Category Archives: Statistics

The Post’s Fact Checker on pre-Roe abortion death rates

four pinocchios

I’m raising this because I found it fascinating when I saw it yesterday, for several reasons. I’m hoping to be able to raise it independently of our respective views on abortion, because this is interesting wherever you are on the spectrum.

That may be overly optimistic on my part, but here goes…

Basically The Washington Post‘s Fact Checker team took a look at the frequent claim from Planned Parenthood and other groups that if abortion bans being enacted in various states succeed in overturning Roe v. Wade, then we’ll be “going to go back in time to a time before Roe when thousands of women died every year.”

Fact Checker ended up giving the claim Four Pinocchios, which is its harshest rating of a falsehood. Having cited that, though, I urge you to read the whole thing, because the numbers are complicated and often murky, so judge the data for yourself.

For my part, I’ll make several observations:

  • As y’all know, I’m a words guy. I try to appreciate the point of view of my numbers-oriented friends out there, even though I think their insistence on reducing everything to digits can lead arguments astray on many issues. In this case, I’ve always looked askance at the numbers, whether someone is claiming the number is high or the number is low — not because I don’t think numbers are important in this instance, but because I think they are unknowable with any precision. You’re trying to count something that happened in the shadows, a dubious exercise at best. So while I see the Post‘s finding as interesting, I don’t necessarily see it as Gospel.
  • Any deaths are too many. Of course as you know (and here’s the only place I’ll refer to my views on abortion) to me every abortion is a death, and a tragedy. If the mother dies as well, then the tragedy is that much more horrific — and yes, more than twice as tragic. It should be society’s adamant goal to prevent that from happening ever. No one’s ideology should get in the way of that.
  • To that point, perhaps the most interesting data points in the piece are these: “In 1972, the number of deaths in the United States from legal abortions was 24 and from illegal abortions 39, according to the CDC.” So aside from the overall number being far, far less than “thousands” (which is the main point of the piece), it turns out that where abortion was legal, there were still more than 60 percent as many deaths as there were where it was illegal. Make of that what you will.
  • That year, 1972, is particularly relevant to the debate, because as the piece points out, a post-Roe America would most likely be most comparable to the time immediately before Roe, rather than to the decades before that (when the estimates of deaths were much higher). The main commonality is this: At that time, abortion was legal in some states and illegal in others. If Roe suddenly disappeared, I expect we’d return to a situation like that one — although my guess is that it would likely be legal in more places than it was then.

Finally, here’s the point that drew me as a journalist, and I hope it is not entirely lost among the media’s detractors: I realize that few critics of the Trumpista variety are likely to ever read this, but it they did they would see as effective a demonstration of this newspaper’s fairness regardless of whose sacred cows get gored. It’s hard to imagine a Fact Checker verdict more likely to cause distress to the political left, which the press supposedly shills for.

So I hope somebody on the right notices it, and has his or her prejudice lessened at least a bit.

Anyway, as I said, it’s interesting on a number of levels, so I thought I’d share it.

336 days, 355 mass shootings

I got this from The Washington Post‘s Wonkblog. How is a “mass shooting” defined for the purposes of this count?

The San Bernardino shooting is the 355th mass shooting this year, according to a mass shooting tracker maintained by the Guns Are Cool subreddit. The Reddit tracker defines mass shootings as incidents in which four or more people, including the gunman, are killed or injured by gunfire.

The Mass Shooting Tracker is different from other shooting databases in that it uses a broader definition of mass shooting — the old FBI definition focused on four or more people killed as part of a single shooting.

It would be also be the second mass shooting just today — in the early morning hours, one person was killed and three were injured in an incident in Savannah, Georgia.

Speaking after the Colorado Springs shooting last week, President Obama urged Americans to not let this type of violence “become normal.” But the data show that this type of incident already is normal. There have been more mass shootings than calendar days so far this year…

So if only three people are hit, it’s not a mass shooting, by this count.