Why Achievement Gap Mania Isn't Cost-Free
As I noted yesterday, my National Affairs essay "Our Achievement Gap Mania" has stirred some conversation. Let's take a moment to address those who've asked, "Rick, why are you trying to stir up trouble? There are no losers here!"
Proponents of the gap-closing gospel cheerfully assure us that everybody wins. Education Trust vice president Amy Wilkins has rejected as a "false choice" the notion "that we have to make a choice as a country between equity and excellence. Our policies need to marry both." That's a swell aspiration. Unfortunately, I think the evidence suggests that focusing our attention and finite resources on some children will frequently come at the expense of others. Now, it may very well be that we should choose to focus them on gap-closing. At the least, though, we owe it to our children and ourselves to be more forthright and more conflicted by the fact that we're privileging the needs of some children over those of others.
And this matters, a lot. For instance, it's hardly the case that the U.S. can afford to be cavalier about the performance of our more advanced students. Stanford University's Eric Hanushek, Harvard University's Paul Peterson, and the University of Munich's Ludger Woessmann reported earlier this year that the share of U.S. students accomplished in math trails those of most other industrialized nations. Thirty of the 56 nations participating in the Program for International Student Assessment math test had a larger percentage of students scoring at the international equivalent of the advanced level on NAEP than we did. Indeed, just 6 percent of American eighth graders scored "advanced" on the 2007 Trends in International Math and Science Study. In Taiwan, Hong Kong, South Korea, and Finland, the proportion of students achieving at the same level was at least three times as large.
A universal and exclusive focus on low-achieving kids ignores the fact that different education strategies work best for different kinds of students. Before they ever enter the classroom, many children from low-income and minority households are at a distinct educational disadvantage. Research demonstrates that children from more educated families tend to start school with much larger vocabularies, more exposure to the written word, more time having been read to, and more of the habits that make for a responsible, successful student.
Kindergarteners from low-income households typically have a vocabulary of about 5,000 words, compared to the typical 20,000-word vocabulary of their more advantaged peers. The disparity results, in part, from the fact that many low-income children don't attend pre-school; low-income parents speak to their children about one-third as much as parents who are professionals; low-income parents read to their children much less than other parents; and low-income children watch much more television than do their peers.
Let's take individual needs and differentiated instruction seriously. Some important differences likely overlay the "achievement gap" divides. From the very beginning, for instance, disadvantaged and advantaged children may have different educational needs and stand to benefit from different kinds of instruction. The kinds of teaching and support that can help disadvantaged students acquire the skills and knowledge that they did not receive at home are often superfluous or inappropriate for children who are ready to move on. In this way, gap-closing can shift from a strategy that lifts up the least proficient students into one that slows up the most proficient.
And children who are ready for new intellectual challenges may pay a price when they sit in classrooms focused on their less proficient peers. In 2008, Brookings Institution scholar Tom Loveless reported that while the nation's lowest-achieving students made significant gains in fourth grade reading and math scores and eighth grade math from 2000 to 2007, top students made anemic gains. He concluded, "It would be a mistake to allow the narrowing of test score gaps, although an important accomplishment, to overshadow the languid performance trends of high-achieving students."
Loveless's findings echo a massive 1996 RAND Corporation meta-analysis. RAND researchers have previously reported that when low-achieving students were placed in mixed ability classrooms, they did about five percentage points better. High-achieving students, however, fared six percentage points worse in such classes--and middle-achieving students fared two percentage points worse than they did when placed in "tracked" classes. Weighing these effects out, the authors concluded that switching to mixed-ability classes in math would reduce aggregate achievement by 2 percent. That may suggest that promoting mixed-ability classes is a sensible and just course of action, but it's not cost-free.
There is, of course, the occasional extraordinary teacher who can make heterogeneous classes work for all students. But such teachers are the exception, not the rule. Value-added testing guru Bill Sanders has reported, based on Tennessee achievement data, that high-scoring students made adequate gains only with the top 20 percent of teachers. Students at lower achievement levels, however, made progress with all but the least effective teachers. In other words, Sanders's research suggests that teacher quality may matter more for high-performing students than for their peers.
As with so much of the "achievement gap" agenda, mixed-ability instruction is not a bad idea per se. But it does impose costs. The gap-closing gospel holds that it is improper or out-of-bounds to discuss such things. That's bad for kids, bad for school improvement efforts and, as I'll talk about next week, likely to undermine the kind of middle-class and suburban parental and political support needed to sustain improvement efforts.