Impacts on health, well-being and performance may be hard to recognize. But indoor pollution levels may be 2-5 times, and occasionally 100 times, higher than outdoor levels, according to the U.S. Environmental Protection Agency (EPA). Studies indicate most Americans spend about 90 percent of their time indoors. Children are especially vulnerable because of the amount of time they spend indoors during the school day. (Ostendorf , 2001, para. 2)

The physical characteristics of aging or poorly designed schools can also inhibit learning with poor lighting, plumbing, and temperature control systems. The decision to build educational facilities with fewer windows in favor of fluorescent lighting may have reduced the amount of heat loss, but may also have created a more serious risk to health and performance. According to Lackney (2000), natural light and artificial full-spectrum lighting has been found to minimize mental fatigue as well as reduce hyperactivity in children, while students tend to react more positively to classrooms that have windows. Further, it has been found that fluorescent lighting may be related to greater amounts of hyperactivity in learners. Thermal comfort is also an important issue in relation to school facilities. Lackney (2000) states that classroom temperatures affect task performance and studentsπ attention spans.

Leaky plumbing systems in poorly ventilated schools contribute to the growth of mold on bathroom surfaces (Davis, 2001). The affects of mold in the environment can be as minor as simple irritation of the sinuses or much more serious depending on the duration of the exposure and the susceptibility of those suffering from the effects. Some people experience temporary effects which disappear when they vacate the premises, while others may experience long-term effects (Davis, 2001).

Certain health effects, such as those related to allergic reactions like irritation of the eyes, nose, and throat, dermatitis, exacerbation of asthma, and respiratory distress, have been proven to be associated with mold exposure. Other reported effects such as fever, flu-like symptoms, fatigue, respiratory dysfunction (including coughing up blood), excessive and regular nose bleeds, dizziness, headaches, diarrhea, vomiting, liver damage, and impaired or altered immune function have been identified in persons who have been exposed to mold via inhalation. (Davis, 2001, p.4)

These maintenance and design issues can have a serious negative effect on the learning environment for students and the working environment for teachers; it is a health hazard for all who spend significant amounts of time in the building. These effects: poor student behavior, lethargy, and apathy are some of the most consistently identified stressors for teachers (Abel & Sewell, 1999; BlasÈ, 1986; Dewe, 1986; Stenlund, 1995).

Beyond the direct effects that poor facilities have on studentsπ ability to learn, the combination of poor facilities, which create an uncomfortable and uninviting workplace for teachers, combined with frustrating behavior by students including poor concentration and hyperactivity, lethargy, or apathy, creates a stressful set of working conditions for teachers. Because stress and job dissatisfaction are common pre-cursors to lowered teacher enthusiasm and attrition (Friedman, 1995; Rosenholtz & Simpson, 1990; Shann, 1998), it is possible that the aforementioned characteristics of school facilities have an effect upon the shortage of teachers.

What is lacking in the body of research related to the effects of school facilities upon student achievement and the performance of teachers is analysis of key characteristics such as lighting, ventilation, acoustics and temperature control in relation to measures of both student performance and teacher satisfaction. According to Schneider (2002), most studies have focused on single environmental media, neglecting the critical issue of interaction effects between daylighting, air quality, noise, thermal comfort, or other factors (p. 4). It is possible that relationships exist between all three areas of the school environment: the quality of the school facility, behavior of students, and teacher satisfaction.

Certainly, more research is needed in this area. In fact, the federal government may act as a catalyst for such research. Section 5414 of the No Child Left Behind Act of 2001 calls for more research into the health and learning impacts of environmentally unhealthy public school buildings on students and teachers (U.S. Congress 2002).

Just as changes in the design of school buildings constructed during the energy crisis were driven by budget concerns created from rising energy costs, any future changes in school design trends are likely to be affected by the cost to taxpayers. Logic suggests the need for research into the specific effects of certain characteristics of school design for which tax monies will be spent before these changes will be realized.

There is considerable debate as to the relationship of funding to academic achievment. According to Schneider (2002), and Hanushek (1989), there is little correlation between capital expenditures and academic achievement. Conversely, Hedges, Laine, and Greenwald (1994), and Lockwood and McLean (1993), state that a correlation between spending and academic achievement does exist.

An analysis by Hanushek (1989) of 37 research articles on the direct effects of spending on achievement stated that ≥detailed research spanning two decades and observing performance in many educational settings provides strong and consistent evidence that expenditures are not systematically related to student achievement≤ (p. 49). However, Hedges, Laine, and Greenwald (1994) re-analyzed data from the same 37 articles and found that there was strong evidence to support a systematic positive relationship between resource input and school output.

Lockwood and McLean (1993) proposed that when the basic requirements of the educational process have been adequately funded, additional monies do improve the educational process. Their study concluded that once a base level of funding has been provided, the result of judicious spending on the instructional program should be evidenced in improved achievement (Lockwood & McLean, 1997). However, a study in Great Britian by Pricewaterhouse-Coopers (as cited in Schneider, 2002) analyzed the effects of capital investment on academic achievement, teacher motivation, school leadership, and other issues and found that relationships were weak. Stricherz (2000) noted that student achievement suffers in inadequate school buildings, but there is no hard evidence to prove that achievement rises when facilities improve beyond the norm.

Schneider (2002) summarized the debate, stating that existing studies on school building quality generally point to improved student behavior and better teaching in higher-quality facilities; however, ≥what is needed is more firm policy advice about the types of capital investments that would be most conducive to learning and to good teaching≤ (p. 9).

The lack of consensus is evidence of a need for further research of the specific effects of school building maintenance and design issues, not only on the student, but also the teacher and his or her job satisfaction, enthusiasm, and commitment to the profession. Should the study of these factors yield significant correlations to student achievement and overall levels of job satisfaction among teachers, it would provide justification to the allotment of monies for the renovation of existing facilities and the design of new facilities to include natural lighting, optimum acoustic and air quality in the classroom, and better temperature control, as well as proper maintenance.

Purpose of the Study

Previous studies have investigated the relationship of poor school building maintenance, including problems with ventilation, poor lighting, mold and mildew, and inconsistent temperatures in the classroom with student health problems, student behavior, and student achievement (Andrews & Neuroth, 1988; Crandell & Smaldino, 2000; Davis, 2001; Hathaway, 1995; Johnson, 2001; Lackney, 1999; Lyons, 2001; Moore, 2002; Sterling & Paquette, 1998; Stricherz, 2000; Tanner, 2000). Other studies have investigated the effects of student behavior on teacher satisfaction levels, having identified low job satisfaction as a precursor to burnout and the decision to leave the profession (Abel & Sewell, 1999; Blase, 1986; Borg & Riding, 1988; Brouwers & Tomic, 1999; Byrne, 1991; Coutanch, 1984); Dewe, 1986; Ingersoll, 200; Kyriacou & Sutcliffe, 1978; Stenlund, 1995).

The purpose of this study was to investigate the possible relationships of specific facility characteristics: light, acoustics, thermal control, and general maintenance, to student achievement, student behavior, and teacher satisfaction. In short, this study sought to investigate a very complicated and pervasive question. How much do design and maintenance issues in our schools affect the well-being and performance of teachers and students?

Research Hypothesis

The research hypothesis was: there is a positive correlation between building maintenance and design, job satisfaction levels among teachers, and student achievement on the Scholastic Aptitude Test (SAT), the (ACT) test developed by the American College Testing Program, Inc., and the Georgia High School Graduation Test (GHSGT).

Implications

Should this study yield a strong correlation between teacher satisfaction levels and the existence of poor climate and maintenance conditions in the school building, it would imply that these conditions affect not only the effectiveness with which teachers perform, and therefore the quality of instruction experienced by students, but also the rate of attrition among teachers. Whether teachers decide to leave the profession directly or indirectly because of the state of the building in which they work is likely to be hard to identify, but a strong correlation between low job satisfaction and a perception of poor facilities and student behaviors symptomatic of SBS would imply a relationship. It is entirely possible that the frustrations teachers experience from poor student behavior and achievement are related to the characteristics of the school facility.

Strong correlations would also imply that a greater devotion to building maintenance and climate control is needed in order to ensure optimum teacher and student performance. This researcher hoped to draw significant conclusions as to the validity of this argument by comparing levels of performance on the SAT, ACT, and the GHSGT of students with poor facilities as indicated on teacher questionnaires to levels of performance in schools where teachers describe the facility as adequate or excellent in this regard.

In the past, studies in this area have been limited to the affects of unhealthy buildings on student performance, and the degree to which student behavior affects teachersπ job satisfaction. The unique aspect of this study was that it considered these charachteristics both in relation to individual characteristics and in relation to a series of grouped characteristics.

Assumptions

1. The sample of teachers who volunteered to participate was representative of teachers in high schools in the Central and North Georgia area.

2. The SAT, ACT, and GHSGT are valid measures of student achievement.

3. Teacher questionnaires provide a valid measure of building conditions and student behavior.

4. The percentage of students receiving free or reduced lunch was a valid measure of socio-economic status (SES).

Limitations of the Study

The study was limited to the following factors:

1. The schools in the study were limited to public high schools.

2. The schools in the study were limited to Central and North Georgia.

3. The teachers who participated in the study were all volunteers.

4. Student behavior and characteristics of each building were rated solely by the perceptions of teachers.

5. Student achievement was rated solely by published scores on the SAT, ACT and GHSGT for the individual schools.

Definition of Terms

1. Sick Building Syndrome: Situations in which building occupants experience acute health and comfort effects that appear to be linked to time spent in a building.

2. HVAC: Heating, Ventilating, and Air Conditioning, related processes designed to control conditions within buildings for comfort or for industrial purposes.

3. Ventilation: A system that circulates fresh air throughout a building, replacing stagnant air or noxious fumes with clean air.

4. Full-spectrum lighting: The use of light fixtures to illuminate the rooms of a building, which simulate natural light from the sun.

5. Mold: A type of fungi which grows in damp, poorly lit environments and can cause allergic reactions, breathing difficulties, and other health problems.

6. SAT: Scholastic Aptitude Test, a test of math, science, and reading skills used to measure a studentπs potential to succeed in college.

7. GHSGT: The Georgia High School Graduation Test, a series of tests to measure student achievement levels in English Language Arts, Math, Science, Social Studies, and Writing skills among Georgia students.

8. ACT: A set of tests used nationally as a criteria for determining student achievement and eligibility for admission to college, which was developed by the American College Testing Program, Inc.

9. Acoustics: The way in which sound travels throughout a building or room, and how it is absorbed or reflected by surfaces such as walls or floors.

10. Thermal environment: The characteristics of the climate control system of a building and the resulting characteristics of air temperature.

11. Building maintenance: The overall cleanliness and working condition of equipment and systems of a building, the state of repair of a building and the equipment within the building.

12. Circadian Rhythm: Circadian rhythms cue daily behavior patterns even in the absence of external cues such as sunrise or sunset, evidence that such patterns depend on internal timers. However, when living things are deprived of normal cues, they display a characteristic ≥free-running≤ period of not quite 24 hours and drift slowly out of phase with the natural world. Light, particularly bright light, is believed to be the most powerful synchronizer of circadian rhythms.

13. High School: For this study, the term high school will refer to public schools including grades 9 through 12.

14. Mean Individual Characteristic Scores: The mean rating for each school for a particular characteristic averaged from teacher responses.

15. Composite Scores: The mean rating for each school for a group of characteristics averaged from teacher responses (eg. Student Behavior, Teacher Satisfaction, Physical Environment).

Organization

Chapter one of the study consists of a description of the problem, purpose of the study, research hypothesis, implications of the study, assumptions, limitations, definitions of terms used, and the procedures used. Chapter two consists of a review of the literature relating to building maintenance and design, air quality and ventilation, thermal conditions, and their effect on student achievement and health as well as their effect on teacher satisfaction and health. In addition, literature describing the effects of teacher enthusiasm and satisfaction on student achievement was considered, as well as the effect of teacher satisfaction on student achievement. Chapter three describes the methodology of the study, the criteria used to select the sample population, a description of the sample population, the instrument used in the study, and a description of how the data were collected and analyzed. Chapter four reports the findings of the study based on analysis of correlations between variables. Chapter five summarizes the findings, presents interpretations and implications, and presents recommendations to consider for future research.

International comparisons of reading literacy among 15-year-olds in 31 participating countries yielded similar results. ≥The socioeconomic status of studentsπ parents was positively associated with performance in reading literacy in the United States≤ (NCES, 2002, p. 56).

Associations between SES and student achievement among Georgia high school students are consistent with national statistics. For example, students who live and attend schools in affluent districts tend to perform better on standardized tests such as the SAT, ACT, and the GHSGT. Schools in the affluent East Cobb area north of Atlanta and the high-income regions of Gwinnett and Fulton Counties typically score in the top ten in Georgia on standardized tests. Although these schools all have over 2000 students, it is not uncommon for almost every student taking the GHSGT to pass on the first attempt, whereas the state average for the English Language Arts, Math, Social Studies and Science sections of the test is a combined 69% and a pass rate of 87% on the writing section of the test (Georgia Department of Education, 2003).

The 2001-2002 Georgia Public Education Report Card reports results consistent with this trend. The top ten schools in SAT total scores are primarily from affluent districts of Cobb, Fulton, and Gwinnett counties, seven of which have an average free and reduced lunch count of less than 7 percent. It is important to note that Woody Gap High/Elementary School was omitted from this comparison since it has only 100 students K-12, and only one student took the SAT in 2002.

The statistics are similar for each section (Math, Science, Social Studies, and English Language Arts) of the GHSGT. The highest performing schools typically have the lowest percentage of students eligible for free and reduced lunch, while the poorest performing schools in the state are typically found to have the highest percentage of students eligible for free and reduced lunch (Georgia Department of Education, 2003).

One theory as to why this correlation exists is that low socio-economic school districts with limited resources fail to keep school facilities in good condition. A recent study in North Carolina (Burton, 1999) indicates that districts which yield undesirable test scores also suffer from inadequate facilities. Burton points out that districts in North Carolina with limited funds have poorer school facilities.

Lacking adequate financial resources, districts are often unable to meet their capital needs. Delaying infrastructural repairs, improvements, or construction often leads to inadequate facilities. Eventually, it is widely believed, students pay the costs for this neglect. These costs manifest themselves through overcrowding; discomfort caused by poor ventilation, heating, and air conditioning; electrical wiring that does not lend itself to advanced technology; and building deterioration, that is, peeling paint, leaking roofs, and inoperable commodes. (Burton, 1999, para.10)

Interestingly, these outcomes do not follow racial lines in Burtonπs study. ≥Poor children, controlling for race, disproportionately attend schools with more deteriorated buildings than their better off peers≤ (Burton, 1999, para. 8). However, she pointed out that ≥black children, controlling for poverty, disproportionately attend schools with better buildings than Whites.≤ This provided evidence that the problem is one of socio-economic status, not race.

In a recent report by Lyons (2001) for The Council of Educational Facility Planners International (CEFPI) the author stated that there is a significant correlation between student achievement and the state of the school facility. ≥Four recent studies that evaluated the relationship between school buildings and student achievement found higher test scores for students learning in better buildings and lower scores for students learning in substandard buildings≤ (Lyons, 2001, para. 43). He stated that a difference of 5 to 17 percentile points existed in test scores, which was a stronger effect on student achievement than the combined effects of family background, socioeconomic status, school attendance, and behavior (Lyons, 2001).

Lyons (2001) names a variety of problems that distinguish a good school facility from a poor one, including age, lighting, ventilation, temperature, and noise. In fact, many of these buildings cannot meet the Americans with Disabilities Act accessibility requirements. More than 75% of our schools were built before 1970 ≠ three decades ago. By age 40, most buildings start deteriorating rapidly, even if all original equipment is replaced. Typical market forces suggest retiring our 42-year-old schools. But their service continues, perpetuating crowded classrooms, outmoded designs, poor communications systems, limited technology, and inadequate security (Lyons, 2001).

Air Quality and Temperature

According to Lyons (2001), ventilation and maintenance problems can trigger asthma, lethargy, an inability to concentrate, and drowsiness in students because allergens are not effectively removed from the atmosphere in the classroom, and high temperatures or inconsistent temperatures make students drowsy and sick or irritable. These problems are partially the result of building tighter buildings to counteract the loss of heat and to save energy during the 1970πs. It was common practice during the 1970πs to reduce ventilation rates from 15 cubic feet per person per minute to 5 cubic feet per person per minute. As a result, the Environmental Protection Agency (EPA) claims that indoor levels of pollutants may be two to five times higher than outdoor levels, and sometimes as high as 100 times as high (Lyons, 2001).

The EPA identifies some such conditions as Sick Building Syndrome (SBS). First employed in the 1970s, SBS describes a situation in which reported symptoms among a population of building occupants can be temporarily associated with their presence in that building (EPA, 1994).

Students and teachers with asthma or allergies suffer the most when exposed to mold and mildew, but even those with no apparent sensitivity to these conditions suffer from lethargy from the build up of carbon dioxide due to poor ventilation, and all suffer when the temperature is inconsistent between classes, or when classrooms are consistently too warm or too cold (Davis, 2001; EPA, 1994; EPA, 2000; Lyons, 2001).

Lackney (1999) asserts that these factors may affect not only the performance but also the overall health of children. ≥Children in åsick buildingsπ have been found to exhibit clear signs of sensory irritation, skin rashes, and mental fatigue ≠ all factors with the potential of decreasing the ability of students to perform≤ (Lackney, 2000, p. 27). ≥Poor indoor air quality has been linked to headaches, sore throats, sleepiness, lethargy, dizziness and asthma. Incidents of acute asthma attacks among children have doubled in the past 10 years and asthma is currently the number one reason American children are hospitalized≤ (Ostendorf, 2001, para. 3).

According to a separate report by the EPA, ≥Children do not perform as well when they are sick or absent from school. Indoor air quality problems can result in absences because of respiratory infections, allergic diseases from biological contaminants, or irritant reactions to chemicals used in virtually every part of the school. Some conditions in the school environment are closely associated with the incidence of sick building syndrome and asthma symptoms, and asthma-related illness is one of the leading causes of school absenteeism, accounting for over 10 million missed school days per year. In addition, persons with asthma or other sensitivities may have reduced performance in the presence of environmental factors that trigger their asthma. All of these building-related illnesses result from the lack of effective indoor environmental quality management.≤ (EPA, 2000, para. 6)

Lackney suggests that strategies for improving indoor air quality, such as increasing levels of fresh-air intake and increased ventilation rates in buildings, can help ensure that students can remain focused on the tasks of learning. According to Ostendorf (2001), prevention and problem solving usually necessitates managing pollution sources and using ventilation to control pollutants. A proactive approach costs less than resolving problems after they develop. It also saves money that would be better spent on educating students (Ostendorf, 2001).

Research by Andrews and Neuroth (1988) concurred, indicating the quality of air inside public school facilities may significantly affect studentsπ ability to concentrate. The evidence suggests that youth, especially those under ten years of age, are more vulnerable than adults to contaminants found in some schools such as asbestos, radon, and formaldehyde (Andrews & Neuroth, 1988).

Another significant health risk related to poor ventilation is the presence of mold spores in the atmosphere and on surfaces. Molds can cause a variety of health problems such as minor allergic reactions, exacerbation of asthma, and even brain damage. In the school setting, mold can grow on wood, paper, paint, fabric, carpet, or even glass and bare concrete, while the spores can travel throughout the school in the atmosphere. Davis stated that floods, leaking pipes, leaking windows, and leaking roofs are all potential sources of moisture which can lead to mold infestation. ≥Increased ambient humidity as a result of inadequate ventilation or improper drying of flooded areas can also lead to mold growth≤ (Davis, 2001, p. 2). According to her report, the increased air tightness of newly constructed buildings can allow moisture to become trapped in exterior walls, creating an environment conducive to mold growth. Also, centralized heating and air-conditioning systems can pick up contaminants and re-circulate them throughout the building thus potentially spreading the infestation.

Consequences of poor indoor air quality in schools include: increasing risk of long and short-term health problems in teachers and students; a negative impact on studentsπ ability to learn due to physical symptoms; reduced productivity of teachers; destruction of school equipment, including text books; and negative publicity for the school resulting in strained relationships among teachers, parents, and administrators. (Davis, 2001, p. 7)

Thermal comfort has been linked to academic achievement in several studies. Thermal conditions below optimum levels affect dexterity, while higher than optimal temperatures decrease general alertness and increase physiological stress (Lackney, 2000). McGuffey (1982) set the threshold of thermal comfort at 80 degrees F. Temperatures above 80 degrees F tend to produce harmful physiological effects that decrease work efficiency and output. (McGuffey, 1982).

According to Harner (1974), both math and reading skills are affected by temperature. He found a significant reduction in reading comprehension and reading speed occurred between 73.4 degrees F and 80.6 degrees F, while mathematical operations such as multiplication, addition and factoring were significantly reduced by air temperatures above 77 degrees F (Harner, 1974). A report by the Environmental Protection Agency agrees. It stated that temperature can affect the ability to perform everyday activities effectively.

In addition to indoor pollution and ventilation, studies confirm that various human activities such as typing or driving a vehicle are diminished when people are demonstrably too cold or too hot. Temperature is also implicated in studies of sick building syndrome. Maintaining temperature at the high end of the comfort zone tends to increase symptoms, while temperatures at the low end of the comfort zone tend to reduce symptoms. (EPA, 2000, para. 7)

The same report indicates that fluctuations in temperature need not be drastic to adversely affect student learning. ≥There is also good evidence that moderate changes in room temperature, even within the comfort zone, affect children's abilities to perform mental tasks requiring concentration, such as addition, multiplication, and sentence comprehension≤ (EPA, 2000, para. 8).

Acoustics

Good acoustics are a key to learning, but noise from the outdoors, mechanical noise, and noise generated from within the classroom because of the hard concrete block walls and concrete floor, make it difficult for students to learn. According to Lyons (2001), ≥students require a higher level of acoustic quality than adults, and to attain the good speech recognition necessary for optimal comprehension and learning, classrooms must limit background noise, carefully manage reverberation of sounds, and keep outdoor noise to a minimum≤ (para. 18). When acoustic quality in the classroom is poor, students may not be able to completely understand instructions from the teacher, causing frustration, and poor performance (Johnson, 2001).

Poor acoustics interferes with speech intelligibility, the ability of a student to hear and correctly interpret instruction of discussion. When a classroom sounds åechoey,π or when outside traffic or noise from the gym class next door interrupts a studentπs concentration, itπs likely that students will miss or misinterpret part of the teacherπs lesson. If this happens too often, a student may tune out because itπs too much of an effort to listen. As a result, learning suffers.

(Johnson, 2001, para. 4)

Johnson (2001) added that students with learning disabilities are at a greater risk of suffering the affects of poor acoustics in the classroom, but that teachers are also affected. ≥They may have to speak loudly to overcome background noise and may be less inclined to repeat information≤ (Johnson, 2001, para. 6).

The acoustic environment of a classroom is determined primarily by two factors: background noise and reverberation time. Background noise refers to any undesired auditory stimuli that interfere with what a child can hear in the classroom. Common sources of background noise include airplane noise, construction, automobiles, playgrounds, gymnasiums, cafeterias, busy hallways, and noise from inside the room from talking or movement (Crandell & Smaldino, 2000). Traffic noise in particular has been linked to deficits in mental concentration. Students make more errors on difficult tasks when traffic noise can be heard in the classroom, and a greater likelihood exists of giving up on tasks before the time allocated has expired (Lackney, 1999).

Background noise in a classroom can inhibit the childπs ability to perceive speech by masking the acoustic and linguistic cues available in the teacherπs spoken message. In general, the spectral energy of consonants is less intense than vowel energy; therefore, background noise in the classroom predominantly reduces consonant perception (Crandell et al., 2000). Reverberation refers to the persistence or prolongation of sound as sound waves reflect off of hard surfaces, such as walls or floors. Reverberant speech energy reaches the listener after the direct sound, and overlaps with that direct signal, resulting in a åsmearingπ or masking of speech. Like noise, reverberation tends to affect consonant perception adversely (Crandell et al.). The combined affects of background noise and reverberation are more serious than either factor alone. ≥The interaction of noise and reverberation adversely affects speech perception to a greater extent than the sum of both effects taken independently≤ (Crandell et al., p.365). In fact, their combined effects on speech perception can equate to a 40% to 50% reduction in speech perception (Crandell et al.).

The effects of classroom noise are not limited to students. Teachers also have been found to suffer ill effects from both background noise and reverberation (Ko, 1979). In a study of the effects of classroom noise on 1,200 teachers, results indicated that noise related to classroom activities and traffic or airplane noise were correlated with teacher fatigue, increased tension and discomfort, and an interference with teaching and speech recognition (Ko, 1979). In addition, Crandell (2000) reported that teachers have been found to exhibit a significantly higher incidence of vocal problems than do the general population. ≥It is reasonable to assume that these vocal difficulties are caused, at least in part, by having to increase vocal output to overcome the effects of classroom noise during the school day≤ (Crandell et al, p.365).

For teachers, sustained exposure to traffic noise in the workplace can lead to an increase in blood pressure. A review of a series of studies in the United States between 1980 and 1986 concluded that significant increases in blood pressure were associated with schools being near noisy urban streets (Evans, Kliewer & Martin, 1991).

These conditions can be controlled, however, with the use of acoustic tile on walls, carpeted floors, and proper location of schools away from airports, industry, or busy streets and highways whenever possible.

HVAC blowers and breakout noise, caused by air vibrating in metal ductwork, are common sources of background noise. A simple solution to both problems is to install acoustic liners inside the ductwork. Melamine foam is especially suitable for this. It resists fungus and microbial growth, and does not contribute to airborne contaminants. High-density vinyl barriers within walls can help stop noise from spilling into adjoining rooms. (Johnson, 2001, para. 10)

Classrooms with suspended ceilings can be fitted with vinyl barriers behind the ceiling panels to reduce noise, and acoustic tile can be installed on the surface of walls to reduce reverberation within the classroom as well as the use of carpeted floors.

Proper maintenance is necessary, however, when using these materials. Although carpet absorbs noise, it must be vacuumed regularly to avoid an increase in dust and mold, and wet carpet must be allowed to properly dry to avoid growth of mold. Acoustic tile on the walls is useful in lessening noise, but one coat of paint on this tile ruins its effectiveness. The location of the school away from busy streets is often impossible, but proper insulation for new schools and adding insulation to older buildings whenever possible is advised.

Lighting

Natural light has been found to profoundly influence the body and mind by affecting our circadian rhythm, according to Lyons (2001). ≥It can alter our mood and is a major source of Vitamin D, required for strong bones and healthy teeth≤ (Lyons, 2001, para. 21).

Heschong (1999) supported this claim in a study of 21,000 students in Colorado, California, and Washington state. Students exposed to maximum daylight were found to have learned much faster. A study by Hathaway (1995) indicated that both attendance and achievement were better in schools with full-spectrum light or full-spectrum with UV enhancement. Nicklas and Bailey compared test scores for over 1,200 students in three schools with natural lighting in North Carolina to scores in the county school system as a whole and other new schools within the county without natural lighting. The study showed that students who attended schools with natural lighting outperformed the students in schools without natural lighting by 5%-14% (1997).

A study by Plympton, Conway, and Epstein (2000) revealed that an increase in the use of natural light was not found to increase costs:

Schools found that increasing the amount of daylighting in school design did not

necessarily represent an increase in school construction and operation costs. Incorporating design components such as light sensors, and optimizing mechanical and electrical systems due to reduced cooling and lighting loads, can actually reduce the initial capital cost because of the reduced size and cost of HVAC equipment. Furthermore, the operations and maintenance costs are reduced due to a smaller electrical load and a smaller number of lighting fixtures to maintain. (Plympton, Conway, & Epstein, 2000, para. 6)

Lighting has been linked to student behavior as well as performance. Ott (1976) found that using full-spectrum fluorescent tubes which more closely replicate natural light than the traditional cool-white fluorescent tubes can show dramatic improvement in some childrenπs behavior in the classroom. Students in standard lighting were observed fidgeting, leaping from their seats, flailing their arms, and paying little attention to their teachers. Students in the full-spectrum lit classrooms settled down more quickly and paid more attention to their teachers (Ott, 1976).

General Maintenance

Some studies indicate that general maintenance affects student achievement by fostering the conditions that inhibit studentsπ ability to perform well. Frazier (2002) stated that deferred maintenance can create an environment of peeling paint, crumbling plaster, nonfunctioning toilets, poor lighting, inadequate ventilation, and inoperative heating and cooling systems, which affects both the health and the morale of staff and students.

The Carnegie Foundation for the Advancement of Teaching found that in those schools which are under funded, morale is low, facilities are decaying, and the dropout rate remains high year after year (1988). This hypothesis was tested by the Washington D.C. school system. While controlling for socioeconomic status and other factors, Edwards (1991) found that as the schoolπs condition improved from one category to the next, for example, from poor to fair, studentπs standardized achievement scores increased an average of 5.45 points. With an improvement from poor to excellent, an increase of 10.9 percentage points was noted.

A recent study of New York City schools revealed that 40% of the 39 schools studied had filthy bathrooms with no soap nor toilet paper. Thirty-three percent had poor ventilation, and 24% had dirty cafeterias. Forty percent reported garbage lying around the school, and 30% had inadequate lighting. The author commented: ≥It is remarkable that at a time when children are being held to higher standards, there are few standards to protect their health from hazards at school, and that existing laws created to protect adult health and safety are being ignored≤ (Neglected Buildings, 1999, p. 2).

In addition to the need for adequate maintenance, some researchers propose the upgrade of school health programs to counteract threats to student health from inside and outside the school. A study by Symons, Cinelli, James, & Groff (1997), identified the health of the student as a primary determiner of the studentπs achievement. It stated that adolescents manifest difficulty learning when they are not in good health and suggested comprehensive health care services for students at school to lessen the effects of poor health. The authors admitted, however, that obstacles such as lack of administrative and governmental support as well as a lack of funding make this unlikely (Symons, Cinelli, James, & Groff, 1997).

Moore (2002) described four types of maintenance necessary in the school setting: emergency, routine, preventive, and predictive. She described emergency maintenance as that which takes place when a system fails; routine maintenance as corrective in nature and scheduled in advance, such as repairing systems or replacing parts; and preventive maintenance as proactive in nature such as lubrication or cleaning. She asserted the need for a relatively new concept in maintenance called predictive maintenance to counteract poor conditions. She identified predictive maintenance as forecasting failures in plumbing, electrical, and ventilation systems and replacing them before they fail. The primary cause for the lack of such maintenance procedures is lack of funding.

≥Reasons for the poor maintenance of our school facilities range from years of under-funding, to the sheer volume of tasks to be completed, to the natural preference to fund growth and development rather than maintenance≤ (Moore, 2002, para. 4). In addition, Moore identified problems with the quality of workmanship among school maintenance staff as part of the problem. A maintenance staff that is poorly trained or unmotivated can cause more harm than good.

Shideler (2001) also asserted the need for improved employee training in maintenance procedures. ≥Training employees in cleaning for health and safety empowers them to help produce cleaner, healthier facilities at less cost, enhances professionalism of a custodial department, raises morale and creates safer working conditions≤ (Shideler, 200, para. 1).

Discussion

The general body of research regarding the effects of SBS and related issues such as poor ventilation, lighting, acoustics, and cleanliness agrees that there is a significant relationship between these issues and student health and achievement in school (Andrews & Neuroth , 1988; Burton, 1999; Crandell & Smaldino, 2000; Davis, 2001; Edwards, 1991; Evans, Kliewer & Martin, 1991; Frazier, 2002; Harner, 1974; Heschong, 1999; EPA, 2000; Johnson, 2001; Ko, 1979; Lackney, 1999; Lackney, 2000; Lyons, 2001; McGuffey, 1982; ≥Neglected Buildings,≤ 1999; Nicklas & Bailey, 1997; Ostendorf, 2001; Ott, 1976; Plympton, Conway & Epstein, 2000; Symons, Cinelli, James & Groff, 1997). Teachers also suffer when the school facility is inadequate in this regard (Crandell & Smaldino, 2000; Evans, Kliewer & Martin, 1991; Johnson, 2001; Ko, 1979; Ostendorf, 2001). Researchers and industry professionals promote an increased emphasis on designing schools with better ventilation systems that move more air than is presently accepted, using full-spectrum lighting and/or natural light whenever possible, and the use of acoustic tile and carpet to lower noise levels in the classroom. Others promote improved training of custodial staff and higher standards of cleanliness.

Teacher Satisfaction

There are many studies investigating the different determiners of job satisfaction among teachers. Some of the variables identified are teacher autonomy, administrative support, relationships with parents, and feelings of efficacy, stress, and student behavior. Many of these issues were beyond the scope of this study, but the effect of the building itself on job satisfaction levels among teachers, whether directly or indirectly through its relationship to student behavior was relevant to this investigation. What is implied here is that poor conditions due to maintenance and design issues are adding to an already serious shortage of teachers and adversely affecting the commitment and enthusiasm they have for their profession, therefore affecting student achievement.

In a study by Richard M. Ingersoll (2001), the author analyzed data gathered by the National Center for Education Statistics through its Schools and Staffing Survey (SASS) and its supplement, the Teacher Follow-up Survey (TFS) involving 6,733 teachers. ≥The data show that, in particular, low salaries, inadequate support from school administration, student discipline problems, and limited faculty input into school decision-making, all contribute to higher rates of turnover, after controlling for characteristics of both teachers and schools≤ (Ingersoll, 2001).

The data in this study show that reasons for teachers leaving the profession are varied, but of particular interest to the facilities management issue are Ingersollπs findings regarding student discipline and motivation. He reports: ≥For example, on a four unit scale, a one unit increase in reported student discipline problems in schools is associated with a 23 percent increase in the odds of a teacher departing≤ (Ingersoll, 2001).

Of the teachers polled, 18 percent cited student discipline problems as a reason to move to another school, and 30 percent cited student discipline problems as a reason to leave teaching altogether. In addition, 10 percent cited lack of student motivation as a reason to move to another school, while 38 percent cited lack of student motivation as a reason to leave the profession. Perhaps the most striking statistics are gathered from urban, high poverty schools. Among these schools, 29 percent cited student discipline problems as a reason to move to another school, and 27 percent cited student discipline problems as a reason to leave teaching, while 27 percent cited lack of student motivation as a reason to move to another school, and 50 percent cited lack of student motivation as a reason to leave the profession (Ingersoll, 2001). In light of the statistics, it seems no small coincidence that urban, poverty stricken schools tend to have poorer facilities overall. The findings of these studies provide evidence indicating that student discipline and motivation are two significant factors in determining the level of job satisfaction among teachers.

Much of the research available on the mental well being of teachers uses the term morale; however, a definition of morale is hard to pinpoint. Mendel (1987) described morale as a feeling, state of mind, mental attitude, or an emotional attitude. Washington and Watson (1976) defined morale as the feeling a worker has about his or her job in relation to where they are in the organization and the extent to which the organization meets their own needs and expectations. Bentley and Rempel (1980) called morale the professional interest and enthusiasm that a person displays towards the achievement of individual and group goals in a given job situation. Hoy and Miskel (1987) stated that morale is high when teachers feel a sense of accomplishment from their jobs. Although researchers do not agree on a specific definition of morale, evidence indicates that certain aspects of low morale lead to teacher attrition and poor performance in the classroom. Because the term morale is difficult to define and even harder to measure, this study focused specifically on the level of job satisfaction among teachers.

A study of the Texas public school system revealed that 44 percent of respondents (teachers) were seriously considering leaving the profession (Henderson & Henderson, 1996). Estes, Stansbury, and Long (1990) examined attrition rates, or ≥burnout,≤ of teachers in the State of California and found that more than 50% of all newly hired teachers leave the profession in that state within 5 years. In a similar study, Colbert and Wolff (1992) discovered an identical rate of attrition.

Apparently the stresses of the workplace lead to lower job satisfaction and eventually the decision to choose another career usually within five years. Although each individual is affected by different aspects of the profession, researchers have tried to pinpoint the most common of these. Dewe identified inadequate resources, and feelings of inadequacy or negative attitudes as sources of stress (1986). Pupil misbehavior and poor working conditions are two sources of stress among teachers identified by Borg and Riding (1981).

In a study by Byrne involving over 3000 subjects, the author identified student behaviors as primary sources of teacher stress. She cited no less than 16 articles of research in support of her claim. ≥In particular, student discipline problems, student apathy, low student achievement, and verbal and physical abuse by students have been shown to be primary sources of teacher stress≤ (Byrne, 1991a, p. 649).

The Environmental Protection Agency lists indoor air quality as one of the determiners of teacher morale. According to Ostendorf (2001), ≥Good indoor air quality can increase productivity, morale and a sense of comfort for teachers, administrators and all school occupants≤ (para. 3).

Effects of Low Satisfaction

As individuals, teachers are affected in different ways by stress. Dewe (1986) concluded that the psychological and emotional effects of stress include general uneasiness, depression, nervousness, anxiety, and a loss of confidence. Behavioral effects include procrastination, impatience with others, low productivity, absenteeism, and withdrawal from teaching (Dewe, 1986).

This researcher theorizes that the cycle of poor student attitudes and performance, poor conditions in the workplace, and low levels of satisfaction among teachers, repeats itself when teachers become so frustrated that their patience with low-achieving students and students with poor attitudes and behavior begins to wane, perpetuating the problem. Overall, educators who fall victim to burnout are likely to be less sympathetic toward students, have a lower tolerance for classroom disruption, be less apt to prepare adequately for class, and feel less committed and dedicated to their work which ultimately leads to increased absenteeism and impetus to leave the profession (Farber & Miller, 1981).

The need to nurture high levels of satisfaction among teachers becomes apparent in light of studies relating to the effects of low teacher enthusiasm. According to Patrick, Hisley, and Kempler (2000), teacher enthusiasm leads to greater student achievement. In an analysis of two studies in this area they concluded: ≥The studies described herein provide strong, consistent evidence, from both the laboratory and the classroom, to suggest that when a teacher exhibits greater evidence of enthusiasm, students are more likely to be interested, energetic, curious, and excited about learning≤ (Patrick, Hisley, & Kempler, 2000, p. 233).

Stress can affect teachersπ job satisfaction and their effectiveness with pupils (BlasÈ, 1986). Stress can also result in mental and physical illness and impair the working relationship between teachers and students as well as the overall quality of teaching (Kyriacou, 1987). Prolonged stress can result in burnout. The consequences of burnout include diminished job satisfaction, reduced teacher-pupil rapport and pupil motivation, and decreased teacher effectiveness in meeting educational goals (Kyriacou & Sutcliffe, 1978). According to Shann (1998), teacher satisfaction influences job performance, attrition, and ultimately, student performance. Teachers who are satisfied with their jobs indicate that the student-teacher relationships are most important. A study by Stenlund (1995) involving teachers from the U.S. and six other nations supported the importance of student-teacher relationships. Teachers questioned clearly identified students as the primary determiner of both their professional enthusiasm and discouragement. Teachers indicated that they almost universally treasure student responsiveness and enthusiasm as a vital factor in their own enthusiasm, and conversely listed low motivation in students as a discourager (Stenlund, 1995).

Burnout

When faced with overwhelming stress and the feeling that what they are doing is no longer useful or effective, teachers often reach what many researchers have labeled ≥burnout≤ (Guglielmi & Tatrow, 1998; Ingersoll, 2002; Ingersoll, 2001; Blase, 1986; Borg, & Riding, 1991). ≥Burnout is a work-related syndrome that stems from an individualπs perception of a significant discrepancy between effort (input) and reward (output)≤ (Friedman, 1995, p. 281). It occurs when teachers perceive they are unable to effectively fulfill the requirements of their profession. Friedman described burnout as a process, not an event, and is the result of unmediated stress over time.

Brouwers and Tomic (1999) identified disruptive student behavior as one of the most prevalent precursors to teacher burnout. ≥When teachers have little confidence in their ability to maintain classroom order, they will likely give up easily in the face of continuous disruptive student behavior. As a consequence, they feel themselves ineffective in their attempts to maintain classroom order≤ (Brouwers & Tomic, 1999, p.249). A report from the National Center for Education Statistics (1998) supported the importance of student behavior in relation to teacher job satisfaction, along with other factors such as administrative support, teacher autonomy, and parental support.

According to Coutanch (1984), job dissatisfaction leads to high rates of teacher absenteeism and turnover as well as increased student apathy, negativism, and misbehavior. Jenkins and Calhoun (1991) concurred, stating that stress is a major factor in teachersπ decision to leave teaching. They prescribed staff development for teachers in stress management techniques and in controlling the circumstances that cause stress. According to Byrne (1991b) teachers suffer serious emotional consequences as a result of burnout, which eventually effect student achievement.

Teachers are purported to exhibit signs of emotional exhaustion when they perceive themselves as unable to give of themselves to students, as they did earlier in their careers; depersonalization, when teachers develop negative, cynical, and sometimes callous attitudes towards students, parents, and colleagues; and feelings of reduced personal accomplishment, when they perceive themselves as ineffective in helping students to learn, and in fulfilling other school responsibilities. (Byrne, 1991b, p. 198)

In addition to the adverse effects that stress and low job satisfaction have on the teachers themselves, the loss of good teachers adversely affects the quality of learning that students receive. Ingersoll (2001) stated that the decision to leave by teachers who do not share the goals and mission of the school is not a negative occurrence. It gives the administration an opportunity to replace them with teachers who do share in the mission, but he described a threshold at which the organization begins to lose experienced teachers who are beneficial to the level of learning available to students (Ingersoll, 2001). ≥After reaching a certain threshold level, however, turnover may become a source of group disintegration, rather than group integration. At such a point, the negative consequences of turnover for organization stability and coherence would begin to overshadow the positive consequences for the organization resulting from the elimination of dissension≤ (Ingersoll, 2001). He suggested that turnover rates of more than 25 percent are likely to have a negative impact on organizational performance.

Summary

Professionals in the field of school building design, educational researchers, and health care professionals have published an extensive body of literature based on the study of these interrelated issues. This literature yields evidence to support the hypothesis that student achievement, job satisfaction levels among teachers, and the health of both students and teachers can be affected significantly by lighting, acoustics, ventilation, and maintenance (Andrews & Neuroth , 1988; Burton, 1999; Crandell & Smaldino, 2000; Davis, 2001; Edwards, 1991; Evans, Kliewer & Martin, 1991; Frazier, 2002; Harner, 1974; Heschong, 1999; EPA, 2000; Johnson, 2001; Ko, 1979; Lackney, 1999; Lackney, 2000; Lyons, 2001; McGuffey, 1982; ≥Neglected Buildings,≤ 1999; Nicklas & Bailey, 1997; Ostendorf, 2001; Ott, 1976; Plympton, Conway & Epstein, 2000; Symons, Cinelli, James & Groff, 1997).

The literature analyzed states that full-spectrum lighting and natural light positively effects student achievement, and that it can positively affect the mood of those exposed to it, while fluorescent light can be detrimental to students, especially those with Attention Deficit Disorder (ADD) or Attention Deficit-Hyperactivity Disorder (ADHD). The lack of natural light or full-spectrum light can lead to irritability, restlessness, and the inability to concentrate (Hathaway, 1995; Heschong, 1999; Lyons, 2001; Nicklas & Bailey, 1997; Ott, 1976).

Poor acoustics has been indicated as a detriment to learning in that it makes it difficult for students to differentiate between vocalized sounds. Poor acoustics in the classroom can also magnify street noise or construction noise, making it hard to concentrate, and making it difficult and frustrating to teach above the noise (Crandell & Smaldino, 2000; Evans, Kliewer & Martin, 1991; Johnson, 2001; Lackney, 1999; Lyons, 2001).

Poor ventilation has been found to cause an increase in levels of carbon dioxide in the building, which can make students and teachers lethargic. It can also harbor dangerous toxins in the air such as chemical fumes, asbestos, or mold spores. These factors lead to poor student achievement and apparent empathy as a result of poor health or lethargy. This effect can be compounded, as teachers perceive their efforts in the classroom to be futile, and eventually lose enthusiasm for their task (Andrews & Neuroth, 1988; Davis, 2001; EPA, 1994; EPA, 2000; Lyons, 2001; Ostendorf, 2001).

And finally, studies have shown that poor maintenance can lead to poor health conditions. Filthy bathrooms, dusty and dirty classroom carpets and floors, and inconsistent temperatures with poorly serviced HVAC systems lead to a variety of health problems which lead to poor student achievement and poor health among teachers (Shideler, 2001; Symons, Cinelli, James & Groff, 1997).

disagree;≤ 10 indicating they ≥strongly agree≤). In some of the questions, 1 reflected a negative characteristic and 10 reflected a positive characteristic, while others reflected the opposite perception. All negatively oriented questions were re-coded to reflect a positive orientation, so that all responses on the final data set reflected 1 as negative and 10 as positive in order to make statistical analysis accurate.

Using the percentage of free and reduced lunch participants at each school, the relationship of socio-economic status to test scores was analyzed. Linear correlations for Pearsonπs product moment coefficient were performed for evidence of a relationship between test scores and free lunch. The r ²value for each was analyzed to determine the amount of shared variance between factors.

Ratings from teachers were averaged to determine a mean score per school for each characteristic. The following characteristics of the facility were measured: cleanliness, working order of the schoolπs equipment, temperature of the classroom, ventilation, windows, the presence of fumes in the classroom, noise levels, the presence of acoustic tile and carpet. The following student behaviors were also measured: lethargy, interest in school, illness, absenteeism, and behavior detrimental to the learning process. Teacher characteristics measured included lethargy, frustration, the number of sick days taken, and job satisfaction. All of the measures tested on the questionnaire were determined solely by teacher perceptions.

Also, a mean score was calculated per school for each of three groups of characteristics: Physical Environment, Student Behavior, and Teacher Satisfaction. Linear regressions were performed utilizing both the means for individual characteristics and the mean group scores in relation to mean test scores on the GHSGT, ACT, and SAT. Control variables used to determine adjusted variables included free and reduced lunch percentages (socio-economic status), teacher experience levels, and teacher education levels.

Linear regressions were also performed to determine the relationship of adjusted test scores to mean student behavior ratings and mean teacher satisfaction ratings for each school. Finally, specific physical characteristics of the school relating to student and teacher health issues were analyzed for their relationship to student lethargy and motivation, student illness, teacher lethargy, and sick days taken by teachers. Health-related characteristics included cleanliness, temperature, ventilation, windows, the presence of fumes, and the presence of carpet.

Reliability

The reliability of the questionnaire was analyzed using Cronbachπs Alpha, which is considered valid for determining the internal consistency of the questionnaire (see Appendix B). Cronbachπs Alpha is a correlation between the test and all other possible tests containing the same number of items constructed from a hypothetical universe of items that measure the characteristic of interest (Huck, 2000). Table 1 reveals the three parts of the questionnaire and the alpha per section.

Table 1

Reliability of the Instrument *

Questionnaire Items Section Construct Standardized Alpha

1 to 13 Physical Characteristics .7023

14 to 23 Student Behavior .8752

24 to 27 Teacher Satisfaction .6818

* See Appendix A for questionnaire items.

Control Variables

Data regarding the percentage of students receiving free or reduced lunch were recorded from the Georgia Department of Education website (Georgia Department of Education, 2003). Teacher education and experience levels were reported by teachers on the questionnaire. The purpose of including the percentage of free lunch in the analysis was to serve as an indicator of socio-economic status per school. The school was the unit of analysis. SAT scores, ACT scores, and percent passing the GHSGT were also recorded from the Georgia Department of Education website. These data were analyzed using SPSS 11.0 for significant statistical relationships (see Appendix C).

The first aspect of analysis required applying a multiple regression to adjust for socioeconomic status, teacher education, and teacher experience. This yielded 13 adjusted variables (see descriptive statistics in Table 2).

Table 2

Grand Means For Adjusted Variables

Descriptive Statistics

	Mean	Std. Deviation	N
GT WRITING
88.33	5.80	27
GT SCIENCE
74.78	11.38	27
GT SOCIAL STUDIES
83.41	9.78	27
GT MATH
92.11	5.22	27
GT ENGLISH
95.96	2.97	27
ACT SCIENCE
19.833	1.539	27
ACT READING
20.111	2.091	27
ACT MATH
19.659	1.615	27
ACT ENGLISH
19.100	2.049	27
ACT COMPOSITE
19.796	1.758	27
SAT MATH
495.37	33.80	27
SAT VERBAL
493.63	29.59	27
SAT TOTAL
989.00	62.57	27

Table 3

Multivariate Tests

Effect		Value	F	Hyp. df	Error df	Sig.	Eta Sq
Intercept
Pillai's Trace	.987	73.540	12.000	12.000	.000	.987

Wilks' Lambda	.013	73.540	12.000	12.000	.000	.987

Hotelling's Trace	73.540	73.540	12.000	12.000	.000	.987

Roy's Largest Root	73.540	73.540	12.000	12.000	.000	.987
YRS. EXP
Pillai's Trace	.715	2.509	12.000	12.000	.062	.715

Wilks' Lambda	.285	2.509	12.000	12.000	.062	.715

Hotelling's Trace	2.509	2.509	12.000	12.000	.062	.715

Roy's Largest Root	2.509	2.509	12.000	12.000	.062	.715
YRS.COL
Pillai's Trace	.608	1.549	12.000	12.000	.230	.608

Wilks' Lambda	.392	1.549	12.000	12.000	.230	.608

Hotelling's Trace	1.549	1.549	12.000	12.000	.230	.608

Roy's Largest Root	1.549	1.549	12.000	12.000	.230	.608
FREE LUNCH
Pillai's Trace	.939	15.471	12.000	12.000	.000	.939

Wilks' Lambda	.061	15.471	12.000	12.000	.000	.939

Hotelling's Trace	15.471	15.471	12.000	12.000	.000	.939

Roy's Largest Root	15.471	15.471	12.000	12.000	.000	.939

a Exact statistic

b Design: Intercept+YRS.EXP+YRS.COL+FREE.LUN

The analysis of the data to adjust the test scores is found in Table 3, where the significant levels range from 0.00 to 0.23. The measure of association, eta, is appropriate for the dependent variables (test scores) measured on an interval scale and the independent variables (years of experience, years of college, and free lunch). Eta is asymmetric and does not assume a linear relationship between the variables. Eta squared can be interpreted as the proportion of variance in the dependent variable explained by differences among groups. Appendix D reveals the tests between-subjects effects, where the eta squared ranges from .000 to .964 for the corrected model.

Physical Environment

Thirteen characteristics of the physical environment of the school were measured utilizing the opinions of teachers: cleanliness of the school, condition of the classroom equipment, temperature of the classroom (both warm and cold), quality of ventilation, presence of windows, cleanliness of bathrooms, presence of fumes in the classroom, noise, traffic noise, presence of acoustic tile, presence of carpet, and cleanliness of the cafeteria. The responses from each school were averaged to determine a mean score for each characteristic per school. A composite score for the total quality of each facility was determined by averaging the means for each of the individual characteristics. Both the individual characteristics of each school and the composite score for the total quality of the physical environment of each school were analyzed to determine the degree to which relationships existed between the physical environment and student achievement as measured on the SAT, ACT, and GHSGT.

Analysis of the individual school characteristics revealed no significant positive relationships with adjusted test scores on the SAT. In fact, the only significant relationships revealed were negative correlations between the presence of windows and the SAT Total (r = -.436, p = .023), the SAT Math (r = -.427, p = .026), and the SAT Verbal (r = -.440, p = .022); as well as coldness of the classroom and SAT Total (r = -.405, p = .036), SAT Math (r = -.419, p = .030), and SAT Verbal (r = -.383, p = .049). This indicated that among our sample of schools, colder schools and schools with fewer windows were more successful on the SAT, controlling for SES, teacher experience, and teacher education levels. The other eleven measures of facility characteristics revealed no significant correlation with the SAT Total (see Table 4). Analysis of the composite physical environment scores revealed no significant correlation to SAT Total.

Table 4

Correlations: Adjusted SAT, Physical Environment

		Adj.MAT	Adj.VERB	Adj.TOT
CLEAN
Pearson Correlation	-.025	.008	-.010

Sig. (2-tailed)	.901	.970	.960

N	27	27	27
EQUIP
Pearson Correlation	.194	.166	.182

Sig. (2-tailed)	.333	.408	.363

N	27	27	27
WARM
Pearson Correlation	-.285	-.249	-.271

Sig. (2-tailed)	.149	.210	.172

N	27	27	27
COLD
Pearson Correlation	-.419	-.383	-.405

Sig. (2-tailed)	.030	.049	.036

N	27	27	27

(table continued)

Table 4 continued

		Adj.MAT	Adj.VERB	Adj.TOT

VENTILAT	Pearson Correlation	-.252	-.265	-.260

Sig. (2-tailed)	.205	.182	.191

N	27	27	27
WINDOW
Pearson Correlation	-.427	-.440	-.436

Sig. (2-tailed)	.026	.022	.023

N	27	27	27
BATHROOM
Pearson Correlation	.014	.068	.039

Sig. (2-tailed)	.946	.735	.847

N	27	27	27
CAFE
Pearson Correlation	.014	.048	.030

Sig. (2-tailed)	.947	.812	.883

N	27	27	27
FUMES
Pearson Correlation	-.025	-.057	-.040

Sig. (2-tailed)	.901	.777	.842

N	27	27	27
NOISE
Pearson Correlation	-.308	-.345	-.327

Sig. (2-tailed)	.118	.078	.096

N	27	27	27
TRAFFIC
Pearson Correlation	-.086	-.079	-.083

Sig. (2-tailed)	.669	.695	.679

N	27	27	27
AC.TILE
Pearson Correlation	.140	.128	.136

Sig. (2-tailed)	.485	.524	.500

N	27	27	27
CARPET
Pearson Correlation	.241	.276	.258

Sig. (2-tailed)	.227	.164	.193

N	27	27	27
MEANPE
Pearson Correlation	-.235	-.212	-.226

Sig. (2-tailed)	.238	.288	.257

N	27	27	27

Analysis of individual school characteristics with adjusted ACT scores revealed no positive correlations. The only significant relationships revealed were negative correlations between the presence of windows and the ACT Composite (r = -.412, p = 033), the ACT English (r = -.424, p = .028), the ACT Math (r = -.417, p = .031), the ACT Reading (r = -.396, p = .041), and the ACT Science test (r = -.400, p = .039). No relationship was detected between the composite physical environment scores and the five sections of the ACT (see Table 5).

Table 5

Correlations: Adjusted ACT, Physical Environment

		Pred COMP	Pred.ENG	Pred.MAT	PredREAD	Pred.SCI
CLEAN
Pearsonπs r	.043	.071	.011	.026	.048

Sig. (2-tailed)	.830	.726	.958	.899	.814

N	27	27	27	27	27
EQUIP
Pearsonπs r	.103	.119	.115	.083	.087

Sig. (2-tailed)	.610	.553	.568	.680	.665

N	27	27	27	27	27
WARM
Pearsonπs r	-.193	-.146	-.248	-.220	-.183

Sig. (2-tailed)	.336	.467	.211	.271	.362

N	27	27	27	27	27
COLD
Pearsonπs r	-.296	-.291	-.334	-.288	-.276

Sig. (2-tailed)	.133	.141	.089	.145	.164

N	27	27	27	27	27
VENTILAT
Pearsonπs r	-.273	-.239	-.297	-.296	-.274

Sig. (2-tailed)	.168	.230	.133	.134	.167

N	27	27	27	27	27
WINDOW
Pearsonπs r	-.412	-.424	-.417	-.396	-.400

Sig. (2-tailed)	.033	.028	.031	.041	.039

N	27	27	27	27	27
BATHROOM
Pearsonπs r	.128	.159	.084	.108	.136

Sig. (2-tailed)	.525	.427	.677	.590	.500

N	27	27	27	27	27
CAFE
Pearsonπs r	.098	.089	.086	.108	.107

Sig. (2-tailed)	.628	.660	.668	.593	.596

N	27	27	27	27	27
FUMES
Pearsonπs r	-.116	-.074	-.125	-.151	-.130

Sig. (2-tailed)	.565	.713	.533	.451	.519

N	27	27	27	27	27

(table continued)

Table 5 continued

NOISE

Pearsonπs r

Pred COMP

-.366

Pred.ENG

-.367

Pred.MAT

-.359

PredREAD

-.362

Pred.SCI

-.365

Sig. (2-tailed)

.061

.059

.066

.064

.061

TRAFFIC

Pearsonπs r

-.067

-.054

-.081

-.075

-.065

Sig. (2-tailed)

.740

.790

.687

.711

.748

AC.TILE

Pearsonπs r

.101

.095

.116

.102

.095

Sig. (2-tailed)

.617

.639

.564

.614

.639

CARPET

Pearsonπs r

.292

.314

.273

.274

.290

Sig. (2-tailed)

.139

.111

.169

.166

.142

MEANPE

Pearsonπs r

-.174

-.140

-.215

-.195

-.167

Sig. (2-tailed)

.384

.487

.282

.331

.404

Analysis of individual school characteristics and composite physical environment scores with adjusted GHSGT scores revealed no positive correlations between adjusted test scores and the physical environment. The only significant correlations revealed were negative correlations between the presence of windows with the GHSGT Writing test (r = -.411, p = 033), Science test (r = - .417, p = .031), Social Studies test (r = -.420, p = .029), Math test (r = -.408, p = .035), and English test (r = -.395, p = .042). In addition, negative correlations were revealed between the coldness of the classroom and the GHSGT Social Studies test (r = -.407, p = .035), Math test (r = -.404, p = .037), and English test (r = -.397, p = .040). This indicates that schools from our sample with colder classrooms and fewer windows tended to score higher on the GHSGT (see Table 6).

Table 6

Correlations: Adjusted Georgia High School Graduation Test, Physical Environment

		Pred.ENG	Pred.MAT	Pred.SOC	Pred.SCI	Pred.WRI
CLEAN
Pearsonπs r	-.068	-.054	-.037	.012	-.014

Sig. (2-tailed)	.738	.789	.854	.954	.945

N	27	27	27	27	27
EQUIP
Pearsonπs r	.135	.151	.165	.115	.116

Sig. (2-tailed)	.502	.451	.411	.569	.566

N	27	27	27	27	27
WARM
Pearsonπs r	-.361	-.340	-.314	-.247	-.287

Sig. (2-tailed)	.064	.083	.111	.215	.147

N	27	27	27	27	27
COLD
Pearsonπs r	-.397	-.404	-.407	-.333	-.351

Sig. (2-tailed)	.040	.037	.035	.090	.072

N	27	27	27	27	27
VENTILAT
Pearsonπs r	-.329	-.310	-.290	-.296	-.314

Sig. (2-tailed)	.094	.116	.142	.134	.110

N	27	27	27	27	27
WINDOW
Pearsonπs r	-.395	-.408	-.420	-.417	-.411

Sig. (2-tailed)	.042	.035	.029	.031	.033

N	27	27	27	27	27
BATHROOM
Pearsonπs r	-.027	-.012	.007

Discussion

Teacher Satisfaction

Population

Sample

Data Collection

Reliability