Few aspects of education have been more thoroughly researched than school size; few findings have been more consistent or more consistently ignored. More than 40 years ago, in 1964, Roger Barker and Paul Gump's book, "Big School, Small School," summarized hundreds of studies that concluded small schools are better.  The optimum size was reported to be about 400 to 500 students. This has been confirmed by numerous studies and reports since then.  Norway,  in 1978, acted on such findings and established 450 as the maximum size for high schools.

Some education reformers, such as Douglas Heath, have said an enrollment of 400 to 500 students is almost too large. Neil Postman and Charles Weingartner suggested 250 students and Ted Sizer has said no school should have more than 200 students.

Herbert J. Kiesling in a study of high schools enrolling from 100 to 4,000 students found a negative relationship between size and achievement.  That is, as the schools got bigger, student achievement declined while the rates of absenteeism, dropouts, discipline problems, disorder and violence went up.  For instance, high schools with 2,000 students average a dropout rate twice as high as those with 667 students.

A survey of 12,708 elementary teachers in Chicago disclosed their belief that there is more progress in small schools than in large ones.

Of necessity, a smaller percentage of students in high schools with large enrollments take part in school events such as athletics, the school paper and class offices.  The percentage of student participation has been shown to peak in high schools with 61 to 150 students.

Those who say small schools are not "efficient" or effective, need to cite the evidence, not just the rhetoric.  The nation's nearly 100,000 public schools with 50,000,000 students average about 500 students each  The nation's 25,000 nonpublic schools have 5,000,000 students, for an average enrollment of only 200.. New charter schools average fewer than 200, and many have fewer than 100. Years ago, the National Association of Secondary School Principals (NASSP), adopted a resolution that no high school should have more than 600 students.

In recent years the Bill and Melinda Gates Foundation and others have awarded large grants to districts such as New York City and Chicago to assist in creating smaller high schools.   Beginning in 2005 Philadelphia started a process to both build new smaller high schools and create some by converting larger ones.  Los Angeles and other large urban districts are joining the trend.

Since 2002 New York City has created 197 new, smaller high schools.  The results: the smaller high schools have higher attendance rates and in 2006 they graduated 79% of their students.   Not least of all, 81% of their graduates apply to college.  That is, about 65% of the students at small high schools apply to college while the larger traditional high school don't even graduate that many (58%).

Many reasons have been given for the decline of SAT scores in recent decades, such as the growth of teacher unions and the weakening of family structures. These factors may well play a role. But so does school size with larger schools coinciding with the lower SATs.

Beginning in the mid 1980s, the Kansas City School District and the state of Missouri, under the order of a federal court judge, spent more than $1.5 billion extra dollars, much of which went to build magnificent new schools. The money was spent in a district with 40,000 students when the project started, tallying up to an extra $37,500 per pupil. Yet dropout rates remained high, attendance and achievement rates were poor and enrollment decreased.

All of this is separate from the usual questions that arise when a new school buildings is proposed, such as is a new school needed, what type should it be, or can it be afforded. Those are significant questions. But the question as to what is the optimum size for the type of school being suggested is virtually never raised. Even when it is, it is usually answered on the basis of personal opinions, rather than knowledge based on actual experience.