Dilbert’s Creator Is a Sad Racist Troll (And Not Very Good at Reading Polls)

 

Scott Adams, the one-time Trump-supporting creator of the Dilbert comic strip, has tanked his career. In the last few weeks, Adams’s strip was pulled from over 300 newspapers, his syndicator dropped him, and he lost a book deal with his publisher.

And all because of his idiotic interpretation of an absurd opinion poll by the grifters at Rasmussen Reports — the GOP-friendly outfit that somehow managed to be the only folks ever to find Donald Trump with majority-favorable approval ratings when he was in office.

According to Rasmussen, only 53 percent of Black folks answered affirmatively when asked if “it’s OK to be white.” Twenty-six percent responded that it wasn’t, while the remainder weren’t sure.

To Adams, this means that roughly half of Black people are hostile or at least ambivalent to white people’s very existence.

As such, as Adams put it, Blacks constitute a “hate group,” and whites should “keep the hell away from them” at all costs.

The sound of Adams’s income drying up faster than roadkill on a West Texas highway is truly beautiful. He has earned his cancellation.

Not merely for being a pathetic bigot who generalizes about 42 million Black people based on what one poll says a minority of them believe, but more importantly, because of his utter inability to read the poll and see how awful it was before going on his tirade.

Anyone who would make such hateful comments based on evidence as weak as this was clearly just looking for a reason to do some racism and jumped at the first opportunity.

He’s like the guy who always wanted to yell the n-word and takes some news item about a Black person who commits a horrible crime as his excuse for finally doing it.

Yeah, to hell with him. Let him crowdsource groceries. Seriously.

Others have written about the problem with the “It’s OK to be white” phraseology, and I’ve also explained my take on it elsewhere. As a phrase that originated as a trolling device on 4Chan among white nationalists, asking someone how they feel about it is complicated. Opposition or ambivalence could easily reflect recognition of the phrase’s troubled genesis or just confusion as to what, precisely, was being asked.

It’s a weird question, after all, especially in the world of opinion polling, where reputable pollsters have all kinds of ways to get at levels of racial bias without utilizing meme-speak created by extremely online gamers and incels for the purpose.

The folks at Rasmussen apparently have a different label for these types.

They call them independent contractors.

In any event, and without belaboring the point about why the phrase is troubling, let’s just leave it at this:

The poll is trash, and anyone who finds it persuasive is embarrassingly bad at understanding basic social science concepts like what a representative sample is or how standard error works.

So let’s do some basics, shall we?

First, please note that the Rasmussen poll in question — the one that impressed the aging cartoonist — surveyed 1000 people.

Don’t misunderstand: that can be a decent sample size for a survey, but if the 1000 people are then split among many different demographic groups, who often look at ideological and political matters differently, that decent sample size becomes several smaller samples, and that’s where the trouble begins.

And that’s what happened here.

This 1000-person poll not only asked an odd, ideologically-loaded question, it asked it of a group that only included, according to a tweet from Rasmussen, 13 percent Black participants, which would be 130 people.

Here’s the tweet:

Or perhaps it was only 12 percent if one believes the Washington Post’s reading of the Rasmussen data, which places the number of Blacks in the poll at 117.

Either Rasmussen is rounding up the numbers very oddly in their Tweet (as 117 is never 13 percent of 1000), or the Post is misreading the chart.

Honestly, I don’t know which it is, and I don’t care enough to spend the $19.95 Rasmussen demands as a fee to see their methodology. Because either way, the number is laughable. Let’s just take the Rasmussen tweet at its word and assume they talked to 130 whole Black people.

So, math, please:

If 26 percent of 130 people say something — in this case, that they disagree with the statement “it’s OK to be white” — that means a blistering 34 people believe this position.

For fun, let’s assume no one was confused by the question, no one knew about the racist origins of the phrase, and all genuinely hate white people the way Scott Adams seems to think they do.

Hell, maybe these were all Black Hebrew Israelites who took time off from yelling at white people about their “cracker babies” on the streets of Manhattan to answer a Rasmussen poll. OK, fine.

Now, there are 42 million Black people in this country, about 30 million of whom are adults and thus among the universe of people typically polled for things like this.

Some more math, please.

Thirty-four people constitute roughly 0.000081 percent of Black people (or 0.000113 percent of Black adults) who said they feel this way in one poll, or a little more than one ten-thousandth of a percent. Is it possible that these 34 people represent 26 percent of all Black folks — around 11 million?

Well, sure, and it’s also possible that Scott Adams is not a flaming racist asshat, but I wouldn’t bet my life on it.

Look, it’s simple. When your sample size is small — and 130 (or 117) people are a fucking joke in statistical terms — your standard error will be huge. The standard error tells us how different the actual population average of something is likely to be relative to the average you obtain from a mere population sample.

The larger the sample, the smaller the error; the smaller the sample, the larger the error. It’s pretty basic.

To obtain a rigorous 95 percent confidence level that your sample average is representative of the whole — and that the larger average would fall within a very close range of whatever number you found — you typically need a large sample, or you must admit to a massive standard error. Rasmussen does neither.

With 130 people, you could easily run this same test repeatedly and get wildly different answers each time.

On the other hand, when you do a population survey of tens of thousands or hundreds of thousands (like the Labor Department, Census Bureau, Justice Department, or genuine social scientists do), you can extrapolate to a nation of 330 million pretty accurately, so long as other methodological formalities are followed.

Scott Adams either doesn’t know any of this or doesn’t give a shit.

He saw a top-line result, pushed out by an overtly right-wing pollster, using deliberately troll-bait language in its question, which it asked fewer Black people than you can find in the first two rows of a small Black church on Sunday, and decided, that’s it, he’s done with Black folks.

Cool boomer, cool.

And now the rest of us, Black, white, and otherwise, can be done with you and your stale-ass comic strip.

It wasn’t ever that funny in the first place.

—

This post was previously published on Medium.com.

***

—

Photo credit: Ol.v!er [H2vPk] on Flickr under CC License