The quality of education and the ability to provide this quality at a lower cost in a larger school setting were advantages touted by large school proponents who were encouraging consolidation efforts (School Renaissance Institute, 2000). Furthermore, fixed costs are distributed over a large school or school district in an economies of scale model (Young & Green, 2005), and this model has been used previously in industry. Presently the model is applied to the educational setting to improve the efficiency of cost and to measure the quality of education (Young & Green, 2005). The size and design of schools have changed over time to accommodate the need to provide a high quality education in an efficient manner (School Renaissance Institute, 2000). Consequently, the number of elementary schools with more than 900 students enrolled increased by 30% from 1987–1988 to the 1997–1998 school year (Office of Policy Planning and Research, 1999).

As educational leaders and communities plan schools, the theory of economies of scale is a critical component of the planning process to ensure an efficient use of the money needed to operate schools (Howley, 1996b). An economy of scale is demonstrated within schools or school districts when larger organizations can derive a lower cost per student due to the ability to distribute the expenses for fixed costs over more students (Porter, 2008). For example, because both large and small schools typically need administrators, teachers, and cafeteria staff, the cost per student might be less in larger schools than in smaller schools.

One study using the economy of scale theory was conducted by Dodson and Garrett (2004) who analyzed data from Arkansas school districts and noted that large economies of scale were present and savings were realized in personnel costs, supplies, and materials. “A consolidated school district can produce an equivalent level of output at a lower cost per student by avoiding redundant expenditures” (Dodson & Garrett, 2004, p. 271). Specific examples noted within this study included the fact that fewer administrators or specialists were needed as small schools were consolidated. Dodson and Garrett proposed that by consolidating Arkansas schools, teacher salaries would show a 30.6% cost savings per student compared teachers salaries at cost per student for each isolated rural district. Further, when smaller districts consolidate, they might be able to offer more competitive salaries than before, thereby retaining more qualified teachers (Dodson & Garrett, 2004).

Other researchers have applied the theory with mixed results. Cotton (2001) reported that at some point the costs continue to rise as the school increases in student enrollment. Imazeki (2006) indicated that in districts in California, economies of scale were demonstrated as well as a lack of economies of scale. Imazeki further reported that the average cost of education was lowest in a district that enrolled 28,992 students, but as the district continued to grow, the cost per student continued to rise. Slate and Jones (2005) reviewed literature and concluded that initially, increasing school size created positive results including training for teachers and the ability to recruit highly qualified teachers. However, as schools continued to increase in size, the effects of the increasing student population and the lack of needed resources tended to reverse the once positive effects of increasing school size. Ferris and West (2004) argued that savings projected using economies of scale might not apply to individual schools as well as the model could be applied to school districts. School districts might not be able to calculate an accurate cost per student because of several varying factors among schools that can affect costs such as school safety issues, teacher development needs, and special needs of students served in the school. As a result, Ferris and West (2004) contended that these external factors can create complex analyses which cannot be included in a cost function model.

Although Young and Green (2005) noted a focus on improving educational opportunities for students as an impetus for consolidation, cost effectiveness has been the motivational factor of the consolidation effort of small schools and districts for many years (Overbay, 2003). In fact, the topic of consolidation has been a controversial issue since the 1800s (Bard, Gardener, & Wieland, 2005). “Some reformers have argued that small rural schools are inefficient and ineffective and that consolidating them into larger organizations is the best, if not only, remedy” (Arnold, 2000, p. 2). Similarly, Conant (1959), a proponent of consolidation in the 1950s, asserted that large schools could provide a more diverse curriculum while maintaining a lower per student cost than smaller schools. As noted earlier, through consolidation, schools increased in enrollments from an average of 170 to 2,300 students per school between 1930 and 1970 and simultaneously, the numbers of schools and districts decreased as the numbers of students attending schools increased (Berry, 2004). As a result, the average school district was 14 times larger in 1970 as compared to 1930 (Berry, 2004).

Even though cost effectiveness has been and continues to be a major factor in widespread consolidation in the United States, researchers have challenged the cost effectiveness argument (e.g., Arnold, 2000; Bard et al., 2005; Howley, 2005; Monk & Haller, 1993). Plecki (1991) asserted that methodological errors occurred in most efficiency research regarding school size and costs. As such, most findings were based on district-level data that did not account for funding sources that might have been available to some schools but not equally to all schools within the district. For example, when student population characteristics differed greatly among schools within a district, the deviations affected the reported district average of available resources and the reported resources might have inaccurately reflected funds available at the campus level (Plecki, 1991). Furthermore, Fox (1981) noted a pattern in expenditures where spending was higher per student at the extreme ends of the school size spectrum. Schools that had the smallest populations and schools that had the largest populations demonstrated the largest expenditures per students. Bickel, Howley, Williams, and Glascock (2001) analyzed “single-unit schools” (p. 22), which contained Grades Kindergarten (K) through Grade 12. The only school in a small rural school district, the single-unit school, typically housed all elementary and secondary students in one building. This setting was deemed to be educationally and economically efficient (Bickel et al., 2001). For example, the researchers stated a cost savings of $1,017 per student expenditures within a single-unit Grades K-12 school. However, schools with such limited grade spans as Grades 3-5 were more costly per student than schools with wider grade spans.

In exploring construction costs of large and small schools, Howley (2005) analyzed two national datasets. He compared the means of construction costs of 168 different projects including large and small high schools and made projections about the final construction costs. Howley discovered that smaller high schools, with 138–600 students, were less expensive than larger high schools, with 601–999 students, per square foot. Howley (2005) reported that this cost difference might have occurred because administrators typically overestimated the predicted student enrollment of larger schools and underestimated the enrollments of smaller schools. Both large and small schools typically cost the same amount per student because smaller schools allocated 26% more space per student than did larger schools. Howley (2005) also reported another relevant variable impacting construction costs: the wealth of the community. If the community wealth surrounding the school was high, then the construction costs were increased to meet the community’s expectations for facility upgrades. As a result, the expected efficiency with larger schools was not evident in Howley’s (2005) study and he concluded that larger schools were not more efficient than smaller schools.

Conversely, Brown, Johnson, Doughty, Cecil, and Keck (2005) analyzed school enrollments and the cost per square foot between two smaller elementary schools compared to one larger elementary school. The researchers wanted to determine which building option for a new school was the most cost efficient. Two building prototypes were examined: (a) a two-school option with 640 students in Grades Pre-Kindergarten–3 and 400 students in Grades 4–5 and (b) a consolidated elementary school of 1,040 students in Grades Pre-kindergarten–5. The consolidated elementary school equated to a space savings of 23,058 square feet and a personnel cost savings in excess of $115,000. Within the first year, the projected cost savings for the one school option was over $3 million. In this same study, Brown et al. (2005) analyzed new building designs with architects to determine if the smaller elementary schools were as cost efficient as the larger consolidated elementary schools. Researchers reviewed schools of varying enrollments and grade spans and concluded that larger schools resulted in a cost savings per square foot and required less square footage per student than smaller schools. Deriving a similar conclusion in South Carolina elementary schools, Carpenter (2006) noted that per pupil operating costs in larger schools were less than operating costs in smaller elementary schools.

In another study related to consolidation, Streifel, Foldesy, and Holman (1991) examined 19 consolidated schools throughout the United States. They analyzed expenditures and revenues for each of the identified school districts for 3 years following the consolidation, comparing these financial data to the average spending for education in each state. Administrative costs were lower after consolidation and the authors noted that the administrative costs might be less than 5% of the overall budget. However, this additional cost might not be a substantial expenditure, especially in small rural districts where the number and cost of administrators are minimal. Expenditures that were not statistically significantly different compared to the state average were instruction, transportation, operations and maintenance, total costs, total revenue, and capital projects. The authors recommended that school district leaders understand their unique communities and that expenditures and revenues should be analyzed from all angles before a decision to consolidate is considered. In another similar study, Chakraborty, Basudeb, and Lewis (2000) analyzed cost and expenditure data from 40 school districts in Utah over a 3-year period and noted that economies of scale were present at both the school district and school level. The researchers concluded that as the school size increased, the average expenditures per student decreased. They recommended considering consolidation of schools or school districts to reduce the cost of education.

Although many of the studies cited thus far have focused on the differing operating costs of large and small elementary schools, Imazeki (2006) refined the analyses by examining subgroups of students and costs related to specific educational needs. Imazeki (2006) defined the base cost of education as the amount per pupil needed in a school district to serve students having a few special needs and the marginal cost as the amount needed to serve students whose needs require more money. Some examples of additional characteristics are a large economically disadvantaged population, a large population of students identified as LEP, or a concentrated number of students with severe disabilities. Imazeki noted that many states, including California, which was the state analyzed, provide some adjustments for these additional costs of education. Imazeki concluded that many of these adjustments were added to the school district funding formula after initial financial calculations were conducted. These additions were not directly linked to the true cost of education nor were these amounts sufficient to meet the unique needs of these special student populations.

Furthermore, Imazeki (2006) analyzed student achievement data for California to determine if school districts that were experiencing financial challenges were able to create school environments where students were demonstrating academic success. Using the financial information for each school district, the researcher examined how students were performing in districts with financial challenges compared to other districts not experiencing the same financial challenges. To measure performance, Imazeki averaged the Stanford-9 scores for each school district and created a summary score, the Academic Performance. Districts with larger numbers of students with more costly educational needs, such as large numbers of students from economically disadvantaged backgrounds, demonstrated lower student achievement in the API than did districts with smaller numbers of students with similar needs. Lower achievement was also predicted in higher cost districts because the funding provided was not sufficient to provide the quality of education needed to overcome the additional challenges. “Summed up across all the districts, the total cost to appropriately compensate all districts for their needs would be just over $49 billion, or roughly thirteen percent more than the cost of maintaining the status quo” (Imazeki, 2006, p. 28).

Other researchers have examined other variables besides financial data to determine effectiveness. Steifel, Berne, Iatarola, and Frucher (2000), in an attempt to focus on the substantial dropout rate in larger schools, refined the definition of cost effectiveness by calculating cost per on-time graduates instead of cost per enrolled student. Furthermore, Ferris and West (2004) purported that economies of scale for schools cannot be measured by production of student achievement alone. Rather, the measurement of the success of the cost of the function should be determined using external factors such as increased school violence and the perception of students regarding the quality of the school environment in addition to student achievement to determine the overall success rate of a school. Slate and Jones (2005) concluded that economic efficiency and educational outcomes represented two separate sets of analyses, each of which has its own correlation to school size. The influence of these factors might be affected by more complex analyses such as poverty level of the student population and the surrounding community. Slate and Jones (2005) recommended carefully considering a comprehensive evaluation of additional analyses that might influence student academic success in the school setting.

Besides cost efficiency and student achievement, another argument used to support smaller schools is that of school connectedness. School connectedness determines the degree to which students feel supported and cared for by teachers and staff in the educational setting (Blum, 2005a). To examine school connectedness, Blum (2005) studied children of military personnel who had to relocate frequently, which often resulted in interrupted schooling and social connections. Blum determined that students responded better and achieved at higher levels in learning environments that were individualized and considerate of the needs of each student. Consequently, Blum theorized that school connectedness resulted in a learning environment where relationships were valued. As such, many of these relationships have been documented in the small school setting.

Cotton (2001) applied the theory of school connectedness in a study and reported that the positive support found in smaller schools has been shown to reduce violence in schools and protect students from dropping out of school. Cotton argued that teachers in small schools are more likely to know the specific learning strengths of each student and such personalized instructional knowledge is more difficult in a large school environment. “When teachers and students are able to build relationships, both are motivated to work and to make a success of the schooling enterprise” (Cotton, 2001, p. 29). In another study, Brookmeyer et al. (2006) noted that connectedness provided students with a safeguard to overcome other barriers that might hinder academic development such as poverty.

Nearly 50 years ago, Barker and Gump (1964) asserted that small schools created an environment conducive to developing relationships and opportunities for student engagement. They compared student engagement of students in small schools compared to large schools and reported that student participation was much higher in smaller schools than in larger schools. In smaller schools, every student was needed as a key member so that teams could compete in athletic, academic, and extracurricular activities. In larger schools, so many possible participants were available that each individual’s participation was not as critical. Later, Lindsay (1982) analyzed extracurricular activity participation rates among 14,668 high school students in 328 schools. The author indicated that schools with 100 students or less had a higher extracurricular activity participation rate, student satisfaction, and attendance when the data were controlled for the socioeconomic status of the student population and student ability levels. Cotton (1996) confirmed earlier research that small schools demonstrated an important relationship between indicators such as student engagement and higher extracurricular participation.

Cotton (1996) reviewed 103 school size studies and research projects and concluded that a statistically significant relationship was present between small school size and higher student achievement in about one half of the studies, whereas the other one half of the studies did not confirm that relationship. In another meta-analysis, Leithwood and Jantzi (2009) reviewed 57 studies of school size effects and student achievement and concluded that students in smaller schools demonstrated higher achievement compared to larger schools. According to the researchers, this trend was critical to students from diverse and disadvantaged backgrounds as they consistently demonstrated higher academic achievement in small schools.

Examining the relationship between discipline and school size, the National Center for Education Statistics (NCES, 2006) surveyed principals and discovered that discipline problems were related to school size. In schools of more than 1,000 students, 34% of principals reported patterns of office referrals occurring at least once a week due to students being disrespectful to teachers. This result was compared to 21% of principals in schools with 500–999 students, 17% schools of principals in schools with 300–499 students, and 14% of schools of principals in schools less than 300 students. Cotton (1996, 2001) also affirmed that students in small schools exhibited positive social behaviors at school noted by the lack of discipline incidents, higher attendance rates, lower dropout rates, a positive concept of school, and the ability to experience success for every student.

Students who have disadvantaged economic backgrounds and who live in rural environments might encounter complex obstacles. Consequently, high academic success has not typically been the norm in high poverty schools in Texas or throughout the United States whether the setting is rural or urban (Kannapel & Clements, 2005). Beeson and Strange (2000) described rural schools as geographically isolated and more likely to have a larger percentage of economically disadvantaged students. Rural areas are more likely to demonstrate significant levels of economic distress in the United States as compared to metropolitan areas. Of the 250 poorest counties in the United States, 244 counties were rural (Beeson & Strange, 2000). Researchers have noted that small, rural schools still face barriers to providing a high quality education to all students. Some barriers include poverty, school finance formulas, isolation, and difficulty competing for high quality teachers and administrators (Arnold, 2000). Some small, rural schools have overcome these challenges, whereas others struggle to maintain successful schools despite a multitude of economic barriers. However, some rural area educators are suspicious of reform efforts for their schools due to negative past experiences with consolidation reform efforts as well as a mistrust of reform efforts initiated by political entities outside the local district (Arnold, 2000).

Small schools have demonstrated a particular effectiveness for impoverished students (Howley, 1996a). Johnson (2006) suggested that consolidating small disadvantaged districts to create larger high schools would likely increase the economic barriers of poverty on student academic achievement. “The larger the district, the more magnified the negative effects of poverty over student achievement, and the smaller the district, the more poverty’s effects are muted” (Johnson, 2006, p. 8). Stevenson (2006) revealed that small schools demonstrated trends of keeping disadvantaged students in school longer, which prepared these students with the skills needed to be more able to fulfill graduation requirements. Stevenson also noted that small schools were better able to fulfill the personal needs of students, resulting in future productive citizens. Cotton (1996) and Stevenson (2006) both observed how small schools added value to the school experience for these disadvantaged students by providing them opportunities for success in a small school setting versus allowing them to become dropouts in a larger school setting. Bickel, Howley, Williams, and Glascock (2001) examined the relationship among school size, student achievement, and the cost of education per student by analyzing 1,001 high schools in Texas. The researchers noted that smaller schools lessened the effects of economic disadvantages for students, whereas larger schools exaggerated these economic disadvantages. Furthermore, Cotton (2001) concluded that small schools can provide opportunities for poor and minority students to perform at high academic levels.

In reviewing the option to consolidate small, rural districts with high levels of poverty, Slate and Jones (2005) proposed that any educational initiative should be examined as to whether it improves the academic achievement of students. Similarly, Bard et al. (2005) concluded that educational stakeholders must consider financial implications of consolidation along with student achievement and the community surrounding the school. “More and more it seems that small schools hold particular promise for helping impoverished students maximize their potential to achieve academically. This hardly means that small schools are the best choice for all students under all circumstances” (Howley, 1996b, p. 5).

The one-room schoolhouse of the 1900s is a rare event in today’s schools (School Renaissance Institute, 2000). In the mid-1900s, students were being educated in 117,000 school districts throughout the United States. Within 50 years, the number of school districts had dropped below 16,000 whereas the population of the United States had grown by 70% (Lashway, 1999). In theory, the consolidation of small school districts provided students equitable opportunities for class choices, labs, library resources, and kept the cost of schools lower while serving more students (Lashway, 1999).

The trend toward large schools continues in the 21st century. Mitchell (2000) reported that about 40% of the secondary schools in the United States now have more than 1,000 students and some schools have enrollments as high as 5,000. Many secondary schools average between 2,000 and 3,000 students. In fact, from the 1990s into the 21st century, “the number of high schools with more than 1,500 students doubled” (Mitchell, 2000, p. 13). Moreover, DeJong and Locker (2006) stated that families are attracted to large schools as they seek out schools that offer a variety of extracurricular activities and diverse course offerings for students.

An example of a very large school was DeWitt Clinton High School in the Bronx, New York that claimed a student population of more than 12,000 in 1934; however, population and demographic shifts caused this high school’s student enrollment to rise and fall over the years (Allen, 2002). In 2002, Clinton High School had a student population of 3,864 and was in the process of whole school reform. Allen described a reform process of creating houses within the school to meet the needs of the diverse student population. These houses, or small groups of students, encouraged the development of relationships with other students and faculty as a strategy for increasing student performance. Allen noted challenges in communicating with parents and the community that educating students in a large school was feasible.

Howley and Bickel (1999) replicated a study in seven states throughout the United States (i.e., California, Alaska, West Virginia, Ohio, Georgia, Texas, Montana). Interestingly, they concluded that students in high socioeconomic status communities might benefit from larger schools whereas students from low socioeconomic status might struggle to perform at high academic levels in larger schools. Johnson (2006) suggested that large schools have the ability to offer a variety of classes. On the other hand, the participation rates of the diverse classes offered at larger schools were lower than in smaller schools. As such, Johnson noted that students in larger schools with a plethora of classes might avoid taking challenging classes and might avoid being noticed.

Creating groups of smaller schools-within-schools was recommended by Lee and Smith (1997). Duke and Trautvetter (2001) promoted updating and redesigning existing buildings in need of improvements with a new design, which would allow for smaller learning environments within larger facilities. Three examples of schools within schools were (a) freestanding small schools, which manage their own budget, administration, and staff; (b) multiplexes, which encompass several small schools in one building with the small schools sharing the same administration; and (c) a school within school, which is a small school housed within a larger school. Overbay (2003) recommended creating the positive attributes of smaller schools within existing schools by redesigning the school and classroom cultures. Terms such as academies and houses have been noted by several researchers (e.g., Clark, Hager, & Nikolova, 2006; Cotton, 1996, 2001; DeJong & Locker, 2006; Duke & Trautvetter, 2001; Lee & Smith, 1997). Houses have a common curricular focus and team approach whereas academies might have a career orientation that offers students opportunities to build skills in a specialized area (DeJong & Locker, 2006; Duke & Trautvetter, 2001). Academic teaming has been another approach used in large schools where teams of teachers share a common group of students to provide a more connected and integrated curricular learning environment (The Office of Policy Planning and Research, 1999). Several of these organization patterns (i.e., houses, academies, teaming) have been offered as a way to make large schools seem smaller and more personal for students.

To address the various barriers identified in larger schools, Bill and Melinda Gates have funded over $284 million towards creating small schools within large schools in fast growing cities such as New York City and Chicago so that these growing urban areas can implement educational reform in existing school facilities (Gates Foundation, 2005). For example, Chicago Public Schools implemented a small schools initiative in 2001 and developed nine small schools from larger existing schools. Moreover, the leaders of the initiative have sought to open 30 more small schools (Gates Foundation, 2003). As planned, the new small high schools housed no more than 100 students per grade with a total student population limited to 400 students. The focus of the initiative was to combine the personalization of relationships characteristic of small learning environments and the breadth of course offerings existing in larger schools (Gates Foundation, 2003). A further example of implementation of small schools within large urban schools is the New York City Smalls Schools Initiative, which has graduated over 70% of its students for the second consecutive year (New York City Department of Education, 2007). The graduation rate of the 47 Small School Initiative Schools was 35% in 2002 before the new program was implemented. More than two thirds of students in the New York City Small School Initiative’s graduation class entered these schools demonstrating academic performance below grade level. In excess of 90% of the students were Black or Hispanic, and 30 of the 47 small schools resided in buildings that were previously planned for closure (New York City Department of Education, 2007).

Johnson et al. (2002) examined the effects of race on student achievement, particularly with African American students. Johnson et al. divided all of the Arkansas schools into the top one fourth of schools that served the highest percentage of African American students and the lower three fourths of schools that served the lowest percentage of African American students. African American students overwhelmingly attended larger schools. This finding demonstrated that the relationship between poverty and school size was three times more powerful in schools that served the highest percentage of African American students compared to other schools. The authors recommended that policy makers and school administrators consider the student academic gains that might be available to larger schools and districts if these schools are divided into smaller schools and districts. Consequently, this recommendation is critical for larger schools that serve a high percentage of African American students. Johnson et al. (2002) noted a statistically significant relationship between small school size and student achievement for African American students. Further, African American students demonstrated higher academic success in schools that had smaller concentrations of economically disadvantaged students.

Slate and Jones (2007b) investigated the relationship between high school size and the academic achievement of Hispanic students in Texas. Data analyzed were TAKS scores in reading and math, end-of-course examinations, SAT I and ACT scores, percent of students who were coded as economically disadvantaged, attendance, dropout rates, and graduation rates for all Texas high schools. In this study, schools were defined in enrollment categories. Small schools were defined as schools that had 400–799 students, medium schools had 800–1,199 students, and large schools had 1,200 or more students enrolled. Slate and Jones (2007b) reported that Hispanic students in larger schools scored higher on the SAT I and ACT than did Hispanic students enrolled in smaller schools. On the other hand, Hispanic students in smaller schools were more likely to stay in school and graduate than students enrolled in larger schools. Their conclusion that small schools were better at connecting with students and keeping students in school confirmed prior research (e.g., Barker & Gump, 1964; Blum, 2005; Cotton, 2001; Howley & Bickel, 2000; Martin & Slate, 1998). Moreover, Slate and Jones (2007b) documented that students in small schools demonstrated higher achievement in reading and writing. Although end-of-course examination scores in algebra and U.S. History were higher in large schools, students in small schools scored higher in biology and English. Attendance rates and graduation rates were higher in small schools.

In another study, Gilmore (2007) examined middle school size and the academic performance of three ethnic groups. African American, Hispanic, and White middle school (Grades 6–8) students performed higher academically when they attended very large middle schools (Gilmore, 2007). In this study, very large middle schools were described as schools that enrolled more than 1,999 students. Gilmore (2007) affirmed other researchers’ findings that larger-sized schools provided students an opportunity to experience more academic success than did smaller schools (Chavez, 2002; Raywid, 1999).