Part 1—What does that mean: Gail Collins [HEART] disaggregation!
Collins makes her feelings clear in her unfortunate new book, As Texas Goes. We reported on her book all last week. For a quick review of those posts, see THE DAILY HOWLER, 9/24/12.
That said, what is disaggregation? And why doesn’t Collins employ the practice she loves?
Those are excellent questions! In truth, you can’t begin to analyze public school test scores unless you understand “disaggregation,” the practice Collins admires.
How much does Collins [HEART] disaggregation? Let’s take a look at the record!
In the part of her book which deals with the Texas public schools and the No Child Left Behind program, Collins heaps praise on former Governor Bush for supporting “disaggregation.”
Throughout this chapter, Collins mocks Bush in ways designed to please liberal readers. But in the following passage, she explains what “disaggregation” is, and she briefly heaps big praise on the former governor, with a few snide remarks scattered in.
What the Sam Hill is disaggregation? Let’s let Collins explain it! In this passage, we’re back in the 1990s, when Bush was governor of Texas:
COLLINS (page 83): To explain how the legislation we now know as No Child Left Behind came into being, we have to begin with some serious praise for George W. Bush.Did middle-class districts hate disaggregation? We have no idea. But disaggregation, a very big word, stands for a simple idea:
While some other states had gone into testing in a big way too, there was one part of the Texas school reforms that was unusual. It’s known as disaggregation. Basically, it means that a school’s score on the test is based not only on how well the students do overall, but also on how much the poor, black and Hispanic kids improve. “The argument was—and I think it was a compelling argument—that in the past some schools had let some sub-populations drop through the cracks and that wasn’t acceptable,” said [Texas pol] Bill Ratliff.
Disaggregation put tremendous pressure on schools to focus on bringing up their poor and minority students. Districts with large middle-class white populations hated it because their schools could wind up with a low rating even if the majority of their kids were doing well. It was a powerful club against all the subtle and not-so-subtle forces that have created unequal educational opportunity in the twenty-first century. And George Bush adored disaggregation. He loved saying the word. Disaggregation was what he meant when he talked about “the soft bigotry of low expectations” for poor and minority kids.
You can’t really judge a school or school district by its overall score on some test. You have to break down a district’s scores to see how well its low-income students did. You see how its black kids did. You have to review the performance of its Hispanic students.
This isn’t a complex idea. That’s why it’s so amazing to see Collins fail to employ this simple technique all through her deeply unfortunate book—to see her mock the Texas schools on the basis of overall scores, without checking to see how the scores look after disaggregation.
Texas, you see, has a very large percentage of low-income and/or minority kids. In every one of the fifty states, low-income kids score substantially lower than their more advantaged peers. Ditto for black kids and Hispanic kids.
We think you know the background:
Due to the ravages of our history, black kids have always scored lower, as a group, than their white counterparts. In the past twenty or thirty years, black kids and Hispanic kids have been scoring much better on the National Assessment of Educational Progress (NAEP), the widely-lauded “gold standard” of American educational testing. One remarkable example: As of 2009, black fourth-graders were scoring higher on the NAEP math test than their white counterparts scored in 1992!
That represents astounding progress. But alas! As of 2009, white kids were scoring substantially higher in fourth-grade math too! For that reason, “achievement gaps” persist between the three major demographic groups, although the gaps have gotten smaller. And in all fifty states, low-income kids score substantially lower than their more advantaged peers.
Texas schools teach a large proportion of black and Hispanic kids. They also teach a large proportion of kids from low-income families. Kids are kids, but on these measures, the Texas student population differs markedly from those in other states.
The difference can be very large. This was the demographic breakdown for two famous states on one of last year’s NAEP tests:
Percentage of students tested by race/ethnicity and income, fourth-grade reading, 2011 NAEPKids are kids. But on last year’s fourth-grade reading test, Texas had almost twice as many kids from low-income families. Massachusetts had more than twice as many white kids—and in every one of the fifty states, white kids still score substantially higher (on average) than black and Hispanic kids.
Massachusetts: 68 percent white, 9 percent black, 14 percent Hispanic
Texas: 31 percent white, 14 percent black, 51 percent Hispanic
Massachusetts: 33 percent low-income
Texas: 63 percent low-income
Kids are kids, but on these measures, the Texas student population differs from those in other states. If you’re trying to determine how well the Texas schools are performing, you can’t simply look at overall scores, for the bone-simple reason Collins explains in her book.
You have to “disaggregate” the data. You have to see how well Texas and other states do with roughly comparable groups.
Collins praises this practice in her book, then utterly fails to employ it. In a similar way, Molly Ivins mocked the Texas schools in a July 2000 column, failing to note that Texas kids were outscoring their demographic counterparts all through the fifty states.
This is how Texas students wer scoring as Ivins mocked the state's schools:
Texas students, fourth-grade math, 2000 NAEPTexas kids were outscoring their peers all over the nation. But Ivins, in a grossly misleading column, said the state’s schools had “rocketed from abysmal to only slightly below average.”
White kids: Second in the nation (of 40 participating states)
Black kids: First in the nation (of 32 states)
Hispanic kids: Second in the nation (of 21 states)
Texas students, fourth-grade math, 1996 NAEP
White kids: First in the nation (of 43 participating states)
Black kids: First in the nation (of 35 states)
Hispanic kids: Second in the nation (of 25 states)
“Our scores are still slightly below the national average,” she wrote—“27th of the 44 states that use the national tests.” That statement can be defended as technically accurate, but it was also grossly misleading. Details tomorrow.
Ivins failed to disaggregate! Twelve years later, in her new book, Collins quotes a wisecrack from Ivins’ column—and she fails to disaggregate too, even after making a point of praising the bone-simple practice!
Collins knows what disaggregation is. But twelve years after Ivins’ error, she’s still too lazy—or too partisan—to employ the practice she loves.
In a rational world, Collins would be disowned by journalists, savaged by liberals.
You don’t live in that world.
Tomorrow: Molly Ivins, twelve years back