The coming American coup…a preposterous notion taking root in our collective mentality? Impression 4.


I’d prefer to think it wasn’t my idea. But the fact such a preposterous thing is even being bantered about in the mainstream (albeit foreign) news shows that the seed of such a notion rests latent in our collective American soul.


These are times of upheaval.


American Coup, #3 (9/10/15)


American Coup, #2 (9/6/15)


American Coup, #1 (9/5/15)


And now, in The Guardian, a YouGov online survey tells us that 29% of Americans would support a theoretical coup in America given dire enough circumstances.


Such a loose online poll is naturally burdened with many statistical inaccuracies and structural shortcomings, but as far as I’m concerned, most noteworthy here are not the results of the lackadaisical poll, per se, but rather, the fact that such a proposition has legs. Most cultural threads of theory tend to start with the faintest kernels of conjectural what-if’s.


Almost a third of Americans could imagine supporting a military coup against their own government, according to a new poll.


The YouGov survey showed 29% of Americans could imagine supporting a coup. Yet, 41% said they could not imagine supporting such an event.


YouGov, which conducts internet polls about “politics, public affairs, products, brands and other topics of general interest”, surveyed 1,000 people online on the issue.


They found that 43% of Republicans would support a military coup in certain instances, while only 20% of Democrats and 29% of independents would.


The overall numbers increased when participants were “asked whether they would hypothetically support the military stepping in to take control from a civilian government which is beginning to violate the constitution”. 43% said yes to this, and 29% said no.


Abraham Wyner, director of the undergraduate program in statistics at the Wharton School of the University of Pennsylvania, said that online polls were “worse than just about any other way you can put together a poll” because they were prone to selection bias, meaning proper randomization was not achieved and the sample was not representative of the population – since people can choose to participate.


“People who are participating in an online poll are generally attracted to that poll because of some variable, some characteristic which is connected typically to one outcome or the other in that poll,” he said.



Of course. Widespread cultural statistical ignorance is appalling and it is important that Wyner’s observation be inserted as a cautionary check into this story in order to hedge against any sort of the marauding hysteria. The news outlet feels compelled to take the high road in tone and self-restraint.


Those are qualities thrown out the window once ideas take root, however. As we are witnessing.