Dear Pew Research Center, Gallup, and other pollsters who poll people,
I am ready to be polled. I’m ready, willing, and eager. So please, I beg of you, poll me. I have been waiting by the phone since this crazy election season began nearly a year ago, and alas, it’s always been offers to lower my nonexistent credit card debt or win a “free cruise.” It’s almost as insulting as the Donald is! I even keep my ringer “on,” which is a rare and risky move in Millennial culture! For months, I’ve watched your polls all over FiveThirtyEight, CNN, Trump’s Twitter, and on every nightly news show. Every time you’ve shown Trump pulling ahead, I’ve screamed into the abyss, “Nobody asked me!”
In fact, I’m a polling virgin! I have been old enough to vote and the owner of a working phone for two national elections prior to this one and I’ve never heard your sweet voices ask me which candidate I would be supporting. If by some chance you’re taking backup data, it was Obama, then Obama again. Also “no” on Prop 8.
My eight years of silence from these polling places made me wonder if anybody is ever actually polled. So, I conducted my own super-scientific poll. I asked my Twitter followers, “Have you been polled for a political poll?” I also clarified that if my poll about political polling was a responder’s first political poll, then they must write “no,” they have never been polled about politics for a political poll. The results: Of the staggering number of replies (15 people total), 60 percent said they had never been polled, and 40 percent said they had. So, that was...tremendously insightful.
The mystery remained. I started recounting my entire life, trying to think of a single friend or acquaintance saying that a pollster had dialed up their Android. And then I did some research.
According to the Pew Research Center, I personally have a one in 154,000 chance of being polled. They calculated that number by dividing the number of adult Americans (235 million) by the sample size of a typical poll (1,500 people). The most common method of contacting responders is through Random Digit Dialing. Gallup, USA TODAY, and CNN have a much smaller sample (about 500 interviewees per poll), which decreases my chances even smaller. Quinnipiac University (which is apparently in Connecticut) adds screening questions before it polls. So, even if you actually answer the phone, you may not be able to participate, depending on the poll’s needs! For the record, I have suspiciously never heard of anybody attending Quinnipiac U, but I’ll let that slide. Still other polls, like Rasmussen and Opinion Savvy use data from Primary and Voter registration databases. If you are not registered, or did not vote in the primary, it’s unlikely you will be polled.
Adding to the imperfection, many of the issues that legitimate polling places face center around the fact that people no longer have landline phones. Some centers, like Gallup, call both cells and landlines to mix it up. Yet others that are location-based suffer from using cells. For example, if you’ve had the same phone number since 2006, but do not live in the same location as you did in 2006, results could be skewed based on your area code. Pew may think it’s polling voters in Chicago, but really it’s polling voters with Chicago area codes.
Additionally, you actually have to answer your phone when it rings, in order to be polled. In 2016, answering the phone is terrifying and most people press “ignore.” In fact, Pew says that the “nonresponse” rate in 2012 was 91 percent, meaning that only 9 percent of people called about elections actually pick up.
I must admit that even after doing this research, as a registered voter, a phone owner, and avid follower of politics, my feelings are still a little hurt. Sure I’ve only got a .0000065 percent chance of being called, but I thought I was special! Until my polling star rises, I’ll be diligently waiting by the phone and looking for exit pollers as I leave the voting booth this November. And for the record, it’s Hillary, Kamala Harris, and yes to almost all of the California propositions.