What Students Do in the Summer

Abstract

WAY BACK in September 1986, I summarized in this column a study that had found results that contradicted a known fact: that there is a phenomenon called loss. Actually, researcher, Maxine Winter of York University in Ontario, stated that literature on summer loss was quite old and contradictory. Winter conducted a study and reported gains for first-graders in reading, knowledge, and mathematical concepts. Third- and fifth-graders showed a decline in math computation. For third-graders decline was statistically significant. In years since, of what happens to students over summer has grown more complex. In summer 2001 issue of Educational Evaluation and Policy Analysis, Karl Alexander, Doris Entwisle, and Linda Olson of Johns Hopkins University provide what they call A Seasonal Perspective on achievement and inequality. They open with a headline from Baltimore Sun: Pupils Lose Ground in City Schools: Longer They Stay in System, More They Fall Behind. Many readers will have seen similar headlines about big- city in their own states. headline and story imply that system is failing. But, researchers wonder, is it that are failing? They point out that students of low socioeconomic status (SES) score well below their affluent peers in first grade. debilitating of their environment are readily apparent even in these early grades. Why do we assume situation would be different later on? The drag of poverty, family stress, and community decay doesn't turn off when children reach age 6 and school's influence begins to weigh in. Given this drag, the achievement gap across social lines would be expected to widen over time for reasons having nothing at all to do with (emphasis in original). They point out that reporting test scores annually, as is customary, confounds effects of home, school, and community. Do exacerbate unequal performance across social lines or do they mitigate such inequality? This is question they want to answer. To seek an answer, researchers use data from Beginning School Study, conducted in Baltimore. In this study students were tested in fall and again in spring. It is thus relatively simple to obtain an indication of what happened during summer by subtracting, say, scores from spring third-grade testing from scores obtained in fall of fourth grade. study placed students into low-, middle-, and high-SES based on interviews with their parents. low-SES students started behind their peers in first grade -- about 0.7 of a standard deviation in both reading and math. By spring of fifth grade, they were more than 0.9 of a standard deviation back. This finding was typical. When researchers looked at fall-to-spring and spring-to-fall changes, though, pattern that emerged was quite different. In fall-to-spring comparisons -- that is, comparisons that measured gains during school year -- all three groups of students gained about same amount. And gains in first two years were substantially larger than in last three. However, in spring-to-fall comparisons, low-SES and middle-SES students lost ground in both reading and math over summer between grade 1 and grade 2 and between grade 2 and grade 3, while high-SES students gained on both. After grade 3, almost all summer loss turned up in low-SES group, and when low- SES group did show a summer gain, it was smaller than gains for middle- and high-SES groups. Thus achievement gap widened during time that students were not in school. Moreover, gains that do appear over summer are quite small in comparison to those achieved during school year, a finding that will no doubt come as a relief, but not a surprise, to teachers. Referring to erroneous but common interpretation of Coleman Report that schools don't matter or to newspaper headlines suggesting that cause failure, researchers have this to say: Schools do matter, and they most when support for academic learning outside school is weak. …