Mythology posing as journalism

It's unusual to see a story with such a low ratio of facts (pretty much none, actually) to self-serving cultural generalizations:

Manners help make a miracle
Southern politeness may have saved lives amid the chaos after Charlotte-bound Flight 1549 crashed in the Hudson River, says a new book
(Notice that a couple of different levels of knowledge are in play here? The hed asserts the role of manners in the "miracle" as a fact; the deck attributes the "Southern politeness" bit to a "new book." Neither one quite corresponds with the teaser on the Web front, shown at lower right: "passengers say." )

Nine months after skidding to salvation on the icy Hudson River, passengers of Charlotte-bound Flight 1549 have their say in a book being released today, one that contains an intriguing hypothesis:

A key reason that evacuation of the jetliner went so smoothly was because it was largely populated by Southerners.

This isn't a "hypothesis" in any meaningful sense (and if there's any evidence that it's "contained" in the book -- for example, a quote from the book to the effect that Southerners made the difference -- the story doesn't report it).

While there were sporadic acts of ugliness in the chaos after the splashdown -- at least two passengers said their seat-cushion floatation devices were snatched by others -- an inbred politeness seemed to be at work, says William Prochnau, author of "Miracle on the Hudson: The Survivors of Flight 1549."

"There's something to that," says Prochnau, who assembled the stories of 118 of the 150 passengers for the book co-written by his wife, Laura Parker. Prochnau said in researching the book, they learned that whenever someone felt a rising sense of panic, others in the group settled them genteelly and guided them through the ordeal. About 100 of the passengers were from the South.

"Something to" what? The idea that some sort of "inbred"* politeness led some people to calm other people down? Or to the reporter's point that about two-thirds of the passengers were "from the South"?

Anyway, to get back to the "hypothesis" the reporter is trying to unearth -- broadly, it's something like "the residence of disaster victims affects their likelihood of panic," but more directionally, it's something like "Southerners are less likely than other Americans to panic during a disaster." It's going to be tricky to test. I don't know about you, but my IRB is going to take a dim view of a proposal to stage airline crashes to examine the effect of regional residence on evacuation effectiveness, so we'd have to use a "quasi-experiment": looking at other crashes to see what seems to be associated with survival.

Do you see some confounding variables emerging here? Aside from the training and performance of the flight crew (both central to having intact people and a fairly intact plane to escape from), how do we make sure we're measuring only what we claim to measure? Do you have natural Southern politeness if your ticket says you're from Waxhaw but you moved there from Jersey three years ago? Do we know it's a regional characteristic, not a national one? Is the performance of "most" passengers essential, or do just a few do the bulk of the moving, calming and organizing?** What effect did the local rescue crews (or the tax structure that supports them, or their post-9/11 training) have?

Turn to the outcome for a second, because to have a research hypothesis (rather than a "null"), it has to be different from what we'd normally expect. I'm not convinced that it is. Disaster behavior has been studied for decades, and one very consistent conclusion is that lots of what people believe is wrong. Panic "is a very rare phenomenon in American disasters," Dennis Wenger wrote back in 1985. "Mass panic is so rare that the few verified events are somewhat historical milestones. ... The problem is not usually panic, it is exactly the opposite, i.e., getting people to move."

I can see why we wrote a story; it's a very local disaster, despite its having happened in New York, and the appearance of the passengers' own stories in book form is worth noting. But I can't see running this story, except as a way of making the audience -- at least, the part that isn't yelling at the other part about whether northerners or southerners are more intolerant -- feel good about itself in general: We weren't there, but we would have done OK if we had been.

The story gives hypothesis-testing a bad name; it makes the whole process sound kinda woo-woo, as if there was no difference between good, testable ideas and angels-on-furry-wings-saved-the-plane ones. And it reinforces one of the core myths about disasters, thus not helping anybody except the people who make bad disaster movies.

And -- how many times do we have to say this? Cut it out with the "miracles" in headlines.

* Not to be picky or anything, but that's a remarkably insensitive choice of words when the topic is Southern heritage.
** This may be overextending, but isn't that sort of like what S.L.A. Marshall reported about infantry action?