Can We Trust Election Polls?

              Now that both parties have competing tickets from which we can choose who will occupy the White House (and the Vice President’s mansion) for the next four years, I’m going to spend some time trying to figure out what the pre-election polls really tell us, as well as what they don’t.

              I’m doing this in response to the Trump campaign’s narrative that Biden’s consistent lead in both the national, aggregate polls and the in-play state polls is nothing more than a combination of: a) mainstream-media lying, and, b) an under-representation of pro-Trump enthusiasm which will easily close the gap. This latter narrative, of course, being a riff on how Trump has ignited and enlarged a political ‘movement’ to ‘make America great again,’ the likes of which have never previously been seen.

              In certain respects, Trump’s 2016 campaign was a very different and unique approach to national political campaigns.  He held large, in-person rallies every day, sometimes two or even three daily mass events in multiple states. He promoted these rallies with an aggressive, social media assault which was never matched or even anticipated by the Hillary campaign. And he cemented together an independent media ‘team’ comprised of FOX noisemakers like Hannity, Carlson and Ingraham, AM shock-jocks like Limbaugh and Beck (plus local wannabees), all of whom said exactly whatever they were told to say via faxes from Trump’s headquarters and the RNC.

              Most important, he made sure that everyone who attended his rallies was told again and again that they were part of a new and important ‘movement,’ (people like to think of themselves as coming together to create something special) and that they were also members of an ‘oppressed minority,’ i.e., folks ‘left behind’ by the un-American elitists who had sold the country out to Deep State enemies both here and abroad. And in case you didn’t know it, people love to feel that they are members of an oppressed, minority group.

              There’s only one little problem with this narrative. In 2004, the Republican candidate, who happened to be the incumbent, racked up 62,039,073 votes. Twelve years later, Trump’s ‘movement’ delivered 62,984,828 votes. Between those two elections, the U.S. population grew by 10 percent, the Trump campaign increased GOP turnout by a whole, big 1.5 percent. That’s some ‘movement.’ Yea, right.

              One of the big problems in evaluating election polls is whether someone who supports a particular candidate will actually show up and vote on election day. It’s what is known in the polling industry as ‘registered’ versus ‘likely’ voters. When you’re still 3 months away from the date of  vote, it’s too early to say for sure that someone who says he or she is voting for this or that candidate will actually appear on election day. So the polls conducted in August and September tend to use respondents from both groups. Once you get into the last 60 days or so, the ‘registered’ voters as poll respondents tend to disappear.

              Allegedly, the shift away from registered voters and towards likely voters is done because the latter category tends to be biased in favor of blue voters.  Gallup explains it like this: “For a long time in American politics, Republicans have represented those voters who are in the majority (or plurality) demographically. A voter who is white, straight, suburban, Christian, middle-aged and middle income is quite likely to be a Republican. However, a voter who deviates from that profile in some respect — for instance, by being black, Hispanic, gay, Jewish, an atheist, young or by having completed an education beyond an undergraduate degree — is likely to be a Democrat. The majority of Americans deviates from that profile in at least one of those respects. That’s part of why a larger share of Americans have long identified as Democrats rather than Republicans.

              The only problem with this analysis is that there were 12 national polls published a week or less before the 2016 election, and 11 of those polls used likely voters for their sample and every one of these polls had Clinton ahead by 3 to 7 points. And in fact, Hillary ended up by getting 4% more votes than Trump.

              Which meant that the 2016 national polls were correct. Which is why tomorrow I’ll look at the 2016 state-level polls.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s