National Geographic and world longevity researchers identify pockets around the world where people lived measurably better.
They found five Blue Zones, places where people reach age 100 at rates up to 10 times greater than in the United States.
That reveals the problem of pre-selection.
There was a virtual certainty that you would find a set of geographic areas which, if drawn based on a criteria, would meet that criteria — in this case “people who live an average of 100 years”.
The question is whether they can identify real factors which led to the measured effect, or if it is just randomness.
To illustrate, imagine you took 1000 pennies, and threw them across a gym floor. Odds are 50/50 whether any penny lands heads or tails.
Now, take a hulu hoop. Walk around the gym floor. Try to find 5 locations where, when you check all the pennies in the hulu hoop, over 80% are heads.
You will likely find more than 5 such locations. And there will be no reason — it’s just how random events happen.