Article ≠ study. There are far different standards as well there should be.
But this isn’t a regular article – it’s an “investigative report”, which rather than simply reporting straight facts, seeks to draw a conclusion from them. Seriously, look at the title of the article and series, and lines like:
“Taken together, they paint the patchwork picture of a quiet phenomenon, tracing a cross-country trail of death and heartbreak.”
…and tell me they’re not trying to create a conclusion.
When anecdotal evidence is all that exists, that is all we can look at.
The plural of anecdote is not data. And you draw conclusions, as the Times does here, from anecdotal data at your peril. For example, there’s “anecdotal evidence” aplenty that women are bad drivers, right? It’s not true, but stories exist, and we’ve all heard them. Think the Times would run a probing series on it? Doubtful. Also, as pointed out above, there is anecdotal evidence that war veterans can be strengthened by their experiences, but the Times doesn’t do a series on that.
This thing stinks on so many levels. The sample size (121 cases) is so small that the inclusion of marginal cases (drag racing, self-defense, drug raids, etc.) skews the “data set” to beyond inconclusive.
Sure, it’s nice that the Times acknowledges the shakiness of their own study, and even gives a whole single paragraph to the Pentagon’s substantive critique of it. That doesn’t change the fact that the basis for this “investigative report” is so thin that it doesn’t support the conclusions it tries to draw.
Finally, I don’t give a damn whether the story is getting play in lefty blogs. It’s getting play in the NEW YORK TIMES, and the fact is that people reading this article are likely to take away two things: the wrenching stories and the phrase “89% increase”, and the poisonous stereotype of the ticking-bomb vet will get a boost.
