A stream of research studies in recent years has added substantially to our knowledge about the relationship between school size and student achievement. It shows that student socio-economic background is a major factor in how school size affects student outcomes. This has important implications for the Stanhope Government’s proposal to close 39 schools and to partially close 5 other schools.
There is a voluminous research literature on the effect of school size on student achievement. It generally shows that students perform better in smaller elementary and middle schools while the results for high schools are mixed.
However, much of this research literature has overlooked the possibility that school size may be associated with different outcomes for students from different backgrounds. This gap has been rectified by a range of state-wide and national studies in the United States since the mid-1990s.
Almost without exception, the studies show that small school size is unambiguously good for students from low socio-economic status (SES) backgrounds and communities with relatively high levels of disadvantage. Students from low SES backgrounds achieve better results in smaller schools. Small schools with high concentrations of students from low SES backgrounds have higher average results than large schools with similar concentrations.
Large schools do academic harm to students from low SES backgrounds. As schools get larger, average achievement among schools enrolling high proportions of low SES students declines. As one study says, increasing school size “imposes increasing ‘achievement costs’ in schools serving impoverished communities”.
Small schools mitigate the effect of low income and poverty on student achievement, the relationship being substantially weaker among smaller schools than among larger schools. In general, the impact of poverty on student achievement in small schools is estimated to be about half that in large schools.
For example, a 2004 Ohio University study of national data shows that students attending the smallest schools experience a 60 per cent reduction in the influence of SES on mathematics performance, a 39 per cent reduction on reading performance, a 50 per cent reduction for science, and a 45 per cent reduction for history.
The benefits of small schools for students from low SES backgrounds appear to be particularly important in the middle years of schooling, when some students start becoming susceptible to dropping out of school in later years.
The methodology used in these studies has undergone rigorous assessment, most recently in May this year by a review at the University of Maine’s College of Education. The findings have proved robust for different technical specifications of the modeling procedures used for the statistical analysis, which is a rare degree of consistency in educational research.
Many of the studies included statistical controls for a range of other factors that influence student achievement. For example, a 2001 study of Texas schools included controls for ethnicity, language, size, expenditure per student, and curricular composition factors including special education programs. The inclusion of these factors did not significantly alter the results.
These studies offer no support for government proposals to close small schools, especially those serving communities with significant levels of socio-economic disadvantage. The following conclusion from one study is representative:
Findings from this study obviously offer no support for arrangements that work to increase the size of already small schools, especially those that serve impoverished communities….In light of the findings from this and other studies, concern for achievement and for reducing achievement gaps means that educators and policy makers must search for ways to meet these challenges without closing schools that are already appropriately small.
Indeed, increasing school size may produce educational effects that are the opposite of those that that the Stanhope Government claims it intends with Towards 2020. School consolidation, without regard to student background, is likely to increase the large inequity that already exists in ACT school outcomes and diminish the outcomes of some students. In particular, it may undermine recent progress in improving Indigenous outcomes.
The ACT Government has ignored these research studies in developing its school closure plan. The Towards 2020 website fails to include them in its links to papers on school size research. This is unfortunate because the studies have clear policy implications.
First, small schools should be maintained in low SES communities and the most impoverished communities should be served by the smallest schools. Second, government policies should strengthen the benefits of smaller schools serving low income families, rather than seek to close them. Third, large schools may not be as cost effective as is often assumed, especially if they lower outcomes for significant groups of students and increase inequity in education.
Trevor Cobbold