
CLINTON WON THE DEBATE
Hillary Clinton can rest easy, for now: as with the first debate, she is widely perceived to have won against her opponent, with 44% of likely voters in swing states (who had seen at least part of the debate) saying that she did the best job, compared to only 33% for Trump.
Both Democrats and Republicans felt that their candidate did the best job in the debate. However, crucially for Clinton, though, she made a bigger impact on Independents, who felt she won by 10 points (v.s. an average of 11 points).
Clinton performed well across all age groups, indicating that younger respondents are not as anti-Clinton as some in the political sphere have assumed.
…BUT THIS IS UNLIKELY TO TRANSLATE INTO INCREASED SUPPORT FOR CLINTON
Even though more likely voters said the debate increased their support for Clinton (v.s. support for Trump), the difference is not as strong as one might have expected given the proportion of viewers who said she did the best job in the debate.
Again, among independents in swing states, Clinton performs strongly compared to Trump (we see again the typical partisan split between Democrats and Republicans).
‘REAL’ GAINS IN A FOUR-WAY HORSE RACE ARE MINIMAL
Our pre-debate poll, conducted entirely after the news of Trump’s leaked comments about women, gave Clinton a 4-point lead over Trump; she is now five points up following the debate, but the change is not statistically significant (meaning her support could be stable). This means that Clinton managed to solidify her support among her pre-existing base, rather than converting undecided voters or Trump supporters to her cause.
This implies that the debate, on its own, hasn’t moved the needle significantly one way or the other. This is in line with conventional thinking that debates usually don’t shift polls in a noticeable fashion unless a candidate makes a significant gaffe, or performs well below expectations (such as Trump in the first debate). In this debate, although Clinton still put in the stronger performance, Trump improved from his disastrous first-debate showing, meaning that his numbers haven’t dropped much from this debate alone.
DESPITE TRUMP LOSING THE DEBATE, HE MIGHT NOT LOSE A LOT OF GROUND
Trump has a stable support base, and despite his loss in the debate (and the leaked tape of his comments about women), his core supporters are not deserting him.
There is no change in terms of certainty of support for either candidate in the pre- or post-debate polls.
THE DEBATE DID NOT CLARIFY POLICY FOR VOTERS
Putting current candidates aside, we found that Republican policies are generally more popular among likely voters in battleground states.
Interestingly, the proportion of voters who said they would prefer ‘neither’ Republican or Democratic policy increased significantly after the debate, implying that neither candidate managed to inspire voters in terms of their party’s platform and policies.
TRUMP IS TRAILING DESPITE REPUBLICAN POLICIES BEING MORE POPULAR
Given that Republican policies are actually more popular than Democratic policies in these states, why is the Republican candidate not leading in the polls?
It implies that either a) Clinton is particularly appealing to voters, even if they prefer the Republican policy platform; or b) Trump is particularly unappealing to voters, and so even if they prefer Republican policies, they are choosing to hold their nose and vote Clinton in this election. Given both candidates’ favorability ratings (obtained by other pollsters), we feel that the latter is most plausible.
(UNSURPRISINGLY) TRUMP’S SUPPORT AMONG WOMEN IS AT RECORD LOW
Not surprisingly, given his recent statements about groping, Trump’s support is abysmal among women (as other analysts, such as FiveThirtyEight, have pointed out recently). Women are much more likely to support Clinton overall, while men are strongly pro-Trump; Trump has a 10-point lead among likely male voters, while Clinton has a 17-point lead among likely female voters.
This is far higher than recent presidential match-ups; Obama led among women as well, but only by 8 points, which is typical for a Democrat candidate (a higher proportion of women identify as Democrats, meaning that Democrats should poll well among a bipartisan sample of women voters). Clinton’s lead is stronger than Obama’s, which emphasizes how poorly Trump is doing with women (and how well Clinton has done to assimilate many disaffected Republican women into her base).
Women are also more likely to be undecided at this point than men; it remains to be seen which candidate will capture the remaining 14% of female likely voters who still haven’t picked a side. Given that Trump’s support among women is collapsing following his recent comments, it seems likely that Clinton will capitalize, and take the presidential race in the process.
LINK TO PODCAST: http://bit.ly/2dKn9oy
Data Collection Method: Interviews were conducted via mobile. We intercept people across 50,000 mobile apps to invite them to participate in the survey on their mobile. Apps are selected dynamically to generate a balanced demographic sample.
Fieldwork Dates: Pre-debate: 9.30 am ET Saturday 8 Oct – 9.30 am ET Sunday 9 Oct. Post-debate: 9.30 am ET Monday 10 Oct – 9.30 am ET Tuesday 11 Oct
Sample: Pre-debate: 2,895 Americans (18+), including 1,439 likely voters. Post-debate: 2,989 Americans (18+), including 1,478 likely voters
Geography: Florida, Pennsylvania, Ohio, North Carolina, Colorado, Iowa, Nevada and New Hampshire. The states were sampled according to their population size.
Weighting: All data is weighted on gender, age, ethnicity and party affiliation to be representative of the population of the 8 swing states examined.
Dutch General Elections 2017: Our latest Poll With VVD to Win by a Significant Lead as PVV Loses MomentumRead More
The one critical skill many data scientists are missingRead More
After Brexit and Trump, Qriously Accurately Calls Italian ReferendumRead More
How We Performed on Our Swing State ForecastsRead More
And the Winner of the 2016 US Presidential Election is…Read More
Samsung Recall: Negative short term impact on image and sales but situation not desperateRead More
Pokémon GO Is Not Just A FadRead More
How is Pokemon GO actually performing?Read More
EU Referendum Post-Mortem: Why Were the Polls So Wrong Again (and why our mobile polls got it right)?Read More