Calm Yourselves Media
You’re scaring the children
Polling is so much hot garbage, and it has been so for at least ten years. Sweet Jeebus though, you wouldn’t know that from watching cable news this week. Devoid of sentient thought, they are all treating the results from a patently ridiculous poll last week that showed Trump winning the swing states by a few points like a precise chemical analysis from a mass spectrometer instead of something one step removed from a haruspex digging through the entrails of a ram in ancient Rome to see what the harvest was going to be like.
First off, let’s talk methodology. Pollsters call thousands of people in order to get many fewer than that number to actually talk to them and answer questions. Then they massage those responses to try to resemble the electorate they *think* is going to actually turn out and vote. For example, if they think 10% of the electorate is going to be Black, but only 5% of their respondents were, they will weight those responses double. Right away you can see two big problems. If the electorate turned out to be 15% Black after the votes are cast, the poll is going to be a poor sample. Also, the more weight you have to assign to a smaller sample increases the chances of wildly magnifying an outlier. If your poll only contacted Kanye West and Candice Owens, you’re going to come away with data that suggests that Black Americans are conservative and deeply crazy.
In 2016, some weird things happened. Some voters were (rightfully) ashamed about planning to vote for Donald Trump so they kept that to themselves, and the concept of the “shy Trump voter” was born. Also, a not-insignificant number of infrequent voters that fit a profile, blue collar, insecure, angry, racist… lets just call it MAGA for short, decided to turn out in unexpectedly higher numbers while generally not talking to pollsters.
This burned the polling industry as a whole and they’ve been making things worse even since trying to compensate for what was most likely a once-in-a-lifetime fluke. The Shy Trump Voters turned into reluctant Democratic voters- the Blue Team wouldn’t have won in 2018, 2020, and 2022/23 if that hadn’t been the case, but the pollsters and the media still insist that there is a mythical 5% of voters out there who just *adore* Trump, but can’t be contacted, polled, or located in any way.
Additionally, pollsters love to ask two questions: “Do you have a favorable or unfavorable view of the President?” and “Do you think the country is heading in the right direction or on the wrong track?”
I don’t know any other way to say it, but these questions are dumb and should be retired instead of forming the basis for all political journalism in this country. Asking if one views the president unfavorably doesn’t account for ideology or causation. Because answering a poll question is a (largely meaningless) thought exercise, and very different from casting a vote, people tend to think for a moment if there were anything they would like the sitting president to do differently. That answer is often “yes”, and people take the free chance to express disapproval over something. Yet many times half the people expressing disapproval are doing so for *opposite* reasons. A deeply indebted college grad may be mad that her student loans were not forgiven, whereas an embittered plumber may be mad that it was even considered.
Same problem for right direction/wrong track questions. Almost all Americans are upset at the near-daily mass shootings in this country. Most people, if asked the direction question in a poll the day of a high-profile mass shooting, would say our country is on the wrong track. Yet within that group, something like a third of them would say we have mass shootings because of atheists and video games and drag queens instead of the availability of military hardware to the mentally ill. A majority of Americans agreeing that the country is on the wrong track does not indicate what they think the right track might be or how they are going to vote.
Yet another problem is the lazy list of standard issues force-fed to the public. Instead of any sort of qualitative interview to see what people care about, respondents are sometimes asked to rank things like “economy,” “immigration,” and “crime.” Fox New famously tends to only report, extremely negatively, on a handful of issues like this, that is cherry-picked topics on which they believe conservatives hold an advantage. Then they poll their own viewers about their top concerns. *Shockingly* viewers report back that they care almost exclusively about the few issues Fox News told them to worry about.
Where you see all this come to a head is on election night. In 2022, and earlier this week, the pundits scratch their heads and say things like, “65% of voters said the economy was their number one concern and that they trust Republicans more to handle it, but the votes are counted tonight and it looks like Democrats will take 54% of the vote, offering a stinging rebuke to the GOP for reasons we can’t begin to fathom… probably something to do with Tiktok and Wokeness?”
Rick Santorum, one of just the worst people to plague the body politic, expressed surprise and dismay at Tuesday’s election results saying that direct democracy was a bad way to do things. Gee, I don’t know Ricky, I think lying to people, telling them what they should care about, and then insulting them when they don’t agree is worse.
