I just returned from a week in Wisconsin during which I tried (not entirely successfully) to ignore the news. A few notes on what I missed:
• I’ve previously discussed the research of Adam Lankford, who claims that the U.S. is home to nearly a third of the entire world’s mass shootings and blames our high level of gun ownership, and I’ve noted his refusal “to share information about his analysis with media outlets likely to criticize him.” He also refuses to share data with rival researchers, and one of them, John Lott, has thrown down the gauntlet.
In a new paper, Lott assembled his own global database of mass shootings (following Lankford’s own definition) and found that the U.S. accounts for just 3 percent of them. This included hiring researchers fluent in Chinese, French, Polish, Russian, and Spanish to uncover incidents that were not reported in English-language sources. Most important, however, is that you don’t even need to ask Lott to share his data: His full lists of U.S. and international shootings are right on his website.
This research is very hard to do, given language barriers and the difficulty of defining “mass shooting.” But the way to arrive at a solid conclusion is for researchers to be transparent about exactly which incidents they’re counting, and it’s great to have a totally public source available as a starting point for future work.
• In the middle of August I noted a paper on “rapid onset gender dysphoria” — cases in which teens and young adults abruptly decide they’re transgender, often amid social pressures to identify this way. One of my observations was that
because the study recruited subjects from “three websites where parents had reported rapid onsets of gender dysphoria” and collected information only from parents, it’s hard to say how common this is, how representative the families involved were, and how trustworthy the parents’ reports are (though it’s worth noting that the overwhelming majority of the parents support both gay marriage and transgender rights).
In other words, the limitations of the study were crystal clear to anyone who read it; even the abstract explained where the data came from. And studies relying on less-than-perfect data are quite common in social science, especially when it comes to areas where good numbers are hard to come by — see, for example, the various studies of gay parenting that relied on self-selected gay parents to report their own kids’ behavioral issues.
And yet this particular study caused an enormous backlash. Brown University, where the study’s author is based, pulled a news story about the research off its website, replacing it with an insufferable open letter from the dean of the university’s public-health school. PLOS One, the journal that published it, is “following up” on “reader concerns.”
The study was an interesting profile of parents who said their kids had identified as transgender with little warning, and it provided a basis for further research. It didn’t pretend to do anything more, and if this study is not up to snuff, then a lot of other social-science work needs to be thrown out with it.
• On a lighter note, I’m on this week’s Political Beats podcast talking about Guns N’ Roses, my favorite band since childhood. Check it out here. I am told this is the most profane group to ever have been featured on the show.