The physical condition of the nation's public schools has been an important topic of discussion among policymakers, educators, and parents in recent years. Newspaper stories (e.g., Nakamura 2000; Seymour 2000) and research studies (e.g., U.S. GAO 1995a) describing broken plumbing, poor ventilation, and overcrowding have raised concerns about the effects of school facilities on teaching and learning. There is also apprehension that older schools with outdated electrical wiring will be left behind newer schools in the effort to connect schools to the Internet. More importantly, some conditions, like sagging roofs or lead exposure, raise serious concerns about student and teacher safety.
Over the past decade, a number of lawsuits challenging school funding for facilities have drawn attention to the poor conditions that many students encounter at school. According to the Arizona Supreme Court, for example:
Some districts have schoolhouses that are unsafe, unhealthy,
and in violation of building, fire, and safety codes. Some
districts use dirt lots for playgrounds. There are schools
without libraries, science laboratories, computer rooms,
art programs, gymnasiums, and auditoriums. But in other
districts, there are schools with indoor swimming pools,
a domed stadium, science laboratories, television studios,
well stocked libraries, satellite dishes, and extensive
computer systems [Roosevelt Elementary School No. 66
v. Bishop, 877 P. 2d 806 (Ariz. 1994)].
Similar descriptions can be found in a number of reports released by the United States General Accounting Office (GAO). According to a 1994 GAO study (1995a), approximately 14 million students attended schools that were in need of extensive repairs in 1994. In fact, at that time a majority of the nation's public schools (60 percent) were in need of repair.
The purpose of this report is to provide national data about the condition of public schools in 1999 based on a survey conducted by the National Center for Education Statistics (NCES) using its Fast Response Survey System (FRSS). FRSS is a survey system designed to collect small amounts of issue-oriented data with minimal burden on respondents and within a relatively short timeframe. The results presented in this report are based on questionnaire data for 903 public elementary and secondary schools in the United States. The questionnaire responses were weighted to produce national estimates that represent all regular public schools in the United States.
This report provides information about the condition of school facilities and the costs to bring them into good condition; school plans for repairs, renovations, and replacements; the age of public schools; and overcrowding and practices used to address overcrowding. Information about the condition of school facilities is based on questionnaire rating scales rather than on physical observation of school conditions by outside observers. The remainder of this chapter provides background information about the condition of school facilities.
Observations of school facilities have appeared in headlines, speeches, and reports that focus on the deteriorating environmental and physical conditions of the nation's schools. The Education Writers Association (Lewis et al. 1989), for example, reported a decade ago that the physical condition of one out of every four school buildings in America was inadequate. Of these, over 60 percent were labeled inadequate due to lack of repair and maintenance. In 1995, GAO released a report on school facilities indicating that things had not improved and appeared to have grown worse (U.S. GAO 1995a). According to GAO, one-third of both elementary and secondary schools reported having one entire building in need of extensive repairs or replacement. Approximately 60 percent of schools reported that at least one building featureneeded extensive repair, overhaul, or replacement.
The features most in need of repair, according to GAO (1995a), included heating, ventilation, air conditioning, plumbing, roofs, exterior walls, electrical power, windows, and doors. Because of this state of disrepair, 41 percent of schools reported poor energy efficiency. In addition, 50 percent of schools reported having at least one unsatisfactory environmental condition, including conditions that violate federal mandates (e.g., exposure to lead and radon gas). In addition, the decay and neglect described by GAO decreased the ability of many older school buildings to meet the proposed technology goals. In some schools, old, outdated wiring makes the use of technology for both educational and administrative purposes impossible (Hansen 1992).
Despite considerable focus on the unsuitable conditions of many of America's schools, Kozol (1991) and others (e.g., Corcoran, Walker, and White 1988; Lewis et al. 1989) highlight an important point: the condition of the nation's schools varies widely. Some schools are in poor condition as previously described, and some schools are in exceptional condition. Most schools, however, fall somewhere in the middle; they are in "adequate" or "better" overall condition (U.S. GAO 1995a). School conditions, whether poor, average, or exceptional, often vary by location (e.g., urban versus suburban) and characteristics of the community (e.g., impoverished versus wealthy). For example, GAO (1996) reported that in 1994, the largest proportion of schools reporting deficient school conditions was in central cities serving more than 50 percent minority students or 70 percent or more poor students. However, GAO (1996) also found that poor conditions exist in many rural areas; one out of every two rural schools had at least one inadequate structural or mechanical feature.
Financing public education in the United States relies on the combined effort of state and local appropriations, as well as funds available from the federal government (Howell and Miller 1997). Because school facilities are funded primarily by local revenues, characteristics of the community, particularly the property tax base, are important factors contributing to the condition of schools (Lewis et al. 1989; U.S. Department of Education 1995). Because community factors tend to vary, they lead to disparity in the funding available for schools from community to community (Augenblick, Myers, and Anderson 1997; U.S. GAO 1995c; U.S. Department of Education 1995; Terman and Behrman 1997).
Disparities in funding for schools in general, and school facilities in particular, are exacerbated by the cost for providing facilities that are in good condition. There are a number of studies on school construction costs. Some of them focus on funds spent on school facilities; others report estimates of funds needed for school facilities. While each one offers a unique perspective, differences between funds spent and funds needed, and differences in the elements each study includes under school construction, make it impossible to draw direct comparisons across the studies.
The 1994 GAO study estimated that $101 billion was needed for repairs, renovations, and modernizations to bring schools' onsite buildings into good overall condition (U.S. GAO 1995a).
The study also estimated that $11 billion dollars would be needed in the next 3 years to comply with various federal mandates that impact school construction, such as asbestos removal and accessibility for students with disabilities. 6
A more recent GAO study of actual school construction expenditures included expenditures for acquired and constructed buildings, land, and equipment such as heating and air conditioning systems (U.S. GAO 2000). This study found that annual construction expenditures for public schools in the United States grew by 39 percent from fiscal years 1990 through 1997, from about $17.8 billion to about $24.7 billion after adjusting for inflation.
In another study on school construction costs, Abramson (1999) reported that public schools completed construction projects totaling $15 billion in 1998, with a projected expenditure of $17.2 billion for 1999. Of the projected spending in 1999, approximately 50 percent was to be used for new construction, 24 percent for additions, and 26 percent for renovations. With advances in technology, new construction is becoming increasingly expensive. Abramson (1999) provided data concerning the facilities that were planned for new schools constructed in 1999.
While these new schools all will have the core features of cafeterias and libraries (among other features), most will have a computer lab (60 percent of elementary, 91 percent of middle, and 85 percent of high schools) and local area networks (LANs) (80 percent of elementary, 95 percent of middle, and 90 percent of high schools), and many secondary schools will have technology labs (33 percent of middle schools and 44 percent of high schools).
A recently published National Education Association (NEA) report looked at funds needed for school infrastructure and education technology (NEA 2000). In this report, school infrastructure included new school construction, additions to existing buildings, renovation, retrofitting, deferred maintenance, and major improvements to grounds. In addition, education technology included computers and peripherals, software, connectivity, networks, technology infrastructure, distance education, maintenance and repair of technology equipment, and technology-related professional development and ongoing support for teachers. Taking all of this into account, the NEA provided a cost estimate of $322 billion needed for school modernization. 7
The funding of public school facilities may be important insofar as it affects the way schools address two important factors contributing to the decline of school facilities: (1) deferred maintenance and renovation, and (2) overcrowding (e.g., U.S. GAO 1995a; Hansen 1992; U.S. Department of Education 1999a).
A number of reports have raised concerns about the age of America's public schools (e.g., U.S. Department of Education 1999b). Older buildings may fall into disrepair or lack the infrastructure necessary for adequate electrical and telecommunications wiring (U.S. GAO 1995b). However, age of the building, by itself, is somewhat less important than is its history of maintenance and renovation, as well as the adequacy of the original construction. Regular maintenance is critical to keeping schools in good condition.
The growing demand for new school construction, as well as choices that school districts must make about where to spend limited funds (e.g., facilities versus instructional programs), has forced many school districts to overlook the maintenance and modernization of old schools. For example, GAO (1995a) found that district officials attributed declining conditions primarily to insufficient funds, resulting from decisions to defer maintenance and repair expenditures from year to year. However, maintenance can only be deferred for a short period of time before school facilities begin to deteriorate in noticeable ways. Without regular maintenance, equipment begins to break down, indoor air problems multiply, and buildings fall into greater disrepair (Hansen 1992). The lack of regular maintenance can also result in a host of health and safety problems, including exposure to carbon monoxide and risk of physical injuries.
Additionally, deferred maintenance increases the cost of maintaining school facilities; it speeds up the deterioration of buildings and the need to replace equipment (Hansen 1992).
Like deferred maintenance and renovation, overcrowding is an important topic to consider when examining the condition of school facilities.
Overcrowding occurs when the number of students enrolled in the school is larger than the number of students the school was designed to accommodate. It both characterizes the condition of school facilities (e.g., the facilities are too small to accommodate the students and teachers who reside there) and contributes to the decline of these facilities (e.g., overcrowded facilities are typically overused facilities that grow old before their time). A number of recent reports indicate that overcrowding is a serious problem in many school districts (Burnett 1995; Corcoran, Walker, and White 1988; Lewis et al. 1989; Fernandez and Timpane 1995; U.S. GAO 1995a; EdSource 1998; Rivera-Batiz and Marti 1995; Lowe 1996; U.S. Department of Education 1998; U.S. Department of Education 1999a). Dramatic increases in enrollment due to the "baby-boom echo," immigration, and migration have led many schools to enroll far more students than they were designed to accommodate. 8 Compounding these conditions are initiatives to reduce class size, resulting in the need for even more classrooms.
To deal with overcrowding, school districts and schools have adopted a number of short-term solutions. For example, many have converted noninstructional space into classrooms. In such schools, students are placed in spaces never intended as classrooms, such as gymnasiums, libraries, cafeterias, and even closets (Burnett 1995). Another popular solution is portable classrooms. These temporary structures are becoming a more permanent feature of schools; some have been in use for as long as 40 years (Lewis et al. 1989). Portables are a prominent adaptation school districts are using to meet space needs; however, recent accounts suggest that they may not be a particularly suitable alternative (U.S. GAO 1995a). Findings from a recent U.S. Department of Education report (1999a) suggest that overcrowding will likely become more serious in the future and that short-term solutions, such as portable classrooms, may not be sufficient to accommodate the long-term enrollment boom that is expected over the next two decades.
There are many ways in which the condition of school facilities may be directly or indirectly relevant to students and their families. The issues of greatest concern surrounding the topic-as reflected in the popular press, the research literature, and courtrooms nationwide-include equal access to adequate school facilities and student safety, as well as more academic matters, such as student achievement.
As described earlier in this chapter, many states have been sued over the degree of disparity in their school facilities funding systems. For example, variations in the quality of Ohio's public school facilities have been cited as key evidence for the violation of the uniform education articles provided by the state's own constitution (e.g., The State of Ohio v. DeRolph, 677 N.E. 2d 733 [Ohio 1997]). The courts' interpretations of uniform education as it relates to facilities have gone beyond the right to have access to adequate facilities and materials; several courts have asserted that widely disparate school funding systems-in particular, funding for facilities-prevent students from attaining equal educational opportunities, achievement, and job opportunities (e.g., The State of Ohio v. DeRolph, 677 N.E. 2d 733 [Ohio 1997]; Roosevelt Elementary School No. 66 v. Bishop, 877 P. 2d 806 [Ariz. 1994]).
Lack of adequate school facilities may also result in conditions that compromise student safety. Students who attend schools in poor condition may be exposed to such health threats as poor air quality, hazardous materials, and sewage overflows. In more extreme cases, students may also be put in more immediate physical danger when parts of the building collapse or when safety features (e.g., fire alarms, sprinkler systems) fail. In fact, witnesses in the Ohio v.DeRolph case testified that arsenic was found in the drinking water of an elementary school and that chunks of plaster routinely fell from the ceilings of several schools. Similarly, GAO (1995a) documented numerous individual accounts of threats to student safety: the ceiling of an Alabama elementary school collapsed less than an hour after students had left for the day; the glass from 70-year-old windows blew into a Washington, D.C., elementary classroom during a windstorm; and many students who suffer from asthma were frequently put at risk by poor ventilation and unhealthy air quality. To compound the problem, in 1994, over 14,000 schools across the country had less than adequate life safety features, such as sprinkler systems (U.S. GAO 1995a). Therefore, dangerous conditions that may normally be prevented or controlled (e.g., fires, carbon monoxide levels) were not adequately monitored.
Many studies have explored the relationship between school conditions and achievementrelated behavior (Burnett 1995; McGuffey 1982; Rivera-Batiz and Marti 1995; Weinstein 1979).
For example, Earthman and Lemasters (1996) reviewed several recent studies of the influence of school conditions on academic achievement. All of the studies found the same relationship, to varying degrees, between school facilities and student achievement. That is, higher rankings of structural and mechanical (e.g., heating and cooling systems) or cosmetic conditions (e.g., how recently the walls have been painted) were correlated with higher achievement scores. This relationship was generally found to be stronger for cosmetic than structural school conditions.
However, Earthman and Lemasters caution readers about the findings; they concluded that despite a large number of research studies, "it is difficult to determine any definite line of consistent findings" (p. 3). These conclusions have been supported by other researchers. For example, Ferguson (1991), in a study of school expenditures, stated that "some [of these] expenditures-particularly transportation and maintenance-have logical relationships to learning that are at best indirect" (p. 484). He concluded that "few would argue that spending more on transportation and maintenance should increase test scores, though most would agree that such spending matters" (p. 484).
Much of the research on the effects of school facilities on student achievement shares a number of general limitations. For example, many of the studies Earthman and Lemasters (1996) reviewed provided no evidence of statistical testing (e.g., Cash 1993; Earthman, Cash, and Van Berkum 1995). Another study (Berner 1993) suffers from different methodological concerns, including a sample size too small to support the analysis the author runs (e.g., a regression model with 41 cases and 9 independent variables) and control variables that are 9 years older than the variables measuring school condition and student achievement.
More importantly, however, many of these research studies did not take into account a number of meaningful differences between schools in poor and good conditions that may explain the findings they report. For example, schools in poor condition may be less likely to have resources important for academic achievement, such as high-quality teachers, effective leadership, high levels of parental involvement, and more appropriate materials (e.g., laboratory equipment, textbooks). These resources, rather than the condition of the physical plant, may explain achievement differences between students in schools of various conditions.
Researchers may have had difficulty establishing a relationship between facilities and achievement because if school conditions do have an impact on student learning, their effects are likely to be indirect. Specifically, student learning is generally believed to be affected by factors such as lost instructional time, reduced attention, and diminished curricular options that may result from facilities-related problems (e.g., school closings, classroom shortages, overcrowding; Duke et al. 1998; Rivera-Batiz and Marti 1995). It is difficult to detect indirect effects in the absence of detailed surveys of students and teachers and suitable measures to control for other differences between schools in poor condition and schools in good condition.
However, researchers who have studied school facilities and the teaching and learning process have found some interesting relationships. For example, a survey of overcrowded schools in New York City found that 75 percent of teachers indicated that overcrowding affects classroom activities, and 70 percent of teachers indicated that overcrowding affected their instructional techniques (Rivera-Batiz and Marti 1995). Corcoran, Walker, and White (1988) found that overcrowding, as well as heavy teacher workloads, created stressful working conditions for teachers and led to higher teacher absenteeism. The survey of New York City schools mentioned earlier found that nearly 40 percent of students indicated that they had problems concentrating in their classes when they were learning something new (Rivera-Batiz and Marti 1995).
The survey on the Condition of Public School Facilities was conducted through NCES' FRSS during summer and early fall 1999. The results presented in this report are based on questionnaire data for 903 public elementary and secondary schools in the United States. Information about the condition of school facilities is based on questionnaire rating scales rather than on physical observation of school conditions by outside observers. While individual schools were sampled, the questionnaires were mailed to the districts with which the schools were associated.
The cover letter indicated that the survey was designed to be completed by district-level personnel who were very familiar with the school facilities in the district. The letter indicated that the respondent might want to consult with other district-level personnel or with school-level personnel, such as the principal of the selected school, in answering some of the questions. The respondent section on the front of the questionnaire indicated that while most questionnaires were completed by district-level respondents, some were completed by schoollevel respondents (usually the school principal).
To maintain the focus on schools, which are the sampled unit, the report refers to schools indicating or reporting various findings, even though respondents were primarily district-level personnel reporting about the sampled school.
Many of the questionnaire items on the FRSS survey are taken from the 1994 GAO survey. The same questionnaire items and analysis variables were used with the intention of providing information about change in the condition of public school facilities between 1994, when GAO conducted its survey, and 1999, when NCES conducted its survey. However, the GAO information included in this report is provided as contextual information only. Statistical comparisons are not provided because GAO does not provide standard errors for the data in their reports, and exact point estimates are also missing for some comparative statements from the GAO reports.
Like the 1994 GAO study, this FRSS study also asked for an estimate of the total costs of all repairs, renovations, and modernizations required to bring the onsite buildings into good overall condition. However, for the FRSS study, schools for which the condition of any type of onsite school building or any building feature (e.g., roofs, plumbing) was less than good provided information about the cost of needed repairs, renovations, and modernizations. Thus, even though the wording of the cost item on the 1994 GAO and 1999 FRSS studies was the same, the two studies include costs for different things. 9 The school characteristics used as analysis variables in this report are school instructional level, school enrollment size, locale (central city, urban fringe/large town, rural/small town), region, percent minority enrollment, and percent of students in the school eligible for free or reduced-price school lunch (which indicates the concentration of poverty in the school). These variables are defined in appendix A.
The questionnaire responses were weighted to produce national estimates that represent all regular public schools in the United States. All comparative statements in this report have been tested for statistical significance using chi-square tests or t-tests adjusted for multiple comparisons using the Bonferroni adjustment and are significant at the 0.05 level. Throughout this report, differences that may appear large (particularly those by school characteristics) may not be statistically significant. This is due in part to the relatively large standard errors surrounding the estimates (because of the small sample size), and the use of the Bonferroni adjustment to control for multiple comparisons. Appendix A provides a detailed discussion of the sample and survey methodology.
The chapters that follow present information about the condition of America's public schools in 1999. Specifically, chapter 2 presents information about the presence and overall condition of various types of onsite buildings, and the condition of nine different building features (e.g., roofs, plumbing). Information is also provided about the cost to put school buildings into good overall condition, and the sources of estimates for those costs. Chapter 3 reports about satisfaction with various environmental factors (e.g., heating, ventilation) in the schools' onsite buildings, and provides information about the status and satisfaction with air conditioning in various areas of the schools. Chapter 4 discusses school plans for new construction and for major repair, renovation, or replacement of building features in the next 2 years. Chapter 5 examines the issue of the age of America's public schools, including determining the age of schools, and how age relates to the condition of schools and plans for repair, renovation, and replacement.
Chapter 6 presents data regarding the extent of overcrowding in public schools, the relationship between overcrowding and school condition, and various practices that schools may use to ease overcrowding. The concluding chapter summarizes the findings of this study and draws some overall conclusions about the findings. Technical information, including a detailed study methodology (appendix A) and tables of standard errors for all data presented in this report (appendix B), are included as technical appendices to the report. The questionnaire is included in appendix C.
6The $101 billion and the $11 billion were collected in two separate questions on the survey. However, GAO frequently presents an estimate of $112 billion needed, which they derive by summing the amounts reported in these two questions. It is possible that the $112 billion includes some duplication of money needed, since the $11 billion needed to comply with federal mandates may or may not have been included by respondents in the $101 billion needed to put schools into good overall condition.
7While NEA describes the study as a 50-state report of school modernization needs, the study received usable responses about infrastructure from only 24 states, and about education technology from only 2 states. The remaining data were derived by various estimation techniques described in the report.
8Migration patterns (e.g., families moving out of particular areas) and decisions families make with regard to their children's schooling (e.g., private school enrollment) may also lead to a decline in enrollments among some public schools. These declines may result in schools that are underenrolled.
9See appendix A for additional information about the 1994 GAO cost estimate, and further discussion of comparability issues for the cost estimates between the 1994 GAO and 1999 FRSS studies.