High-stakes testing, homework, and gaming the system.

Readers of the Los Angeles Times were recently shocked by published findings of the Harvard University Civil Rights Project concerning California school dropout rates. According to the study (presented March 24, 2005, at the Harvard conference, "Dropouts in California: Confronting the Graduation Rate Crisis"), only 60 percent of Latino students and 56.6 percent of African American students graduate at all from California schools, and only 47 percent of African American and 39 percent of Latino students graduate on time. What is worse still, this record has been largely hidden.

Confronted with the pressures of high-stakes "Leave No Child Behind" testing, some administrators have, in fact, left many children behind. They've cooked the books: they've tolerated or even encouraged high-risk students to drop out while failing to keep tabs on the dropouts. As a result, schools' statistical performances artificially improved as most of their minority children were being deserted.

Should we be surprised? After all, such results are a likely consequence of the confluence of two destructive trends in U.S. education: high stakes testing and homework intensification. These trends are premised on the notion that U.S. public schools are failing because of the laxity of teachers and laziness of students. Youths must work harder; be monitored by objective, quantifiable standards; and be sanctioned harshly when they fail to meet those standards. For teachers, the proposals favor merit pay based on student performance as a way to increase the productivity of schools.

This sounds like a no-nonsense approach to education, but its results have been less straightforward. Administrators' careers now depend on the results of one standardized test--a test that few educators had a voice in fashioning. As a result, administrators have found ways to reshape the playing field. Recent research by Sharon Nichols and David Berliner for the Educational Policy Studies Laboratory at Arizona State University (edpolicylab.org) suggests that scandals like California's dropout rates are an example of Campbell's Law: "The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor." Nichols and Berliner find many instances across the country where, in response to these pressures, administrators have encouraged low-performing students to drop out and then failed to report these dropouts to state officials.

Educators and political leaders should also examine the role that homework plays in occasioning dropouts among the most vulnerable segments of the population. Parental involvement in schooling is widely regarded as a key to academic success. And nowhere is that involvement more important than in the completion of homework, so the argument goes. Homework is regarded as an answer to what ails poor and minority communities. But homework's contribution even to solidly middle class children is exaggerated at best. Homework is often inattentive to the learning styles and limits of particular children, can merely exhaust already tired children, and too often is done or aided by others--in many instances without the teacher's knowledge. Even many staunch defenders of homework now admit that the case for it in the lower grades is weak and defend it only for the discipline it purportedly provides.

But homework provides no more discipline than does concentrating during the seven or eight hours of the school day or working on a hobby at home. When schools demand homework as a condition of advancement, many middle class children complete it--often with parental help--even if homework does little to enhance their long-term academic strengths. This practice isn't as easy for poor children, however, as we will show--which means that homework often punishes poor children merely for being poor.

In studies we conducted among high school dropouts in rural Maine, we asked each student, "Was there a point in your school career at which you were sure you weren't going to make it?" We had anticipated stories about harsh teachers or intimidating peers but, to our surprise, the former students all had tales of homework. They described difficult personal circumstances--from lack of quiet space to siblings in need of care to parents who couldn't provide assistance.

By the same token, how can California educators expect students to learn from homework or in school when they face situations like that of Gabriela Perez (detailed in the March 25, 2005, LA Times article)? Perez described a lack of textbooks and science classes, and counselors who placed her in work programs rather than academic classes designed to prepare her for college. She explained how her immigrant parents couldn't help her with her homework. She did want schools to expect more of students--but also to provide them with more support. Her story appears to be a common one for Los Angeles students.

Unfortunately, whether by conscious design or benign neglect, decisions by students like Perez to drop out seem to bother too few administrators. If homework drives out marginal students, a school's rating goes up as long as dropout numbers are fudged. And when minority students drop out because of homework, it confirms and reinforces some of our worst cultural stereotypes: that students fail because they are lazy and that minority students and their families don't care about education.

Minority children can do better in school. But the answer lies in remedies that a focus on lazy children and cultural stereotypes hides. According to a recent RAND Corporation study, smaller class size, especially in the elementary grades, better trained teachers, and adequate pre-school programs have all contributed more to educational excellence. At the high school level, students can benefit from independent work, but such work is most effective when all have adequate facilities and trained adults who can provide assistance and support.

Just as basically, many poor minority children also do less well in school than their middle class peers not because they don't do homework but because of the lack of cultural opportunities. Johns Hopkins University researchers point out in a recent study of children's academic progress that the more successful children "more often went to city and state parks, fairs, or carnivals and took day or overnight trips. They also took swimming, dance, and music lessons; visited local parks, museums, science centers and zoos; and more often went to the library in summer."

Students and teachers should both work hard and be held accountable. All children, however, deserve and benefit from the proper educational and cultural support. Working together with parents and other community members, teachers and students can craft reasonable workloads and standards that most understand and accept. When we treat teachers and students as perennially lazy, and whole groups as uninterested in education and in need of a short, continuous leash, should we be surprised if the most disadvantaged students give up on a system stacked against them, the most creative teachers flee that system, and others merely game it?

Etta Kralovec directs the master's program in education and teacher credentialing at Pepperdine University in Los Angeles, California, and is co-author of The End of Homework and author of Schools that Do Too Much. John Buell, a columnist for the Bangor Daily News, is the author of Closing the Book on Homework.