Wednesday, September 19, 2007

Lower-Income, High-Achieving Students Fall Behind

Lower-Income, High-Achieving Students Fall Behind

It's a popular belief that gifted kids are "the kids with disposable income," as Mara Sapon-Shevin wrote in her book Playing Favorites. Since all gifted kids supposedly have well-off parents, there's no need for schools to concern themselves too much with their welfare. After all, these are the kids best able to fend for themselves.

Now a new report from Civic Enterprises, a think tank, and the Jack Kent Cooke Foundation, called The Achievement Trap, shows not only how wrong that assumption is, but that it's a dangerous assumption as well. Fully 3.4 million K-12 students with family incomes below the national median test in the top quartile for academic achievement. Unfortunately, a higher proportion test in the top quartile in first grade than later on. In other words, as these students progress through school, their achievement suffers.

As I mentioned in the last post, there is much to dislike about No Child Left Behind (though I think there's some evidence that it has created incentives to reverse failure in the worst schools). One thing to dislike is that these millions of high-achieving, low-income students are in fact getting left behind. Even as they fall out of the top quartile, they rarely test as "failing." And so they're ignored both by the standards movement, and by the folks who dislike gifted education because it's only for the kids with "disposable income."

I'm glad this report is out, and I hope it gets some attention, because it shines a spotlight on what many of us who are involved in gifted education know. The leveling impulse in schools hurts all gifted kids, but it does not hurt all gifted kids equally. Well-off parents can afford to move to a different district, put their kids in private schools, hire tutors, pay for summer courses, or give up a parent's income to homeschool. Low-income families are simply stuck with the schools they get. And so, these promising children wind up falling out of the top quartile, and going to college at lower rates than other children whose potential they matched in first grade.

Can anything change it? Under the current educational set-up, I think the best approach is to measure individual children's progress, not just the progress of groups and schools. Education Secretary Margaret Spellings has hinted at shifting NCLB to this concept. What gets measured gets changed, and if schools are held accountable for the low-income children who drift from the 90th percentile in first grade to the 70th percentile in fifth grade, maybe that's a few more children who won't fall through the cracks.


Taia said...

The article measures academic achievement in terms of being in the top quarter academically. Only 25% of students can be in the top quarter. We can't increase the fraction of low-income students in the top quarter without decreasing the fraction of high-income students, the NCLB approach.

I would be more interested in how absolute scores of ALL students can be improved, on a test like the NAEP, ACT or SAT.

I also wonder how correlated parental intelligence is with parental income, and if we will ever know the genetic heritability of intelligence. Unfortunately, nature has not bestowed her gifts equally.

Karen said...

Tracking test scores by group rather then individual makes NCLB worthless for the following reasons.

1. Schools in neighborhoods with well off, well educated parents get credit for the efforts of the parents while schools in areas with poor illiterate parents get penalized.

2. Schools are motivated to screen incoming students.

3. Schools are motivated to force out poor performing students to improve test scores.

4. Schools with high turnovers suffer, especially those near embassies. Last years English Language learners are now back in China and have been replaced by Israelis fresh off the plane who speak no English. But the test doesn’t care; the test expects the group to make progress even though it isn’t the same group.

5. There is a lot of pressure to hold back students so the school will look good on the tests while there is no way to move up because concerns about the test.

Another major problem with NCLB is teaching to the test. This is a problem because.

1. It is possible to teach kids to score well on a test without actually teaching them the material. Consider basic division on a multiple choice test. A number is divisible by 2 if it is even. By 3 if the digits add up to 3, 6, or 9. By 4 there is 50% if it even or teach to divide by 2 twice. By 5 if it ends in 0 or 5. By 6 if it is even and divisible by 3. By 7 if none of the other answers work. By 8, there is a 25 % chance it is divisible by 8 if it is even or divide by 2 three times. By 9 if the digits add up to 9. By 10 if it ends in 0. Now by teaching a child to divide by 2 they can get at least 90 percent on a basic division test without really knowing much division.

2. Long term education goals are forgotten. There is no reason a child can not be introduced to the periodic table in Pre-K. Right after the Alphabet and the numbers. It is simply another basic set that is key to understanding a major area. The alphabet is the basic set for all words and language. The numbers are the basic set for all mathematics. The Period table is the basic set for all science and matter in the universe. It fits perfectly in the pattern and could easily be painlessly introduced with well designed materials. Doing so would allow children to notice and understand chemical formulas on products, the mention of elements on TV or in everyday life and allow them to enter later science classes with a much greater understanding. But because the period table is not part of any test until much later if ever it will not be discuss in early childhood education and may not be discussed in elementary school at all.