We Know Our Schools

All school evaluations, like all politics, are local

Citizens like their local schools much better than they like the nation’s public schools in general. According to the 2009 Education Next survey, 60 percent give their local elementary school an A or B, while only 18 percent give the nation’s schools one of those two grades.

How do people evaluate their local schools? Are their ratings based on reliable measures of effectiveness? Or do they base their evaluations on other kinds of information? With this issue of Education Next, we can now answer this question. Using an innovative technique made possible by Internet surveys and geo-coding technology, Martin West and his colleagues at Harvard’s Program on Education Policy and Governance (“Grading Schools,” research) were able to match each member of a nationally representative sample of adults to the specific elementary and middle schools that serve his or her neighborhood. As a result, respondents’ grades for their local schools could be compared to the actual performance of those schools on state math and reading tests. The analysts also collected publicly available information on the school’s average class size, racial and ethnic composition, and the percentage of students who were of low income.

From their findings, we learn that American citizens know quite a bit about the local schools. Indeed, schools that score high on statewide tests receive high evaluations from those surveyed. Within the larger population, parents turn out to be particularly adept at determining which schools are good and which are not—welcome news, indeed. And despite all the hoopla over class size, citizens’ judgments about a school’s quality are unrelated to how large or small its classes are.

Critics of school choice often claim that parents ignore quality when evaluating schools and draw their conclusions on the basis of the school’s racial or ethnic composition. But this study shows that parents are indifferent to student race as long as a school’s pupils perform well. (They do, however, give higher marks to schools with fewer low-income children.)

Citizens are less impressed with their local middle schools. Only 49 percent were willing to give them an A or B, and they were almost twice as likely to assign middle schools a D or F than they were elementary schools (12 percent vs. 7 percent). In the second research study in this issue (“Stuck in the Middle,” research), Jonah Rockoff and Benjamin Lockwood show that judgment is also right on target. Their analysis of student achievement in New York City middle schools confirms parents’ conclusion that children learn more if they stay in an elementary-school setting through grade 8 than if they move to a stand-alone middle school.

That finding called to mind what it was like when I was introduced to junior high school in 7th grade many years ago. Suddenly, bells rang, kids ran around, teachers shouted, lockers banged, and no one learned a thing. Not at all like the tranquil elementary school I had previously attended.

A final caveat. Parents tend to compare their local school to others within their own state. Those living in parts of the country with lower-quality schools apparently have little idea that schools in other states are, on average, a lot better. Could such provincialism be corrected by grading all schools on a common, nationwide scale, such as national standards advocates propose? Or are all school judgments inevitably just as local as streetwise politician Thomas P. “Tip” O’Neill said of all politics?

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428
Email Education_Next@hks.harvard.edu

For subscription service to the printed journal
Phone (617) 496-5488
Email subscriptions@educationnext.org

Copyright © 2024 President & Fellows of Harvard College