Reporter Volume 26, No.24 April 13, 1995 By PATRICIA DONOVAN News Bureau Staff A new study by UB education researchers casts a new light on the performance of 100 of Western New York's 102 school districts (data was insufficient for two districts). It supports past studies showing that student academic achievement depends on many variables independent of the schools themselves, including a district's per-capita income and parents' educational levels. The study finds that in this region, some districts, even if unable to produce average achievement levels equal to those of upper-middle-class school districts, are far more effective than many wealthy districts in teaching the children they get. The report, "School District Effects and Efficiency," was produced by Austin D. Swanson, professor and chair of the Department of Educational Organization, Administration and Policy in the UB Graduate School of Education, and Frank Engert, a doctoral student who will receive his degree in May. The purpose of the report, said Swanson, is to offer a fresh perspective on how well and for what reasons various school districts achieve their goals. "Most of the analyses that have been made of school-district effectiveness have focused on student achievement alone," he said, an assessment method to which Swanson and most other educators take exception. "The accurate measurement of school district quality is very complex and can't be captured by a single statistic," he insisted. "In this study," he said, "we've broadened the focus to include not only student achievement, but the unique effect that a school district has on student achievement." Swanson and Engert did this by ranking school districts by achievement test scores before and after controlling for the social and economic factors that influence academic achievement levels. The study points out that the RAND Corp. study undertaken in 1994 found that the most important family characteristic influencing student performance in school is the parents' education levels. Income, family size and mother's age at child's birth were modestly significant. Other signifiers of academic failure include a family whose primary language at home is not English. Whether the mother worked or not had a negligible effect and single-parenthood, while not significant in itself, highly correlated with other factors that have negative effects on achievement, such as family income. Since the presence of these factors differs markedly among school district populations, the work of some school districts, like those in Williamsville, Amherst, Orchard Park and Clarence, say the researchers, is assisted by favorable environmental factors. In districts like Buffalo and Niagara Falls, for instance, the work is frustrated by unfavorable environmental factors. Swanson and Engert say that by controlling for important risk factors in each district when considering achievement levels, the study offers a much more accurate picture of how successful -- how effective -- various districts are in educating the student population with which they must work. The study also includes a measure of school district efficiency, which is a ratio of school district effectiveness to the resources each one uses in the educational process. When looking only at achievement scores, he said, the greatest credit has tended to go to wealthy suburban districts because they are seen to be responsible for turning out students with high test scores. Predictably, many of the school districts in the area that have historically ranked low in tests of academic achievement are city districts and several rural districts with an average household income of $21,000: Jamestown, Ripley, Friendship, Salamanca, Hinsdale, Dunkirk, Gowanda, Buffalo, Limestone and Niagara Falls. No suburban school districts are among the bottom 10 in student-achievement level, but most districts in the top 10 in achievement are suburban districts with an average household income of $38,003. The public may be most surprised by the fact that the Buffalo city schools, whose school population has the highest at-risk rate (10.9%) of any of the 100 districts studied and ranks in the bottom 10 when considered in terms of achievement scores alone, nevertheless ranks among the top 10 districts in terms of its effect on student achievement. Other high-effect districts are Scio, LeRoy, Newfane, Frontier, Portville, Bolivar, Pavilion, Belmont and Pine Valley. wanson pointed out that these districts, because they are usually "rated" publicly only by achievement test scores, are not normally recognized for their positive impact. The study underscores that in Western New York, as in the rest of the country, student academic achievement levels can be predicted principally by a district's household and community per-capita income levels, parents' education levels and primary language spoken at home. When these factors are taken into account, the study points out, the percentage of students entering school in districts like Williamsville and Orchard Park who are at risk of academic failure is minute (.1 percent and 0.0 percent, respectively). City districts like Niagara Falls, Dunkirk, Jamestown, North Tonawanda, Buffalo and Salamanca, and many small, rural districts, must educate students that are, from the outset, at considerably higher educational risk. The percentages of at-risk students in Buffalo, Dunkirk and Jamestown city schools, for example, are 10.9 percent, 8.1 percent and 6.4 percent, respectively. "Published reports -- especially those that make comparisons among school districts and schools -- tend to oversimplify this very complex subject," said Swanson, "and often lead to misleading conclusions about how well a particular district is doing its job." Children likely to attend urban or some rural schools are much more dependent on the schools for their intellectual development than are children attending school in wealthy suburban districts. The study concludes that districts populated by people of relatively lower income and lower education levels confront a much more difficult task than upscale districts when it comes to educating children. That is why, when the study controlled for such influences, many of the smaller and poorer districts, including Buffalo, exhibited very high levels of effectiveness and efficiency in educating children. "I don't wish to imply that the wealthy school districts aren't good," said Swanson. "They are very good schools and work hard to complement work of the families in educating the pupils in their schools. "The point is, however, that for the wealthy districts, the effort is a complementary one, with parents and schools jointly providing incentives and an environment that enhances and encourages learning," he added. "In poorer districts with low per-capita education levels, the schools alone have to carry much of the responsibility for intellectual development of students. It's a very difficult job, but many of them do it extraordinarily well. "We hope this report, which offers what will be to the public a fresh analysis, will clarify the issues here and give credit where credit is due," he said. Policy implications of this analysis, Swanson and Engert said, include the suggestion of better use and coordination of resources already available for educational and social support services, and the need for additional resources for districts that serve high proportions of students from lower socio-economic status families.