Includes updates and/or revisions.
It鈥檚 been a truism for decades that students鈥� learning slips during the summer, and that low-income children fall farther behind than their classmates, but no one had connected the longitudinal data dots to show just what the cumulative consequences of the summer slide might be. Until now.
A by sociology professor Karl L. Alexander and colleagues at Johns Hopkins University in Baltimore concludes that two-thirds of the reading achievement gap between 9th graders of low and high socioeconomic standing in Baltimore public schools can be traced to what they learned鈥攐r failed to learn鈥攐ver their childhood summers.
The study, which tracked data from about 325 Baltimore students from 1st grade to age 22, points out that various characteristics that depend heavily on reading ability鈥攕uch as students鈥� curriculum track in high school, their risk of dropping out, and their probability of pursuing higher education and landing higher-paying jobs鈥攁ll diverge widely according to socioeconomic levels.
鈥淚 call this the Harry Potter divide,鈥� said Alan B. Krueger, a professor of economics and public policy at Princeton University, referring to a 2000 poll by the Princeton, N.J.-based Gallup Organization that asked adults if any of their children were reading the wildly popular series of eponymous books. The poll results showed a wide gap in the responses, based on income.
鈥淐hildren from low [socioeconomic-status] backgrounds don鈥檛 get that reading enrichment,鈥� said Mr. Krueger, who was chief economist of the U.S. Department of Labor in the Clinton administration.
Pace Parallel During Year
The study, which appeared in , makes use of data from reading tests that were administered to the same students twice yearly, enabling researchers to isolate reading comprehension gains made during the school year with those made鈥攐r lost鈥攄uring the summer.
Although the limited national data available on the subject had suggested that the gap between rich and poor would be wide, Mr. Alexander said the numbers on summer from his Baltimore study took him aback.
鈥淲hat surprised me was the size of the summer learning difference,鈥� he said.
By the end of 5th grade, the differential in cumulative scores reflecting what students of high and low socioeconomic classes learned outside of school in the summer was stark.
The summer learning among students in relatively well-educated, economically secure homes had effectively added a total of about 47 points to their test scores by that point in their school careers. Students in relatively low-income, poorly educated families had been reduced by about 2 points over that period.
By contrast, in data covering five winters, when test scores reflect mostly classroom learning, the socioeconomically disadvantaged students kept pace with their more-advantaged classmates.
鈥淪chools are in fact compensating for a shortfall of quality learning experiences outside of school,鈥� Mr. Alexander said. 鈥淚 don鈥檛 fault parents鈥攑arents by and large are the best advocates for their children鈥攂ut the reality is that many parents lack the effective tools for helping.鈥�
Daria L. Hall, the assistant director for K-12 policy development for the Washington-based Education Trust, a nonprofit group that promotes high academic standards for disadvantaged children, worries that the findings will take policymakers鈥� focus off the need to close a different kind of gap.
鈥淲e can鈥檛 allow the problems of the out-of school inequities to overshadow the problems of the in-school inequities,鈥� she said. 鈥淗owever way you look at it, low-income kids and kids of color get less than their fair share of quality teaching, curriculum, and resources.鈥�
Mr. Alexander鈥檚 research has also attracted interest outside of academia. Democratic presidential candidate and U.S. Sen. Barack Obama of Illinois is co-sponsoring the , a bill that cites Mr. Alexander鈥檚 research.
The legislation would authorize $100 million to be divided among five states selected by the U.S. secretary of education for summer programs that combine fun and academics for children who are eligible for the federal free-lunch program. States would have to match the federal contribution of $1,600 per child per summer.
鈥淭hat would be wonderful if the states would actually sponsor high-quality programs,鈥� said Meredith Phillips, a professor of public policy and sociology at the University of California, Los Angeles.
About the study itself, Ms. Phillips said the methodology is sound and the data depth is enviable, even if the sample size is small and all drawn from one place. 鈥淭his is the only data set available to study this question鈥攚e can鈥檛 do any better than this,鈥� she said. 鈥淭he one limitation is that we don鈥檛 know how generalizable the results are from kids in Baltimore to kids nationally.鈥�
Mr. Alexander acknowledged such limitations but said he was sure 鈥渢hat you鈥檇 see much the same results in high-poverty school systems across the country.鈥�
Asked what would ameliorate the problem his study highlights, Mr. Alexander suggested two words: more school.
鈥淢ost advanced industrial countries have more schooling than we do鈥�230 to 240 days a year, some of them,鈥� he noted. 鈥淭he key, though, is that whatever we do, it needs to be done well.鈥�
