The Changing State of Polling

Something I've been thinking a lot about this election cycle is the impact that polls can have on the election itself. This is not particularly surprising, given the role polls played in the 2016 election, and the way many in the media sphere publicly decried them (even though many of these polls weren't actually that inaccurate). But lately I've been thinking much more about the way polls are changing, and how just their mere existence impacts our voting preferences.

To start with, getting good data for election polls has proven to be more challenging in the modern era. Traditionally, polling has relied on live or recorded phone conversations to gather data, a method that has proven more difficult as more individuals rely on cell phones and caller ID, and are less inclined to pick up the phone for an unknown number. And while online polling is likely the method of the future, the results have not yet proven to be successful in generating accurate data. So, for now, we find ourselves in a bit of a weird spot, relying primarily on an older way of collecting information (resulting in perhaps skewed results) while we wait for newer methodologies to become more advanced.

All that said, polls released now are typically more accurate than they ever have been historically. This brings me to my larger item of concern, which is the way in which polling impacts who we vote for. It is widely known that the simple knowledge of who is a front-runner or underdog in a race can largely dictate outcomes, especially considering people like supporting a winner. This becomes doubly complicated when you have a primary system set-up like ours and multiple candidates are running. Say you prefer a progressive candidate who's polling at 5%, but there's a second progressive candidate currently polling at 25%. Because there is currently no ranked choice or runoff voting in presidential primaries, you may decide to ultimately support the progressive candidate polling at 25% because you think they have a better overall shot of winning the nomination, and you much prefer a progressive to a moderate. All of which is to say, you vote for someone who is not in fact your top pick.

While all of the above information I've summarized is very well-known (and has been for at least a couple decades now), I bring it up because recently there seems to have been a significant cultural shift surrounding the conversation of elections and polling. We are certainly more distrustful of polling results now than we used to be--but perhaps our recent dissatisfaction with polling is just misdirected frustration for something else going on. Perhaps what is really ailing us is not the current status of polling, but rather the sentiment that there is something deeply unsettling about the currently state of our election process.