What is the State of Polling in 2020?
JT collaborators Bill McInturff and Micah Roberts, partners at Public Opinion Strategies (POS), share their thoughts on the state of polling ahead of the 2020 election.
The 2020 election is in full swing, so JT reached our to our frequent collaborators, Bill McInturff and Micah Roberts, partners at Public Opinion Strategies (POS), about the state of polling in this critical election cycle. POS is a leading national survey research firm based in Alexandria, Virginia, that we have worked with on dozens of public affairs efforts over the years. This is the first of several posts on this subject.
JT: What is the state of polling going into the 2020 election? Are polls still reliable?
POS: What we do has always been hard — and, to be clear, it has gotten a lot more difficult because of a number of factors. But, overall the state of polling is very strong in the U.S.
Pew Research Center wrote a report entitled “Election 2020 Field Guide”, and here’s their bottom line:
“Here’s a myth that can be set aside right at the start. Polling is not ‘broken.’ Well-designed and carefully administered surveys still work, and there’s plenty of evidence to back this up.”
Separately, Nate Silver of FiveThirtyEight made a similar point, saying: “Polling is about as accurate as it has been historically and traditional polling methods continue to be more accurate than non-traditional ones.”
JT: So, you’re sounding pretty positive that things are A-OK with polling. That seems hard for many to digest after 2016. Why are things better in 2020?
POS: Well, there have already been some questions after Iowa and New Hampshire, and we are sure that at some point during the Democratic nominating process, “something” is going to happen to cause a rash of, “OMG, the polls are broken” coverage. For instance, after the results in Iowa and New Hampshire, some have said that Joe Biden’s strength was inflated through 2019. Here’s why that may not be a fair critique:
First, there is the ethnic representativeness of those two early states and the fact that Biden’s strength among African Americans is one of the most consistent aspects of his “front-runner status.”
But also, Joe Biden is only getting 8% of the vote among Democrats ages 34 and under. Sanders receives only 10% of the vote among primary voters over 65 years old. That’s as big as it gets in terms of differences by age. Even small changes by age in terms of composition of the electorate are going to have a critical impact on the final result in a state.
Before Iowa, almost 7 out of 10 Democratic primary voters nationally said they were enthusiastic or comfortable with either Biden, Sanders, or Warren, and a majority had not made a definite vote choice.
As we saw in the 2020 Iowa and New Hampshire primary contests (and this was also true in Michigan, Wisconsin and across the Rust Belt in 2016), voters can break late towards a candidate and even polls three to five days earlier will miss this last minute decision-making.
Winning a state quickly moves voters towards the winning candidate in ways that can be breathtaking to watch in terms of quick changes in ballot support.
These are the realities of a long, multi-candidate presidential primary and do not necessarily mean that we should anticipate problems with polling leading onto the general election in the fall.
PUBLIC AFFAIRS
Can a poll help advance your business or policy goals?