Smaller Schools Research

Big Trouble: Solving Education Problems Means Rethinking Super-Sized Schools and Districts

by David N. Cox
View as a PDF

We Utahns love to have everything from our fast-food meals to our trucks "super-sized"; you might even say our families are "super-sized"! In many cases bigger is better, but not always. When it comes to schools and school districts, we as a nation have typically consolidated districts to make them larger under the assumption that bigger was better, specifically, more cost-effective with better academics. Yet several decades of super-size districts have called that notion into question for quality of services and even cost efficiency. Some districts might be too big, and might do better to DE-consolidate.

Over half of all students in Utah reside in only four of our forty districts. Our two largest districts, Jordan and Granite, have over 70,000 students each. Nationally we have about 14,800 school districts. Jordan, Granite, Davis, and Alpine are in the top 100 for size.

Many parents in giant school districts nation-wide are not happy with their schools. Those who can, demonstrate their displeasure by "voting with their feet" and moving their children into private schools. Robert W. Jewell found a strong correlation between non-Catholic private school enrollment and large district and school size. He states,"...the larger the districts and schools, the greater the non-Catholic private school enrollments among the states." Utah seems at present to be bucking the trend. We have the smallest percentage of private school enrollment of the states. Perhaps our larger family size has made that option too expensive. However, there is a growing segment of the electorate that are demanding vouchers and tuition tax-credits. If passed, these options would make it possible for many more to "opt-out".

Support for school choice is becoming almost a litmus test for the majority Republican Party here in Utah. Even a former Democratic State Party Chair is involved in the push. School administrators, who have stated their solid opposition, need to address the why, even if they win temporary battles now. Is the growing desire for private schools because our schools and districts have become too big?

Big Roots of Dissent

What root problem of public education has disaffected so many parents in the last twenty years that they want to place their children in private or charter schools?

Are poor test results causing parents' dissent? Test results can indicate that a problem exists, but they are not the problem itself. Some people suggest that declining test scores show that students today are not learning as well as they used to; others argue that the test results are being misinterpreted. Still others say that the tests themselves are not valid indicators of education progress. In any case, the test results themselves are not driving the school choice movement.

Is it curriculum? Utah has not yet had the math wars and reading wars that California and other states have had. That likelihood is becoming very real now that whole math and the latest version of whole language "balanced literacy" have been adopted by the state office. The constructivist philosophy of avoiding the discipline parts of learning (which include direct instruction of the basic skills of grammar, times tables, and phonics, etc.) is creating a division among parents and between teachers. Many parents and teachers want these taught directly but find it increasingly impossible to influence district and state curriculum committees. Intimidating statements like, "You don't want to be one of those 'drill and kill' teachers do you?" or "Don't let those old (experienced) teachers influence you as you begin teaching." or "You're the only one who has voiced this concern." are used to silence or isolate curriculum debate. This truly is alienating many would be friends of education.

Consider the debates over the teaching of evolution versus biblical creation of the early twentieth century, sex education, and many others. There have always been disagreements about curriculum, but people did not consider leaving public education over it. Regardless of what is done in the next twenty years, curriculum will still be an issue, as well it should be. Every generation needs to learn, focus, and decide what is of importance and how and to what degree it should be taught. Beliefs about curriculum, though very germane to the problem, have not caused the call to get rid of public education.

Are taxes driving the dissatisfaction with public schools? While Americans have protested taxes since before our nation was formed, in most every state and national poll for the last twenty years Americans have expressed support for increasing taxes IF it would improve schools. This sentiment opens the debates about where and how these funds should be spent, and whether more money improves schools at all. But the fact remains that many parents and citizens would be happy to pay much more for education if they had some control of it.

Control is the key. Who controls the quality, financing, and curriculum of our schools? While many citizens intuitively feel these decisions are best made by the parents and teachers of students in the local neighborhood, under our current system, for various reasons, the power to make those decisions rests in the hands of school boards and administrators who oversee increasingly large districts.

Big, how did we get there?

In 1900 there were more than 150,000 school districts in the United States. Single-school school districts were common. There were even some districts that were so small they employed only one teacher! As the assembly-line factory-model took hold in industry, the economies of scale began to be understood and applied in other settings such as schools. Utah had 380 districts until 1915 when they were consolidated into 39 and later the 40 districts we now have. After World War II there was a stronger push nationwide to consolidate school districts in rural areas. Many people were leaving agriculture for the city, resulting in less students and taxpayers for schools in rural areas. There were still about 128,000 districts in 1930, but by the end of the 60's that number had shrunk to 36,000. Today there are less than 15,000 school districts nationwide. During this time our cities were growing and with them their school districts. The net effect was fewer, bigger districts. We now have 24 districts with more than 100,000 students.

Parents in the middle of the century were convinced by administrators to allow consolidation of schools and districts for the economies of scale and for increased opportunities. Many moved to the city for the same reasons; school taxes would be less and schools would offer more because of a bigger base. At that time the diseconomies of scale (being too big) weren't even on the horizon because our cities and districts weren't as big as they are today. In 1959, J. B. Conant, in his report "The American High School Today," called for an increase in high school size to 400 students. Educational Research Service, 1971, referring to 26 studies completed between 1939 and 1969, stated that the most common recommendation for district size was 10,000, and that, "The decrease in the total number of school districts has been 85.9%....The job is, however, far from completed." indicating the need to further consolidate.7 Since then we have increased both school and district size way beyond what the proponents of larger size were suggesting at that time. We now have high schools of 5,000 and districts up to 1 million students.

Big disappointments

These large districts have not brought the envisioned savings financially. "Webb & Ohm (1984) found smaller districts more efficient than larger ones in both dollars per student and numbers of administrators per student...." Antonucci found that there are "penalties of scale." Instead of making up a larger percentage of the budget as school districts size increases, the percentage spent on teachers, books, and teaching materials goes down! He writes, "Paradoxically, the larger a school district gets, the more resources it devotes to secondary or even non-essential activities." McGuire, in a 1989 study found, "As specialization in staff grows, program offerings expand, and administrative personnel increase, problems of coordination and control also increase. And in large systems, time and energy are more likely to be shifted away from core service activities." Antonucci also writes, "And let's not forget the labor implications. Which district is more likely to have difficult contract negotiations or work stoppages? The district with 15 bus drivers, or the one with 677 bus drivers?"

1999 Utah administrative costs per pupil show little difference between large districts and small unless one gets below 1000 students. Below 1000 students the administrative costs go up. (See Figure 1) The three lowest districts in administrative cost per student were Logan with 5,840 students - $181 cost per student (cps), South Sanpete with 2,878 students - $198 cps, and Juab with 1796 students - $207 cps. Alpine with 45,208 is next with $237 cps. In the 2000 legislative audit on class size reduction monies, the smaller districts were better able to account for specific funding than big districts because the big district budgets were so complicated.

Nor have bigger districts necessarily provided better education. Cotton, in her review of 100 research projects observes that, "The states with the largest schools and school districts have the worst achievement, affective, and social outcomes." According to Webb the researchers have fallen into two camps, those who see no advantage for big districts and those who find "...that achievement drops as enrollment levels rise." She states that this is even more evident in lower socio-economic populations,"...there was a strong, consistent negative correlation between district size and student achievement in low-SES populations." Walberg goes even farther to show a direct line, negative relationship of the states with large district size and test results. Jewell states point blank that, "Students in states with smaller districts and smaller schools have higher SAT and ACT scores."

Comparisons with Utah school districts 1997-2000 SAT scores are revealing. (See Figure 2) In both 5th and 8th grade test scores, the smallest districts score highest within their expected range (99% and 74% of expected range). Medium small districts score next highest (65% and 47%), medium size score next (54% & 37%), with the largest districts scoring lowest in their expected range (36% & 32%). 11th grade scores do follow the socio-economic expectations which are upward with district size but this may be because the lower scoring students in bigger districts have dropped out. Larger schools, which are mostly in the larger districts, have higher dropout rates prompting the analysis that while smaller schools are slightly more expensive, the cost per graduate is less in smaller schools.

Big Districts = Big Schools

Very closely tied to the district size issue is the school size issue. What seems to be true with districts has even more research showing it to be true with school size. Cotton and Irmsher in their reviews of over 100 research projects regarding size show conclusively that bigger is not better once schools go beyond a certain size. Irmsher writes, "Michael Klonsky (1995), and Mary Anne Raywid (1995), and others report that large school size hurts attendance and dampens enthusiasm for involvement in school activities. Large schools have lower grade averages and standardized-test scores coupled with higher dropout rates and more problems with violence, security, and drug abuse." She says that with smaller schools, "Security improves and violence decreases, as does student alcohol and drug abuse." Cotton states, "Behavior problems are so much greater in larger schools that any possible virtue of larger size is canceled out be the difficulties of maintaining an orderly learning environment."

Larger schools are not necessarily less expensive either. "Generally, there is agreement that unit costs are higher in the smallest and largest schools. Various studies characterize per-pupil school costs as having a U-shaped average cost curve, where costs are high in both the smallest and largest schools." When cost per graduate is calculated, smaller schools are a better deal.

It appears that district size is probably the most direct cause of school size. Jewell says that large schools are concentrated in large districts. He shows that Utah is 6th in average district size and also 6th in average size of school. This connection makes sense when one considers how school buildings are funded. Buildings are funded through voter approved bonds. When a district is large, administrators find it hard to get positive votes for bonding in areas that do not need the new school. These areas are reluctant to increase their taxes because they don't see the need and won't feel the benefit. Over time some bonds don't pass, forcing administrators to build fewer schools. Those that are built must then be made bigger. To gain voter approval in non-growth areas, administrators then offer additions to existing buildings making them bigger as well. Therefore in big districts both old and new buildings end up bigger.

There may be other reasons that promote larger schools. Patrons needing another school may want everything that was built or installed in the last school built. The booster club may want more enhanced sports facilities, for instance, making it politically necessary to add something to the bond for the older schools to buy voter approval in those areas. This may make the bond bigger than desired and so instead of building two smaller schools that are needed, administrators compromise and build one larger one with all the extras.

Utah's big districts have public perceptions that the other side of the district is getting more benefit than their side. Davis and Alpine's north and south sides and Granite and Jordan's east and west sides have fought each other for years over which direction the tax dollars are flowing. Smaller districts may be more likely to pass bonds because the community as a whole would see the need and feel the benefit. They would not have to fight another community for political power and tax dollars. They would not have to take from another area for their own benefit with the resulting negative feelings. And they might decide that the extras weren't worth the extra taxes.

In areas of high growth, school size continues to grow, seldom reversing itself. One reason may be that the growth also means the district grows and with it, the aforementioned problems. An additional
problem presents itself here for future discussion, how does a district build more or smaller schools on land that already has houses on it? If a school is not built with a new development, the land
for a new school becomes very expensive and usually can only be acquired by condemnation procedures. This makes it politically very difficult and prohibitively expensive to purchase.

Some districts have tried to solve big school problems with the "school within a school" approach. This is an attempt to gain the recognized benefits of smaller size without really becoming smaller and has resulted in some mixed improvement. The idea is to reorganize into two or more smaller groupings within the existing building. "The major challenge to schools within schools," writes Mary Ann Raywid, "has been obtaining sufficient separateness and autonomy to permit staff members to generate a distinctive environment and to carry out their own vision of schooling." In 1996 Howley reported, "If size is a structural phenomenon, however, caution is warranted in approaching the simulation of small size through such mechanisms as "schools-within-schools" and "house plans" (Meier, 1995; Oxley, 1994; Raywid, 1996). In general, despite substantial popularity, research on the effectiveness of simulating small size as a way to restructure is very limited (Raywid, 1996)." Transportation is not alleviated with this model either.

Similarly some big districts have tried to gain the advantages of smaller neighborhood districts using ideas such as subdistricts (Los Angeles School District, 2000), cluster and cone councils that some Utah districts use, or attempting to push more control to local school levels. It is very difficult to really turn over control and responsibility to these groups, however, because they are not legally the entity in charge. They cannot raise revenues or allocate them, and this leaves them as mere arms of the larger organization. "Modern means of decentralizing funding and governance within multi-layered educational organizations - state accountability schemes, school-site management, New York City's community boards, and Chicago's local school councils - have yet to prove their value. Nor have 'home rooms' and 'schools within schools' shown that they can recapture the advantages of small schools." 'Kincaid concluded: 'Virtually all of the factors most associated with academically effective education are school- and neighborhood-based. Yet, we have shifted more control and financing of education to state and national institutions."

Big Conclusions

What does the current state of affairs portent for the future? According to the 1999 Fall Enrollment Report from USOE, 85% of Utah's high school students are in schools that research show are too big. Utah is among the highest in district size. It may not be possible to build smaller schools without creating smaller districts. The current situation is not necessarily less expensive either. Excessive transportation costs and traffic problems resulting from schools that are too big have not been addressed in this report. Utah's smaller districts score better, not because the educators of big districts try less, but because the bureaucracy, which a large district must have for control, ties their hands. It certainly does not engender community spirit toward a common goal in education. Substantial studies show that bigger is not better and that 'bigness' is alienating citizens toward our educational institutions. This cannot help but influence other governmental entities such as the relationships with cities inside these megadistricts.

This alienation is depleting local support, and that will result in increased state controls and diversion of funding for basic instruction into accountability schemes in an attempt to force quality, quality that will become increasingly elusive. Utah's scores seem to have leveled out or are starting to decline. Growth of population is continuing and will result in ever bigger schools and further distance of the districts from the citizens. Back in 1989, Walberg intimated the following as what we could expect as a result of becoming too big. He listed:
  • declarations of educational bankruptcy and state appointment of "receiverships" of new boards and central staff;
  • breaking up large-city districts into free-standing smaller units;
  • suits by parents for failing to employ state-of-the-art educational practices;
  • litigation by graduates for fraudulent services and diplomas;
  • magnet schools and choice plans within and outside districts and the public sector;
  • and vouchers and tax rebates for private or public tuition.
He further wrote, "These schemes are motivated by the desperation of some legislators, business people, citizens, and parents who wish to employ the courts or market-like competition to improve the efficiency of schools, particularly those in large districts, that seem unable to respond constructively to their clients and society."

Most of these consequences we have begun to see coming true, if not in Utah then elsewhere in the country. Of the six, only one has any future for really improving the education of the general public and that is breaking up our large districts. State takeovers would be completed socialism. How could the state possibly run a district better? Suits and litigation tie more red tape around a bureaucracy that is already floundering. Magnet and charter schools are only imitations of what could be found with smaller schools in smaller districts. A charter school is essentially a one-school, new district that has no local tax base. Vouchers and tuition tax-credits, though very enticing to true believers in individual freedoms and responsibilities, will not gain wide-spread voter approval or improve public schools if implemented. They too, will get wrapped up in litigation.
Setting a limit on the size for both districts and schools and creating an orderly way for setting up these new districts, will achieve better academics and a more efficient use of tax dollars long term. It will encourage more participation by both students and citizens. Cities and schools will work together to solve transportation and traffic problems around schools, as well as future development problems. Why? Because only when our noses are pushed into our own messes do we seem to cleanup after ourselves and solve similar problems for the future.

Smaller districts and schools bring the problems and opportunities back to the local level. This spurs innovation, flexibility, and commitment by both parents and teachers. We cannot build students without building parents at the same time, and that builds citizenship as well. Only by creating new, smaller districts will we return liberty and responsibility to the local parents and teachers. Only then will true accountability be accomplished. Only then will true educational quality and efficiency be possible to achieve.


  1. Robert W. Jewell, "School and School District Size Relationships, Costs, Results, Minorities, and Private School Enrollments" Education and Urban Society, 1989 pp. 140-153.
  2. As one example, BYU's 1999 Poll of The Public's Attitudes Toward the Public Schools in Utah, by the Department of Educational Leadership, shows a steady increase from 1994's 57% support to pay more taxes to help raise the standard of education in Utah, to 59% in 1997, 62% in 1998, and 64% in 1999. Gallup polls of the 1990's have also confirmed this nationwide.
  3. Minutes of Education Interim Committee, Utah Legislature, September 16, 1987.
  4. Pipho, C., "Rural Education" Phi Delta Kappan, 1987 pp. 6-7. See also: McGuire, Kent, "School Size, The Continuing Controversy" Education and Urban Society, 1989 p. 173.
  5. Antonucci, Mike, "Mission Creep: How Large School Districts Lose Sight of the Objective -- Student Learning" Alexis de Tocqueville Institution Brief #176, November 17, 1999.
  6. Conant, J. B., "The American High School Today" Carnegie Series in American Education, 1959.
  7. Webb, Florence R., "A District of a Certain Size" Education and Urban Society, 1989 p. 127-128
  8. ibid, p. 130.
  9. ibid, p. 134-135.
  10. Walberg, Herbert J., "On Local Control: Is Bigger Better?" University of Chicago, No. 59 Nov. 22, 1993.
  11. Cotton, Kathleen, "School Size, School Climate, and Student Performance" Northwest Regional Educational Laboratory, Close-Up #20. (
  12. Irmsher, Karen, "School Size" ERIC Digest 113 July 1997. (
  13. Walberg, Herbert J., "District Size and Student Learning," Education and Urban Society, 1989 p. 128.
  14. K. McGuire, "School Size: The continuing controversy," Education and Urban Society 21 (2, February): 164-174
  15. Langdon, Philip, "Students do better in Small Schools so why have we been making schools bigger?," The American Enterprise, January 2000 p. 25.
  16. Costs were compiled by taking the total reported administrative costs in each district for 1999 and dividing by the number of students in each district.
  17. Raywid, M. A. "Family choice arrangements in public schools: A review of the literature." Review of Educational Research, 55(4), p. 455, (1985).
  18. Howley, Craig, "Ongoing Dilemmas of School Size: A Short Story" ERIC Digest, December 1996.