Fun with Spin and Stats
Stats and Spin: Thursday I was stuck reading a paper copy of the NYT and learned that if you come to the paper with a pile of fresh statistics they will obligingly give them the spin you want (or, if you don’t have an obvious agenda, the spin various hovering causists want). [Need three ex.-ed . Got ’em!]
Example 1: “Segregation Prominent in Schools, Study Finds,” by Motoko Rich.
Across the country, 43 percent of Latinos and 38 percent of blacks attend schools where fewer than 10 percent of their classmates are white, according to the report, released on Wednesday by the Civil Rights Project at the University of California, Los Angeles.
That 38% figure–the Civil Rights Project wants us to think it’s high, but doesn’t it seem encouragingly low? I would have thought it would be much higher. If you check the text of the report, you learn that it has risen slightly since 1992 (when it was about 33%). But, still, some perspective please. In 1969 it was 64%. And the percentage of blacks in schools with fewer than 1% of their classmates are white has continued to fall since ’92 (from 18% to 15.5%).
Example 2: “Fewer Children Are Found Exposed to Violent Crime” by Erica Goode, has some “breathtaking” good news. It is buried deep in the paper below the fold.
From 1993 to 2010, the number of children living in households where another member had been a victim of a nonfatal violent crime decreased by 68 percent, to 2.8 million from 8.7 million, according to the report, released on Wednesday by the Bureau of Justice Statistics. The crimes included domestic violence, rape, sexual assault and robbery.
The “sharpest drop” was from 1993 to 2001, though after that the decline continued “at a slower rate.” The decline “also comes in tandem with substantial declines in other indicators of childhoood distress,” such as suicide, bullying, sexual abuse, risky sexual behavior, teen pregnancy and feelings of danger at school.
Buy why the decrease? A professor of criminology cites “increasing effectiveness of services available to victims.” The director of a research center points to the ubiquity of cellphones (to summon help) and anti-depression medications. The economic boom of the late 90s isn’t even mentioned as a possible cause–perhaps because it can’t explain why the improvement continued after the bust of 2000. But it’s not hard to come up with at least three unmentioned theories that don’t have that problem: 1) Lead removal. The argument that removing lead from the environment lowered pathological behavior is controversial, but it roughly fits this timeline, no? 2) Social networking. Maybe all the new Internet connectedness is more good than bad, giving people who would be swamped parents or social misfits in earlier eras new communities and relationships; 3) [yes] Welfare reform. When the 1996 reform law passed, opponents on the left predicted a rise in domestic violence. The reverse seems to have happened. Why? Could it be because unprecedented numbers of women from the most at-risk families–poor single-parents–went into the labor force, and work turns out to be a centering, de-isolating, generally healthy experience? (That’s a suggestion backed up in Jason DeParle’s American Dream.)
These all could be BS explanations. But they seem at least as plausible as “increasing effectiveness of services.”
Bonus example: “Income Data Shows Widening Gap Between New York City’s Richest and Poorest,” by Sam Roberts.
Among poor New Yorkers 16 and older, a third had worked full or part time within the preceding year.
“The statistics demonstrate quite clearly that our most vulnerable neighbors are far from a recovery,” said Jilly Stephens, executive director of City Harvest, which helps get emergency food to hungry New Yorkers.
Hmm. Two-thirds of poor people over 15 in New York did no work at all in the preceding year? That strongly suggests the problem of poverty in the city isn’t low wage levels. It’s non-work. The non-work could be because in the current recession jobs are scarce. Or it could be because of … other factors, like a culture of non-work, or single-parents too busy parenting to get work, an un-reformed welfare system, or lots of seniors and disabled people–or even idle people with means (e.g. assets) who find they can get by without work. We don’t know. But only City Harvest’s take makes the paper.
These were just in one day.