— HuffPost (@HuffPost) November 7, 2016
Of course you do. Our memories are longer than the media may like to believe.
How quickly all the “models” unraveled, the pollsters left scratching their heads again, the exact same thing that happened with Brexit just a few months earlier was happening all over.
Inside the industry, data scientists have been discussing their methodologies and figuring out how and why they’ve been talking to the wrong people for so long. Unfortunately for those concerned with accuracy, their management teams and sales departments have other ideas.
Cue today’s “disapproval rating” stories about President Trump, conveniently dropping before Labor Day in order to dominate this weekend’s news cycle.
Nobody really speaks of “disapproval ratings”, by the way. It’s usually been “approval ratings”, even if in the negative. Many Western leaders have disapproval ratings well below those of President Trump, by the way. Another thing that is rarely ever mentioned.
The Washington Post on Friday led with: “Trump’s disapproval rating hits a high point in new poll”.
Scott Clement, who runs the Washington Post’s “independent” in-house pollster (how does that work, when WaPo is clear about its agenda?) reports that “a majority of the public has turned against Trump and is on guard against his efforts to influence the Justice Department and special counsel Robert S. Mueller III’s wide-ranging probe.”
Here’s why they’re wrong … again:
The poll isn’t independent. Let’s be clear about that. While Clement and WaPo might have the visage of neutrality to some, the firms they use to do the behind-the-scenes magic simply aren’t. As I’ve mentioned before, WaPo uses a company called Abt Associates.
Abt counts its board members as staffers that have worked for Bob Dole, Bill Clinton, Barack Obama, Planned Parenthood, the World Bank, and Deloitte. I described it previously as a “microcosm of Davos”.
Anne Marie Slaughter — one of their two board members with a focus on the political side of things — was President Obama’s Director of Policy Planning in the U.S. State Department.
Random Digit Dialing
There aren’t a lot of low-cost options to get around this, so most pollsters trying to reach people on the telephone use it. WaPo discloses this if you bother to go and read through the data sets (which most news consumers don’t).
“A dual-frame landline and cell phone telephone sample was generated using Random Digit Dialing procedures by Survey Sampling International (SSI). Interviewers called landlines and cellular mobile phone numbers, first requesting to speak with the youngest adult male or female at home. The final sample included 350 interviews completed on landlines and 653 interviews completed via cellular phones, including 422 with adults in cell-phone-only households.”
Well, who do you think you’re going to reach more of if you dominate your samples with cell phones and cell-phone-only households? Younger people. But just to make sure of it, they also then ask for the youngest person in the household — who trend more liberal.
More Democrats/ Independents
It doesn’t really matter how young or old the respondents are if you’re dominating your samples with leftists, anyway.
That’s why Clements will point to the crosstabs in the poll and say, ‘No, look, older people don’t like Trump either!’ Because WaPo/Abt has stacked the deck with moderate and leftists.
Their data reveals that out of 1,003 adults they spoke with, 33 percent identified as Democrats, 37 percent identified as Independents, and just 25 percent identified as Republicans.
You’d think after 2016 — when pollsters were left humiliated by failing to speak to people who identified as Republican or with conservative values — they’d have learned their lessons. Apparently not.
Perhaps it is for these reasons and more that Christopher Hitchens once decried so much public polling as junk used “not to interpret the world but to change it.”
In Harper’s magazine in 1992 he said of the Washington Post and CNN specifically:
“Poll, poll, poll. Try reading a news story of watching one aired on TV without encountering the word. Readers of the Washington Post of 5 February, to take but one example, were offered seven stories on the front page, and of these, three — about the pessimism of Washington’s residents, the souring on capitalism, and it should go without saying, the Clinton campaign — were based on polls. Not content to wait a day or two for results, the Cable News Network (CNN) pioneers the viewer phone-in poll, inviting nightly news-watchers to glimpse a minute-long story, then dial an instant opinion.
“Opinion polling was born out of a struggle not to discover the public mind but to master it. It was a weapon in the early wards to thwart organized labor in the battle against Populism, and it later became a rather favorite in the arsenal of ‘mass psychology’ parties of the European right…
“…Polls are deployed only when they might prove useful — that is, helpful to the powers that be in their question to maintain their position and influence. Indeed, the polling industry is a powerful ally of depoliticization and its counterpart which is consensus.”
It was ever thus.
Raheem Kassam is a Lincoln fellow at the Claremont Institute and the author of two bestselling books: No Go Zones: How Shariah Law is Coming to a Neighborhood Near You and Enoch Was Right: Rivers of Blood 50 Years On
The views and opinions expressed in this commentary are those of the author and do not reflect the official position of The Daily Caller.