Caveat: I’m not a statistician, but my job requires that I interpret large pools of raw data to interpolate outcomes for even larger systems. Think using 30 respondents and a week of observation/collection to represent 2500-10,000 respondents over the course of a year. The way I collect, analyze, interpret, and present my data is scrutinized heavily from every angle (as it should be) but still very susceptible to biases. Fortunately, I’m super autistic and it’s basically impossible for me to lie so I’m considered really good at what I do.

These surveys are typically done via methods that are prone to significant statistical errors. For the most part they use phone calls, Facebook ads, emails that most people under the age of 65 just delete (or never even see because it’s caught by spam filters). This, combined with the fact that the type of people who will reply to these sort of questionaires are always more right-leaning skews the data even further.

If someone approached you on the street asking you to take a survey, about Pervert Hoover what would you do? Uncle Jerry, who’s known for providing his unsolicited opinions about librulz and the gays, and Aunt Elaine who doesn’t want to get backhanded at home so she just stands there usually count as two respondents.

That’s not even touching the bias almost always introduced by the question and answer formats. “Do you agree with the President’s policy of deporting people with legal residency status due to excercising their First Amendment Rights? (Y/N)” is a totally different outcome to “Do you agree with the President’s national security policies? (Y/N)”

These articles are always written to imply they’ve surveyed everyone in the country, done the math to adjust for selection bias, and written the survey questions to be fair, but time and time again that’s proven to be untrue, or at least inaccurate. The interesting (and honestly the only good news we hear anymore) is that even with all of the cards literally and figuratively stacked in the orange skidmark’s favor, he’s only getting as positive an outcome as he is.

And there’s also the need for clicks and ad revenue.

TL; DR: these things are bullshit and always produce wildly inaccurate results skewed in favor of the dumbest outcomes.

As Kylo Ren, the late great Kylo Ren–we love Kylo, don’t we folks? Ky-lo. As he said to me, he said it. He said, “Sir,” he said, “MORE,” and we all have to agree, don’t we?