The election of Donald Trump to the U.S. presidency surprised almost everyone, including apparently Trump himself.
On the morning after the 2016 election, my teenage son made snarky comments about the state of polling and statistical science. As a trained statistician, I took offense. However, I had no background in political science and really no idea what had gone “wrong.”
So I decided to put him to work, gathering and entering vote totals and poll data from 2016 and past elections, to judge for ourselves. In our analysis, we examined the performance of presidential poll-based predictions and proposed a new, improved model.
The 2016 election caused considerable hand-wringing over the state of opinion polling. However, the best evidence is that current polling is generally sound, but tune-ups to one particular aspect of how polls are collected – notably the practice of aggregating poll data – would be helpful.
Polling versus prediction
After the unexpected election outcome, most observers concluded that the polls, on average, underestimated support for Trump. Such a systematic error is known as “polling bias.” Numerous reports and think pieces have tried to explain the bias, pinpointing specific problems in how polls assessed likely voters and were weighted by voters’ education levels. Public misunderstanding of the concept of uncertainty also played a role.
To understand the issues, it’s important to recognize the distinction between polls, which represent samples of individuals at a particular time using a particular methodology, and poll aggregation.
Poll results can vary greatly, depending on who is sampled and what methods the pollsters use to weight respondents to account for nonrepresentative sampling, or assess who is likely to vote. With more and more people not answering pollsters and using cellphones, it’s an amazing testament to the field that opinion polling still works as well as it does.
When it comes to predicting an election, one must reconcile the often disparate poll results. That’s the role of poll aggregation sites, such as FiveThirtyEight, The Upshot and HuffPost, which average recent polls to produce a consensus.
For presidential elections, the sites go further and make predictions for each state, to tally a final electoral college outcome. Flashy graphics and accessible content make the sites hugely popular, driving public perception of the likely outcome.
Did the poll aggregators really miss?
Our own research suggests that the polling bias was actually not very large – that is, pollsters may have underestimated the support for Trump but not to a large degree. However, due to a statistical quirk, the prediction models were unable to recognize the dropping support for Hillary Clinton just prior to the election.
We examined the state-level predictions across all 50 states, plus D.C., in 2016, as well as their stated uncertainty.
We found that FiveThirtyEight and The Upshot showed statistical bias and overestimated support for Clinton, but with enough uncertainty that their probabilities left room for a Trump victory. HuffPost had similar state-level predictions, but was overconfident in these predictions, markedly overestimating the chance of a Clinton victory.
When the polling data up the eve of the election were fully taken into account, we estimated the chance of a Trump victory as at least 47 percent. The main novelty in our approach was to use polling data from multiple states to “fill in” the sparse information from state-level polling.
In contrast, the popular poll aggregation sites gave much lower chances, ranging from 2 percent on HuffPost to about 29 percent on FiveThirtyEight.
An alternate explanation
But why would the polls have biased against Trump in the first place? An extensive report from the American Association for Public Opinion Research (AAPOR), an association for public opinion and survey research professionals, examined a number of possibilities, dismissing some popular theories.
Our own analysis suggests an additional possibility, hinted at in the AAPOR report: The polls weren’t highly biased and were roughly correct at the time. However, pollsters conducted too few state-level polls just prior to the election.
Remember, poll aggregators must average several polls to make a good prediction. Due to sparse state-level polling, predictions were “stuck” on values from about two weeks prior to the election, when support for Clinton had been higher.
Note that this sparse polling scenario indicates that polling methods are generally sound, although more frequent polling of swing states would be helpful.
Our rationale also explains why the estimate of the popular vote – 3.3 percent estimated margin for Clinton versus 2.1 percent actual – was largely accurate. National polls were conducted more frequently, and so the national averaging could include polls closer to the election.
Can pollsters do better?
The voter environment in 2016 election was unusual, with a sharp drop in support for the leading candidate prior to the election. Although it’s tempting to attribute the drop to then-FBI Director James Comey’s letter regarding an investigation into Clinton’s email server, we and others have noted that the drop in support for Clinton started in mid-October.
Although the 2016 experience was unusual, we proposed a statistical model designed to be sensitive to a national trend. The model combines information across numerous states, instead of relying only on polling within each state. The model estimates of the Democratic-Republican vote spread and overall win probabilities for the 90 days leading to an election.
Although our analysis was conducted after the election, we plan to try it out in 2020 in real time.
Black Georgia lawmaker accuses white man of demanding she ‘go back where she came from’ in supermarket diatribe
On Friday evening, Erica Thomas, and African-American Democratic lawmaker in the Georgia House of Representatives, was shopping at a Publix supermarket in Mableton when a white customer came up to her and shouted at her, telling her to "go back where you came from" — words echoing President Donald Trump's recent racist attacks on four Democratic congresswomen of color.
Thomas' crime? She had too many items for the express checkout line.
Today I was verbally assaulted in the grocery store by a white man who told me I was a lazy SOB and to go back to where I came from bc I had to many items in the express lane. My husband wasn’t there to defend me because he is on Active Duty serving the country I came from USA!
Trump offers to guarantee bail for rapper A$AP Rocky
US President Donald Trump offered Saturday to guarantee the bail of rapper ASAP Rocky, detained in Sweden on suspicion of assault following a street brawl.
Trump tweeted that he had spoken with Prime Minister Stefan Lofven, who he said gave assurances that the singer would be treated fairly.
"Likewise, I assured him that A$AP was not a flight risk and offered to personally vouch for his bail, or an alternative," Trump wrote.
There is no system of bail in Sweden.
Trump said he and Lofven had agreed to speak again over the next 48 hours.
Fans, fellow artists and US Congress members have campaigned for the 30-year-old artist, whose real name is Rakim Mayers, to be freed since his arrest on July 3 following the fight on June 30.
The best Civil War movie ever made finally gets its due
On Sunday and on July 24, Turner Classic Movies and Fathom Events are presenting big-screen showings in theaters nationwide of “Glory,” in honor of the 30-year anniversary of its release. The greatest movie ever made about the American Civil War, “Glory” was the first and, with the exception of Steven Spielberg’s “Lincoln,” the only film that eschewed romanticism to reveal what the war was really about.
The story is told through the eyes of one of the first regiments of African American soldiers. Almost from the time the first shots were fired at Fort Sumter, S.C., the issue of black soldiers in the Union army was hotly debated. On Jan. 1, 1863, as the country faced the third year of the Civil War, Abraham Lincoln issued the Emancipation Proclamation, rapidly accelerating the process of putting black men into federal blue.