By Julian Betts
, adjunct fellow, Public Policy Institute of California
This opinion article appeared in the Riverside Press-Enterprise
on October 14, 2007
With the latest release of numbers from the Academic Performance Index—the major measure of academic performance in public schools—California is awash in test-score data. So with the school accountability system now entering its eighth year, where do we stand? Is the system working?
When California set up its accountability system in 1999—with annual student testing, content standards in major subjects and interventions for underperforming schools—many believed it would improve teaching at all schools and radically narrow achievement gaps between socioeconomic groups.
While California’s schools have clearly boosted student achievement in math and reading, progress has been stubbornly slow and wide gaps remain among students from different backgrounds. In short, it is becoming clear that the state’s accountability system alone has not achieved its lofty goals.
The main difficulty has been finding interventions that truly help struggling students and boost performance in underperforming schools. California’s system includes a program of state subsidies for implementing reforms at the lowest-performing schools. However, an evaluation of the program by the American Institutes for Research suggests that the reforms have produced only “negligible” gains in achievement.
What, then, should the next set of reforms include to significantly affect student achievement? Class-size reduction in the lower grades has probably boosted achievement moderately. My research with colleagues in San Diego suggests that smaller classes somewhat increase reading and math scores, especially among English Learners. However, we found no evidence that class size makes a difference at higher-grade levels.
A second intervention is deceptively simple: Give struggling students additional time to learn. A growing body of evidence from school districts throughout the nation, including Chicago and San Diego, suggests that summer school can help lagging students catch up to their peers. Results from San Diego’s bold but controversial literacy reform earlier this decade also suggest that afterschool-reading programs can be effective.
Similarly, the federal No Child Left Behind law requires districts to offer “supplemental services,” which consists of tutoring, to students at schools deemed as failing. Evaluation of this program shows that tutoring has produced moderate gains.
All of these findings suggest that additional time for learning can truly help students.
A third area to consider is investing more in teachers themselves. Our research finds that teacher qualifications, including credentials indicating subject-matter knowledge, do matter for student learning. But unlike the pattern for class size, teacher qualifications matter more in high school and less in the lower grades. So is a policy of “fully qualified teachers for all state schools” the magic bullet? Probably not. Overall, research suggests that these kinds of standards would only slightly narrow the huge achievement gaps that exist across the state’s schools.
A different approach to improving teachers’ skills could be through professional development. But this is difficult to research effectively because professional development varies wildly and is difficult to characterize. And second, there are simply too few rigorous studies that examine the effects of professional development on student achievement.
The need for better research applies to professional development, but also leads us to a final, broader issue that is intrinsic to improving student achievement: California urgently needs to raise the quality of its evaluation of new educational reforms.
Two steps are essential. First, instead of implementing major new reforms statewide, California should do so on a limited-trial basis. Second, the state should hold a lottery to determine which schools or districts would receive the initial funding. Researchers could then compare the progress of students in schools that were not selected in the lottery with that of those who were. This approach— which is similar to the evaluation of new prescription drugs before they are allowed on the market—would provide compelling scientific evidence on the effects of a given reform.
And while a limited rollout could also be fiscally prudent in the short run, the real money savings would be gained in the long run. If an evaluation revealed that a particular reform did not work, the state would not spend resources to extend that reform statewide.
And best of all, the state’s students would be spared from suffering through the latest unproven educational fad.