Sometimes I wonder whether the Associated Press reads the Associated Press. (I don’t know why they would; if I weren’t professionally obliged to do so, I would not.)
So, here’s a typical AP story from today:
More anti-gay, religious-motivated crimes reported
That’s the AP’s headline. Salon’s is more typical of the way these stories get played:
Anti-gay, religious-motivated crimes up
FBI data shows 11 percent increase in crimes based on sexual orientation
Salon’s headline is slightly misleading, to my ear. It reads like, “People religiously motivated to attack gays committing more crimes,” but that isn’t what the article says. And the AP’s headline, while technically correct, is misleading, too, which you will only discover if you are one of those rare people who reads down into the story past the first dishwater-dull paragraph. In fact, there is no year-on-year increase in any meaningful sense, because the FBI does not make year-by-year comparisons. There’s a good reason for this: The number of local law-enforcement agencies reporting these things to the FBI varies wildly from year to year. This year, more agencies reported than last year, so that the total number of crimes is up ought not be surprising. How do I happen to have this arcane data? Unlike the headline writers, I read paragraphs six through nine:
The FBI does not compare year-to-year trends in hate crimes, saying the number of agencies reporting changes too much. In fact, the bureau cautioned that the increase reported Monday might well be due to more agencies tracking such incidents.
Brian Levin, director for the Center for the Study of Hate and Extremism at California State University-San Bernardino, warned that the national numbers may be misleading because some states — like California, New Jersey, and Ohio — are good at reporting hate crimes while others — Georgia, Hawaii, Mississippi and Pennsylvania — are not.
“The quality of the data is so variable and in some instances so bad that it makes trend analysis extraordinarily difficult, if not impossible,” said Levin. “Generally, states that have effective data collection also have effective training and procedures to address these crimes.”
In 2008, 2,145 different agencies reported hate crimes incidents, while the year before 2,025 agencies did this reporting.
So, color me shocked: More agencies report their hate crimes, and the total number of hate crimes reported goes up. Yeah, thanks for the Muppet Newsflash, geniuses.
The real headline should be this: “FBI, in spite of its gigantic budget and zombie army of coddled bureaucrats, has no reliable data on hate-crime trends.”
But none of this will stop this story from being cited by every diversity-pimping NGO and taxpayer-fleecing government agency operating as a full-employment program for sociology majors as proof of a ”disturbing increase in hate crimes” as they seek to swell their budgets. Examples here and here. And here. How about here? You bet you’ll find it here. Here? Inevitably.
That last example, from the Human Rights Campaign, contains an interesting sub-head:
Local and state law enforcement agencies urged to report statistics to FBI.
Why? They are not “urging” the FBI to actually do what is necessary to track the trend, but only encouraging law-enforcement agencies to swell up the entirely meaningless raw numbers. Those big (fictional) spikes in hate crimes look really dramatic in fund-raising letters. The ho-hum reality — that tracking broad trends in hate crimes is very difficult to do and produces few satisfactory answers — will not open very many checkbooks.