Hey Nancy,

What's Up With That

Just another WordPress site

Optimal School District Size

We had some fascinating education headlines last week.  Perhaps none more interesting than the report of comments made by Mark Elgart, CEO of AdvancED, the accreditation conglomerate that owns many regional accreditors including the Southern Association of Colleges and Schools (SACS).

Here’s some background – Georgia is constitutionally limited in the number of school districts to 159 county districts and 21 city district.  Last year, Rep. Tom Taylor filed HR 486; a bill calling for a statewide vote to amend the constitution to allow new school districts to form under certain conditions.  A feasibility study was commissioned for the City of Dunwoody to determine if an independent school district was viable from a revenue standpoint.  The study’s results indicate that a city school district would be financially feasible and, at current millage rates, would produce a healthy surplus.

Speaking before the Buckhead Business Association days after the feasibility study was made public, Dr. Elgart stated the current 180 school districts in Georgia are “far too many.” According to The Reporter Newspaper, he went on to state, “Georgia does not need to expand the number of school systems it has in the state, … It needs to contract it so it can use its resources differently than it currently does.”

I’m puzzled why the head of an international accrediting agency would comment on a state political subdivision matter.  The organization of school districts is a self-determination made by the good citizens of our state.  Notwithstanding that fact, the suggestion that Georgia has “far too many” school districts is not supported by the research on the topic of optimal school district size.

Here are just a few quotes from scholarly articles on the subject of school district size that support the need for Georgia to break-up its large districts.

  • In a study to examine if consolidating smaller school districts in Michigan would save taxpayers money, Andrew Coulson estimated the most cost-effective school district size in Michigan and the cost savings that would result from merging small districts and breaking up excessively large districts. From his analysis, Coulson found that the most cost-effective district size for schools in Michigan was 2,900 students. Districts that were either larger or smaller in size would generate higher per-pupil costs (Coulson, 2007). Consolidating smaller school districts to achieve this optimal size was estimated to result in a cost savings for the state of Michigan and local governments of approximately $31 million annually. In comparison, breaking up large school districts would produce an annual savings of $363 million. The savings from breaking up large districts is estimated to be 12 times greater than the savings that would be generated from merging small districts.

Center for Evaluation and Education Policy
Education Policy Brief
Revisiting School District Consolidation Issues
Terry E. Spradlin, Fatima R. Carson, Sara E. Hess, and Jonathan A. Plucker

  • Small size is good for the performance of impoverished schools, but it now seems as well that small district size is also good for the performance of such schools

The Influence of Scale on School Performance: A Multi-Level Extension of the Matthew Principle
Robert Bickel, Marshall University; Craig Howley, Ohio University and AEL, Inc.

  • A study of Pennsylvania districts found that the lowest costs per student were in districts enrolling between 2,500 and 2,999 students (Standard & Poor’s School Evaluation Services, 2007).
  • A North Carolina report compared the district sizes of the five states with the best and worst SAT and ACT scores, high school graduation rates, dropout rates and retention rates. The study found that the states performing at higher levels on these performance indicators had smaller average district sizes (Sher & Schaller, 1986).
  • A Nebraska study demonstrated that smaller school systems academically outperformed larger ones within the state (Johnson, 2004). Researchers in Maine found that their 15 smallest districts produced higher graduation and post secondary enrollment rates than their 15 largest districts (Bowen, as cited in Driscoll, 2008). In Massachusetts, a task force found that smaller districts had lower average dropout rates, higher attendance rates, greater extra-curricular participation, and were more likely to meet Annual Yearly Progress (AYP) targets than the state average (Driscoll, 2008). A study of small rural districts in New York found that students in these small districts tended to learn the basics at average or above average levels, when compared to students in other districts (Monk & Haller, 1986). In a series of five studies, researchers found that smaller districts and schools had greater achievement equity than larger districts and schools (Howley, 1996; Bickel & Howley, 2000).

An Exploration of District Consolidation:
By:Kathryn Rooney and John Augenblick
Augenblick, Palaich and Associates, Inc.
May, 2009

The abundance of research indicates that the optimal district size is certainly much smaller than DeKalb’s current enrollment.  That research shows us that per-pupil costs are minimized in much smaller districts; completely negating the argument of economies of scale with large districts.  Furthermore, academic achievement measurements are better in smaller districts, particularly for the economically disadvantaged.  In the face of this type of evidence it is difficult to understand any defense of the status quo or advocacy for even larger districts.  The evidence is clear and compelling that our students and taxpayers would benefit from breaking up large districts.

(10) comments
  1. Thank you for taking Elgart to task on this ridiculous comment. I did my own very unscientific research after reading his statement to see how many school districts high performing states like Massachusetts, New Jersey, and Minnesota have relative to their population. Surprise – they have two to four times more school districts than we do relative to the number of residents.

    I went to school in a small town district with one high school. It was probably pretty close to the 3000 pupil optimal size these studies point to. It was not super high-performing, but it was accountable. A decade after I graduated the system was forced to consolidate into a county-wide system after losing a case that the town took all the way to the North Carolina supreme court. My friends who still live in the community bemoan the loss of their accountable city schools.

  2. Thank you Nancy for bringing data and facts to the conversation. It is clear Mr. Elgart does not care about student achievement, only about allocation of resources, money.

  3. @Refugee – You’re welcome! The facts from a financial and student achievement standpoint support the idea of districts that are much smaller than DeKalb.

    Texas has 1080 districts – approximately one district for every 4300 students
    Massachusetts has 327 districts – approximately one district for every 2900 students
    Michigan has 576 districts – approximately one district for every 2900 students
    Missouri has 534 districts – approximately one district for every 1700 students
    New Jersey has 598 districts – approximately one district for every 2300 students
    California has 1082 districts – approximately one district for every 5600 students
    Georgia has 180 districts – approximately one district for every 9200 students

  4. @Kirk – Thanks! If Elgart were concerned about resource allocation, he’d support the efficiency of smaller districts. The cost curve is U-shaped with a much smaller district providing optimal resource allocation (i.e., minimizes per-pupil administration/overhead costs).

  5. I find it interesting that you look to other states to determine what is an “appropriate” size of school districts, instead of examining the success of districts of varying size within Georgia. One of Georgia’s most successful school districts is its largest, Gwinnett County, where nearly 165,000 students at 132 schools are served. While the size of the district may be a factor, it is one of many. Leadership and community involvement in local schools are just two factors that I would argue have a much greater impact than the number of students or schools. Often, having too many school districts leads to too many central offices, where a great deal of resources may be wasted in a place that is not the classroom. There’s no doubt that Dekalb County Schools are a mess, but there needs to be a solution that benefits all students. It’s an important discussion to have.

  6. Nancy, you definitely have my vote. So many of us that have grown up elsewhere and seen the efficiency of small districts. How the biggest benefit a small one high school large district does is it ensures buy-in with the community. It insures neighborhood stability, people don’t have to worry about the central palace workers moving school attendance lines yearly to justify their jobs or provide lucrative contracts for friends. My Grandmother bought our family land in Western New York State at a Depression era Foreclosure auction in 1934. Both my parents went to the same school. In 1960 they split the school district into two districts to keep it at one high school large. Our family trust owns the property on the line. My siblings and I went to the new school district created in 1960 with the new school district. My nieces and nephews are split between the two districts. It really doesn’t matter because both of the one high school large districts are always in the News Week top 1000 performing Schools. They offer three diploma tracks Regents with AP, Regents, and General Vocational. They share a Vocational school that high school jr. And seniors go to 1/2 day their senior year and learn an actual real trade. For example, you have all the hours needed for a cosmetology license, so you don’t have to pay for one of these expensive Private schools, or you can learn carpentry, welding, pipe fitting. the graduation rates in these districts are 99%. Everyone’s kids from the Doctor’s to the trash collectors go to the public schools. if you do choose to go to Private school, the prices for those are a 1/3. There are no Private High schools within 30 miles since there is no market. Supt. Make $150 K and directly supervise the Admin. and Principals. The school boards are elected but volunteer like in a condo or HOA. Their are no big Contracts to skim off.

  7. Nancy – you position on school choice? I believe education money should follow the student, to whichever school that student is enrolled in. Comments?

  8. John E,

    I don’t know Nancy’s position, but I agree with you to a point. Money should follow the student to any school that is not faith based.

    Also, I am interested in Nancy’s view of rural districts that are smaller than 2,500 students. Should they be consolidated?

  9. John E – Thanks for commenting on my blog. I am an ardent supporter of school choice. I have written and spoken on this topic for several years. Here’s a link to one of my articles published in the AJC last year:


    Re: funding
    Your idea about money following the student is important. Sadly, our current funding formula is byzantine and is not attached to the student. On my website (the “About” page) I state:

    “We must simplify funding, attach it to the child, and improve transparency and disclosure to the citizens of Georgia.

    I want to work with other elected officials in our state to develop and implement a simpler funding method that accrues to each pupil. Our current system is byzantine and requires excessive administrative time to utilize. It is fraught with the potential for errors and manipulation and not easily audited.”

    Thanks again for commenting. Best Wishes for a Happy New Year!

  10. Kirk,
    Studies generally show that there is a size interval for school districts where administrative costs per pupil is minimized. The cost curve appears to be U-shaped where districts below and above that interval have higher administrative costs per pupil.

    So, you’ve asked if districts below 2500 should be consolidated. As far as merging school districts, that would be something that those, legally established, municipalities would need to decide for themselves.

    Thanks for commenting!

Leave a Reply

Your email address will not be published. Required fields are marked *