40% followup.

Previously on WCD, we talked a little about the “40% of all gun sales are done without background checks” figure.

Ted Cruz has pointed out some of the same holes we discussed in those figures, and “Politifact Texas” decided they’d fact check him.

What did they find?

The 40 percent figure originated in a 1997 National Institute of Justice study by researchers Philip Cook of Duke University and Jens Ludwig of the University of Chicago, who examined data from a 1994 telephone survey about gun ownership taken months after the Brady Handgun Violence Prevention Act of 1993 took effect, mandating background checks of individuals buying guns at gun shops.

The law doesn’t apply to private sellers at gun shows, flea markets or people who post firearms for sale on the Internet.

What Politifact means by “people who post firearms for sale on the Internet” is unclear. If they mean people who sell to other people locally, that’s quasi-true. If a firearm – any firearm, rifle, pistol, or shotgun – is being shipped rather than sold face to face, the gun has to be shipped from a licensed dealer to a licensed dealer, and the receiving dealer has to do a background check before turning over the firearm.

Of the 2,568 households surveyed, only 251 people answered the question about the origin of their gun.
In those answers, Cook and Ludwig found that 35.7 percent of respondents reported obtaining their gun from somewhere other than a licensed dealer. (That has been rounded up to 40 percent.)

Rounding does not work that way!

When the Washington Post’s fact-checking project, the Fact Checker, asked Ludwig to revisit the data, the newspaper said, results suggested that purchases without background checks amounted to 14 to 22 percent, not 40 percent.

By the way, Politifact also quotes the Post as pointing out the small sample size means there’s a large margin of error: “plus or minus six percentage points”. So it could be anywhere from 8% to 28%. And by the way…

“…the survey was taken in late 1994, eight months after the Brady law went into effect, and the questions were asked about gun purchases in the previous two years. So some of the answers concerned gun purchases that took place in a pre-Brady environment.”

And more:

For his part, Cook, the other study author, told us he has no idea whether the 40 percent figure remains reliable. “This survey was done almost 20 years ago. … It’s clear there are a lot of transactions that are not through dealers,” Cook said. “How many, we’re not really clear on it. … We would say it’s a very old number.”
Other scholars had similar views.
“I don’t see how anyone could know that number,” said James Jacobs, Center for Research in Crime and Justice at New York University School of Law.

More:

The surveys took place after the law took effect, Cook said, but it’s correct that the vast majority of gun acquisitions identified by respondents took place before then, given that respondents were asked to focus on guns acquired within two years, meaning 1992 and 1993.

And there were 18 states that required background checks prior to Brady. But:

Even so, Cook told us, he and Ludwig did not use information on individual states in doing their analysis. He emailed that “the survey provides an accurate representation of the nation as a whole but not of individual states.” Also, he stressed, the survey had no questions about whether there had been a background check and respondents also weren’t asked the states where they had acquired their guns. He agreed there is “no way to reach any conclusions about background checks for any of the transactions that are reported in the survey.”

So, summing up: the figure was never 40%, but 36% (see, that’s how rounding works), and is probably closer to 20% before the margin of error kicks in, which could make the actual figure anywhere from 8% to 22%. And the survey was conducted in a pre-Brady environment. And the researchers themselves say (again I quote) there is “no way to reach any conclusions about background checks for any of the transactions that are reported in the survey”.

PolitiFact’s rating: “Half True”. I see nothing in their summary that refutes anything Ted Cruz said, or that justifies the “half true” rating.

One Response to “40% followup.”

  1. […] is all over Polifact Texas “checking” Ted Cruz’ statements about untracked gun […]