Creating a better school may be as simple as creating a smaller one. The results of two recent studies indicate that small schools may be the remedy for lots of things that are wrong with public education, especially for the nation's poor children. The separate studies credit small schools with reducing the negative effects of poverty on student achievement, reducing student violence, increasing parent involvement, and making students feel accountable for their behavior and grades.
Educators have long known that poverty hurts student achievement. Researchers Craig Howley, of Ohio University and the Appalachia Educational Laboratory, and Robert Bickel, of Marshall University, set out to find out whether smaller schools could reduce the negative effects of poverty on student achievement. In four separate studies of seven states, they repeatedly found that poor kids do better if they attend a small school. In fact, in the most recent four-state study, the correlation between poverty and low achievement was ten times stronger in larger schools than in smaller ones in all four states. Howley and Bickel found that the benefit of smaller schools was particularly important in the middle grades, when children are most at risk of dropping out.
The researchers initially evaluated schools in California. Howley then replicated the research in schools in West Virginia and Alaska.
More recently, the Rural School and Community Trust, a national nonprofit organization, asked Bickel and Howley to study School Size, Poverty, and Student Achievement in Montana, Ohio, Georgia, and Texas. The study included 13,600 urban, suburban, and rural schools in 2,290 school districts.
They found that at least one-fourth of the schools serving moderate- to low-income communities in Texas, one-third in Georgia, and two-fifths in Ohio are too large for students to achieve top performance. In Montana, among kids in grade 8 in larger districts, the power of poverty over achievement was 2.5 times greater than in smaller districts overall and three times greater in elementary-only districts.
SIZE MATTERS"Everyone knows that there is a strong association between social class and achievement and that this association works very much to the disadvantage of economically disadvantaged students," Bickel told Education World. "The California research, however, had the virtue of demonstrating that this disadvantage was exaggerated as school size increased."
Each time Bickel and Howley conducted another study, the results were very similar. "It's very unusual in education research to find this degree of consistency," Bickel said.
Howley and Bickel did not base their findings on a definition of what constitutes a large or small school but looked at school size on a continuum. They found that poor students from relatively smaller schools outperform poor students from larger schools.
TEST SCORES DROP IN LARGE SCHOOLS
SMALL SCHOOLS MAKE 'CENTS'"A common argument for making schools larger is expressed in terms of economics of scale: Large schools save money," Bickel said. "Recently, however, using the Texas data set, we have found that 116 districts that have only one school for all grades have an expenditure per pupil that averages about $389 lower than the more conventionally modern schools. These schools tend to be small, they have at least 13 grade levels from kindergarten to grade 12, and the students are distributed more or less evenly across grade levels.
"Hardly sounds like a modern consolidated school," Bickel continued. "So perhaps cost in dollar terms is not a barrier to making schools more equitable places."
'TEACHER SATISFACTION WENT WAY UP!'Another study also linked student achievement with small schools. The two-year study, Small Schools: Great Strides, was conducted by Bank Street College of Education and funded by the Joyce Foundation.
A team of seven researchers took a close look at 150 small schools in Chicago, many created as part of education reform that started in the city during the past decade. The schools had enrollments between 200 to 400 students, far below the national average of 741, said Pat Wasley, one of the principal co-investigators of the study.