Poll of the Week
A Monmouth University poll released Tuesday generated a ton of headlines. It found Republican Ed Gillespie leading the Virginia gubernatorial race by 1 percentage point, 48 percent to 47 percent. Democrat Ralph Northam has led in most surveys of the race, and if Republicans win Virginia with President Trump so unpopular, you can expect a full-blown freakout among Democrats. Adding to the confusion: Quinnipiac University released a poll on Wednesday showing Northam up 14 percentage points, 53 percent to 39 percent.
So what the heck is going on in the Virginia governor’s race? Nothing. The split between the Monmouth and Quinnipiac results is big, but it’s not unnatural. In fact, it’s a sign that pollsters are doing their job.
Polling averages work best when pollsters are working independently. You have different pollsters using different methods and making different estimates of the electorate, and you get a more accurate picture of the race by averaging their results together than by looking at any individual poll. It’s kind of like the old “wisdom of the crowd” principle.
That doesn’t work if pollsters “herd” — which my colleague Nate Silver defined as “the tendency of polling firms to produce results that closely match one another.” When pollsters release results that are closer to each other than is statistically plausible, it may make individual polls more accurate, but it makes the average less so. That is, there should be a big spread among polls of the same race. Unfortunately, herding happens, particularly as Election Day approaches.
But it doesn’t seem be happening in Virginia. The October average of Virginia gubernatorial surveys has Northam leading Gillespie by 7 percentage points, 50 percent to 43 percent:
|POLLSTER||END DATE||SAMPLE SIZE||NORTHAM||GILLESPIE||DIFF.|
|Fox News||Oct. 17||697||49%||42%||+7|
|Quinnipiac University||Oct. 17||1,088||53||39||+14|
|Monmouth University||Oct. 16||408||47||48||-1|
|Christopher Newport University||Oct. 13||642||48||44||+4|
|Roanoke College||Oct 13||607||50||44||+6|
|Emerson College||Oct. 7||318||49||44||+5|
|Christopher Newport University||Oct. 6||928||49||42||+7|
|Washington Post/GMU||Oct. 2||720||53||40||+13|
With an average Northam lead of about 7 points, sampling error alone1 suggests that some polls should find Gillespie up by a little and some polls should find Northam with double-digit advantages. That’s exactly what we’re getting.
There are also additional sources of error besides sampling error, which should make highly divergent results even more likely. For example, some pollsters in Virginia are relying on random digit dialing and asking whether someone is a voter, while others are calling lists of known voters. Put another way, pollsters aren’t necessarily calling the same universe of people. Pollsters also weight their results to ensure the population polled is representative of what they believe the voting electorate will look like. This weighting makes the results more accurate on average, but it also adds to the possible error in the polls.
The bottom line: The results we’re getting in Virginia are totally normal and fine, and people shouldn’t be demanding that every poll show the exact same thing.
Other polling nuggets
- Democrat Doug Jones and Republican Roy Moore are tied at 42 percent in a Fox News poll of the Alabama special Senate election, but Moore held a 51 percent to 40 percent advantage in a Strategy Research survey. The three previous polls of the race taken since last month’s GOP primary runoff had Moore up by between 5 and 8 points.
- Democrat Phil Murphy looks like he will be New Jersey’s next governor. He held leads of between 14 and 18 points over Republican Kim Guadagno in surveys released over the last week by Fairleigh Dickinson University, Fox News and Stockton University.
- Trump’s attacks on the “fake news” media seem to be working, as 46 percent of voters told Morning Consult that they believe the media makes up stories about Trump and his administration. Republicans are far more likely than Democrats to believe this (76 percent to 20 percent).
- The percentage of Americans who consider themselves NFL fans is down to 57 percent, from 67 percent in 2012, according to Gallup. That decline is mostly attributable to declining fandom among independents and Republicans.
- The percentage of Americans who approve of Trump’s response to recent hurricanes dropped from 64 percent last month — before Hurricane Maria devastated Puerto Rico and the Virgin Islands — to 44 percent this month, per a CNN poll.
- A majority of women, 54 percent, said in an ABC News/Washington Post survey that a man had made an unwanted and inappropriate sexual advance towards them. Fifty-eight percent of those who said the incident occurred at work told the pollster that they didn’t report it.
- According to a new Pew Research Center survey on gender equality (it has lots of interesting findings, check it out in full), 48 percent of Americans feel that more women working outside the home and more men taking on household chores and childcare have allowed women to lead more satisfying lives. Just 36 percent of Republicans feel that way, compared with 58 percent of Democrats.
- 14 percent of Americans said men were better emotionally suited to politics than women, according to an Emerson College survey. Forty years ago, that number was 47 percent.
- Only 19 percent of Americans told YouGov that they’ve never lied to their significant other. A near-majority, 49 percent, said they have lied more than once.
- As we head into tonight’s pivotal Game 6 of the American League Championship Series, this New York Yankee hater takes some refuge in the fact that only 15 percent of baseball fans believe the Yankees are most likely to win the World Series, per a YouGov survey.
Trump’s job approval rating
The generic ballot
Democrats are ahead of Republicans 48.1 percent to 37.6 percent on the generic congressional ballot. That’s a wider advantage than Democrats held last week, when they were ahead 46.6 percent to 39.0 percent.