The Annual Misuse of Hate Crime Statistics

Mark of New Jersey

Mark is a Founding Editor of The League of Ordinary Gentlemen, the predecessor of Ordinary Times.

Related Post Roulette

21 Responses

  1. greginak says:

    everybody should take at least half a year of stats in HS and read the book How to Lie with Statistics.

    good stuff mark. i’m surprised Think Progress whiffed on this,they are usually good.Report

    • Mark Thompson in reply to greginak says:

      Thanks, Greg! I don’t know what it is about this particular report, but it seems to invite this kind of a misread every single year. The previous example I cited was even worse in some ways, because the headline was disseminated by CNN. I suspect it has something to do with the fact that the report is statutorily mandated, ostensibly for the purposes of keeping track of annual trends, so it’s very easy to fall into the trap of thinking that it’s a meaningful report.

      The positive is that this year, at least, it looks like CNN reported the data appropriately:

  2. Kyle says:

    Sometimes stuff likes this makes me think there are Americans who hate America(ns) and they aren’t just from make believe stories Republicans tell their children to scare them.

    What really bothers me here is the all too common way in which people combine good intentions with an irresponsible lack of thought or competence.

    In this case, the goal is to highlight crimes against gays, women, blacks, etc…, unaffectionately called “hate crimes,” in order to advance awareness and preventative action. Yet, instead of advancing that cause, they’ve given legitimate fodder for critics to harp on – the accuracy of their claims and a tendency to distort data to serve their purpose.

    Which, in turn, leads to an argument between the good people who “get it” and “care” versus the bad people who don’t. When, in fact, the good people are just really bad at crafting their message and rather than own that cast aspersions. Rinse and repeat.Report

  3. Will says:

    This was an incredibly informative post, Mark. Good stuff.Report

  4. greginak says:

    FYI- think progress has posted this update “The FBI adds a disclaimer that it “doesn’t report trends in hate crime stats.” So increases or decreases may be due to changes in the number of actual hate crimes, or changes in “the number of agencies who report to us varies from year to year.”

    Whatever the faults of web news and opinion, many of them are willing to update and admit mistakes far more then the MSMReport

    • Kyle in reply to greginak says:

      I don’t know I find print outlets are good at admitting mistakes, they just tend not to put them front and center in a way that is structurally more likely to happen with a blog.

      Though, it’d be nice if – aware of the irony of not doing this myself – blogs did more ombudsman work. Sully sort of does it with his dissents but otherwise, not so much.Report

  5. Hudson says:

    For a long time government has kept statistics on various types of crimes. Statistics are not the real topic here. What is the real meaning of hate crimes/hate speech? Who wields this term the most often and to what purpose? It seems to me, more and more, “hate” is a nasty epithet hurled by the Left against a broad swath of opponents of its various programs. It is a short step from hate speech to thought crimes. If you are interested, you can read more in my blog entry “Hate Speech” at Link:

  6. zic says:

    So the report is more a measure of how agencies report crimes, and it shows an increase in police-agency reporting hate crimes against gays.

    Though it’s counterintuitive, perhaps it’s good news in the struggle for equal rights.Report

  7. ThatPirateGuy says:

    I think that we can draw a clear conclusion from this data.

    Our data gathering for these statistics is deeply screwed. It would be nice if we could at least get all agencies reporting.Report

    • That’s certainly part of it. But a good chunk of this is also that each state has wildly varying hate crimes laws. So even if every agency had a formal mechanism for reporting hate crimes, you’d still get impossible-to-interpret data simply because of the different definitions of what constitutes a hate crime, to say nothing of variations in local law enforcement priorities.Report

      • zic in reply to Mark Thompson says:

        I’d guess there’s general under-reporting to the authorities, as well. I’m sure many crimes committed simply because a victim is gay, like many sexual assaults against women, aren’t reported because of the stigma involved; particularly for people who haven’t come out. I’m sure other crimes are reported as assault, vandalism, etc. though there was a ‘hate’ aspect to the crime.Report

      • ThatPirateGuy in reply to Mark Thompson says:

        One problem at a time!Report

  8. Jaybird says:

    If I may ask a wicked question, did the run of the mill, just a garden variety assult and battery, this was only a random guy who happened to be Muslim who shot people while yelling “Allahu Akbar” kinda crimes go up or down?

    I ask because, if *I* were a reporter whose favorite headline was “WE ARE ALL GOING TO DIE” and the crimes that consisted of little more than random people killing other random people for reasons we can’t really determine went *DOWN*, I might be tempted to read these numbers and say “Hey, I might be able to use my favorite headline after all…”Report

  9. BobN says:

    Discussing the uselessness of the data as a measure of trends seems rather, well, distracting. I, for one, would like to know how the seven bias-based killings last year break down by bias. Oddly, the stats don’t seem to offer those details…Report

    • Mark Thompson in reply to BobN says:

      It’s not distracting if the primary way in which the data is being used is as a measurement of trends, which is in fact how the data is being touted. It is thus important to point out that the data does not, in fact, show any trends.

      As for how various crimes break down by bias, table 4 of the report provides that information. It shows 1 race-based killing, 3 anti-gay killings, 2 anti-bisexual killings, and 1 anti-Hispanic killing. However, it’s difficult to read anything into those numbers given not only the very small sample size but also, again, the fact that state hate crime laws vary so wildly.Report

  10. BobN says:

    Ooops, take that back. It’s in Table 4.