“Not a real cluster” (Dominican Republic deaths)


If something is truly random, it will sometimes show hints of an apparent pattern. For example, 2, 12, and 22 might be part of same winning PowerBall combination, or the same defense industry worker may be picked for a drug test three times in a row.

These and similar instances sometimes lead observers to infer that these streaks or clusters are evidence of the phenomenon being nonrandom. So a frustrated lottery player may attempt to find a nonexistent way to beat the system, or a jaded employee may be convinced management has it in for him.

Clustering illusions are amusingly demonstrated by the Texas Sharpshooter Fallacy. In this tale, a man fires several dozen pistol rounds into the side of a barn, picks out a number of holes close to each other, and claims that’s where he was aiming. When done in real life, an observer can wrongly think there is a connection in random phenomenon, be it related to crime, economics, or sports.

It’s too early to know, but this misinterpretation may be manifesting itself regarding American deaths in the Dominican Republic. Nine U.S. citizens have died there over the last 12 months, usually the result of a heart attack, though there is some concern that poisoning may be involved.

Metabunk’s Mick West, who saw the clustering illusion first-hand when he developed landscapes for computer games, writes that heart attacks are scattered throughout the population, but also correlate to demographic factors. This means heart attack numbers will generate both genuine and illusory clusters. With regard to the specific subset of U.S. tourists in the Caribbean, which type of cluster is in play?

West further makes note of the Frequency Illusion, where one pays more attention to something if they’ve heard of it lately or been impacted by it. For example, a motorist who buys a new Pontiac will likely notice other such vehicles more often than they would if they had not made the purchase.

So someone hearing of a handful of deaths afflicting those of a specific nationality in a certain place will likely draw a conclusion about what this means even if that’s not a justified reaction. West writes, “People have died of heart attacks, and even of alcohol poisoning, in the Dominican Republic before, and they will again. Is it any more common this year?

He crunches some numbers and notes that .1 percent of U.S. adults died from sudden cardiac arrest in a 12-month period. Meanwhile, about 2.7 million Americans go there each year. Assuming an average trip time of seven days, this would extrapolates to an expected 52 Americans dying of a heart attack in the Dominican Republic each year.

So while there could be something malevolent going on or some yet-to-be identified virus in the Santo Domingo vicinity, the numbers at this point support no such notion. If you have a trip planned there, go ahead. It’s likely to be as safe as always been and besides, there are clusters showing that people die unexpectedly while sitting in their living room.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s