![[Table of Contents]](/icons/up.gif){kind=icon}
Archived Information
A National Study of Charter Schools - July 1998
In addition to the findings discussed above, the survey data show that almost all the children in a significant number of charter schools are minorities, economically disadvantaged, or students with disabilities. We estimate that approximately one-fifth of charter schools may serve such a particular student population. At least 32 charter schools serve more than two-thirds African-American students, 13 serve more than two-thirds Native American children, 22 have more than two-thirds Hispanic students, and eight serve more than 50 percent students with disabilities.
Such concentrations are not accidental. Many charter schools have been founded specifically to meet the needs of a particular population of children. In the telephone survey, we asked charter schools to tell us why their school was founded, and then we asked them to tell us their most important reason. The second most cited reason was to serve a special population of students. Sixty charter schools (17 percent) stated this was their most important motivation for starting a charter school.
Information that we gathered from the 91 sites we visited provides a deeper understanding of schools that serve special populations.[18] About one-third of the fieldwork schools serve almost exclusively either a particular racial/ ethnic group, students with disabilities, or at-risk students and dropouts.[19] Many of these schools, but not all, target particular populations and have designed programs that charter developers believe will address the challenges posed by working with that group of students. The following sections provide examples from our fieldwork of such schools.
Some charter schools serve one particular racial/ethnic group because they are in a district that primarily serves that group (that is, they are not racially distinct from their districts). This situation is certainly true for the majority of predominantly White schools that are located in predominantly White districts. It also holds for other racial groups.
Predominantly African-American Schools. Of the 32 charter schools that have at least two-thirds African-American students, one-quarter are in districts that have an average of at least two-thirds non-White students. The following school from our fieldwork sample illustrates a case of a charter school that is 100 percent African American and is in a district where the student population is more than 80 percent non-White:
The pre-kindergarten through 12th grade school, located in a mixed-income residential neighborhood in a large metropolitan district, enrolls almost 200 students. Forty percent of these students are eligible for free and reduced-price lunch. The school represents the merging of two independent private schools--an elementary/middle school and a high school--which had been created to address some of the academic, cultural, and political gaps in the education of children from the African-American community. Each school was the product of activist teachers and parents who wanted their children educated within a more culturally appropriate setting than public, private or parochial schools offered at the time. The schools were founded during the mid-1970s and merged under a charter in 1995 into one school in order to offer the specialized educational programs to other students in the community who were unable to afford the tuition at the private schools. The high school is a college preparatory academy that was developed to receive students from the elementary school.
The school's educational program is built on seven principles, including unity, self-determination, collective work and responsibility, cooperative economics, purpose, creativity, and faith. The underlying premise of the school is the belief that building a strong Afrocentric identity will give the youth the power and strength to succeed in life. The school is organized into three divisions, each covering multiple grades; students are assigned to divisions rather than to traditional grades. The school organizes students into cross-age groupings based on achievement level, interests, and social maturity. Students take placement tests at the beginning of each semester and a plan of study is developed for each student based on those plans along with teacher, student, and parent input.
Within the groupings, teachers introduce basic or extended concepts and encourage students to explore those concepts more deeply through the completion of projects, participation in debates, and making presentations to their peers. The school has a strong focus on monitoring their progress toward student achievement goals. Student growth is measured through teacher observation of the level of originality and quality of student work, and through demonstrated mastery of subject matter based on examinations, demonstrations, and contents of student portfolios. Students are also assessed using the state achievement test, and two additional standardized tests. The school analyzes data from all sources to make decisions on ways to improve the school's educational programs. The school offers summer programs, after-school care, and literacy and other education programs for parents and the wider community.
In this case, both the school and the surrounding district are primarily African American. However, approximately 40 percent of charter schools that enroll predominantly African-American students (greater than 50 percent) are in districts where the average White enrollment is more than 50 percent, including three all African-American schools in districts that average more than 60 percent White students, making these schools racially distinct from their districts. Since they are publicly funded schools, they are open to all students, but the educational programs in these and other charter schools have been developed for a particular racial/ethnic group.
Predominantly Native American Schools. We focus here on one of the 13 charter schools that serve more than 50 percent Native American students. The following example is of a newly created charter school that began operation to serve Native American students living on a reservation. Its students are essentially all Native Americans, while the district in which the school is located has an average White student enrollment of 60percent.
The small, rural K-12 school is located on a reservation. Several community members founded the charter school because of their concerns about the negative outcomes experienced by Native American students in other schools in the district. These problems included high suspension and dropout rates and the over-representation of Native American students in special education programs. The charter founders believed that a new school could develop a program that built on the students' culture and language and address a complex of neglected issues for this population. For example, the school offers community-based services and publicly funded treatment programs that young people had been unable to get because they were unavailable on the reservation.
The school enrolls about 50 students, 92 percent of whom are Native American and 44 percent of whom are eligible to receive special education services. The director reported that what is important about the school are the "human relationships" that are developed between the students and teachers. The school focuses on assessing the strengths of each student and building an individual educational experience around the abilities of each student to allow and encourage student success. The resulting educational program is one that emphasizes flexibility in response to students' interests and needs.
Predominantly White Schools That Have a Distinctively Higher Proportion of White Students than the District. Our final examples in this section concern schools that have predominantly White students within a district where the average percentage of White students is lower by 20 percent or more. Our earlier analysis located only 16 charter schools (four percent of the charter schools where both school and corresponding district information on ethnicity was available) that serve 20 percent higher percentage of White students than the district. In the fieldwork sample, we had five such cases out of 70 sites (seven percent). None of these sites show any signs of a deliberate selection of White students or a restriction against non-White students, as the following synopses illustrate:
One charter school is located in a small district that includes several small towns and many rural areas. The school is located in a rural area surrounded by isolated, rolling hills, and wooded property. The school, which enrolls fewer than 200 students, is both an alternative school and an institutional support for families who are home-schooling their children. The alternative high school was formerly a regular district school and is designed to meet the needs of students who have not succeeded in the regular high-school program. The alternative school program provides both classes in the core curriculum--mathematics, English, and history--as well as elective courses. Classes are small--12 to 15 students--and faculty seek to create an educational environment that is both relevant to the needs of students and engages them in the educational process.
The school also supports parents who are home-schooling their children. School staff--primarily part-time employees--meet approximately weekly with parents and students in their homes. The staff give support to parents who are creating educational experiences at home. The school has a curriculum library and materials including mathematics manipulatives, computer programs, textbooks, and resource materials that are available to parents. Some of the high school-aged, home-schooled students attend classes on site with the students at the alternative school. Home-schooling parents and students in the alternative program report that they are attracted by the individualized attention for students.
The home school draws students from the entire district, but the alternative program draws primarily from the nearby rural and remote areas of the district, which are primarily White. The percentage of White students is much higher at the charter school than in the surrounding district--99 percent of students at the charter school are White compared to 67 percent of students in the district. School staff reports that the home-schooled population in the district has traditionally been White.
Another example illustrates a different type of education program that draws predominantly White students.
One charter school was previously a private Montessori School that converted to charter status. It is located in a district where the population is almost 60 percent White and the school population is 85 percent White. The school, which enrolls more than 250 students in grades pre-kindergarten through eight, converted to charter status in part to provide a Montessori education to students who were unable to afford tuition at the private school. The director also reported that the school converted to charter status in order to accommodate students from the surrounding neighborhood. Although the school staff reports that the school population is more diverse than before conversion, the school retained most of the students who were enrolled prior to its conversion to charter status. The school does not attempt to enroll particular groups of students, but because it is a Montessori school, it does try to enroll parents who believe in the Montessori philosophy.
The school's curriculum and instruction vary significantly from a conventional school, although they are representative of other Montessori schools. Students are given a contract of work they need to accomplish, but are free to work at their own pace. Much of the instruction is either individualized or in small groups. Class sizes are small, with the goal being no more than 20 students per class. In the middle grades, students spend their morning doing individual work, while the afternoon is reserved for group work. Their curriculum is integrated and focuses on the natural, social, and physical worlds.
The examples in this section shed some light on charter schools that serve a particular racial group, whether or not they are distinctive from their district. We see that the cases vary from charter schools where the students mirror the racial category of the district to sites where charter developers design their program with a particular racial/ethnic group in mind. In the latter case, charter developers clearly believe they are providing an educational program and other services that their students were not able to get from the public school district.
EXHIBIT 4-13
ESTIMATED PERCENTAGE OF LEP STUDENTS IN CHARTER SCHOOLS (1996-97)[1] AND IN ALL PUBLIC SCHOOLS IN THE SIXTEEN CHARTER STATES (1994-95)[2]
| State |
LEP students of statewide enrollment |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Charter schools |
All public schools |
|||||||||||
|
# of students |
% of students |
# of students |
% of students |
|||||||||
|
California |
8,702 |
20.7% |
1,262,982 |
23.6% |
||||||||
|
Arizona |
1,732 |
10.6% |
98,128 |
13.3% |
||||||||
|
Michigan |
268 |
2.6% |
47,123 |
3.0% |
||||||||
|
Colorado |
59 |
0.9% |
26,765 |
4.2% |
||||||||
|
Minnesota |
186 |
8.7% |
21,738 |
2.7% |
||||||||
|
Massachusetts |
312 |
7.8% |
44,211 |
4.9% |
||||||||
|
Texas |
247 |
10.9% |
454,883 |
12.6% |
||||||||
|
Wisconsin |
1 |
0.1% |
20,541 |
2.4% |
||||||||
|
Georgia |
115 |
1.7% |
12,726 |
1.0% |
||||||||
|
Florida |
1 |
0.2% |
153,841 |
6.4% |
||||||||
|
New Mexico |
634 |
17.1% |
80,850 |
24.9% |
||||||||
|
Louisiana |
0 |
0% |
6,336 |
0.8% |
||||||||
|
Alaska |
3 |
3.3% |
29,929 |
24.0% |
||||||||
|
Delaware |
0 |
0% |
1,684 |
1.6% |
||||||||
|
District of Columbia |
71 |
40.8% |
5,151 |
6.4% |
||||||||
|
Hawaii |
28 |
3.4% |
12,186 |
6.7% |
||||||||
|
Total[3] |
12,359 |
12.7% |
2,279,074 |
11.5% |
||||||||
2 Source: Office of Bilingual Education and Minority Languages Affairs, Summary Report of the Survey of the States' Limited English Proficient Students and Available Educational Programs and Services 1994-1995: Washington, DC: U.S. Department of Education, 1996.
3 In interpreting the meaning of the figures in the Total row, it is important to realize that California has the largest number of students in charter schools, representing almost half (46 percent) of all charter school students for the time period shown in the table. California also accounts for about 70 percent of all LEP charter school students, while the state represents 55 percent of the LEP students in public schools in the 16 charter states. These figures imply that the total percent of LEP charter school students is positively affected by the California numbers and the total percent of LEP public school students is influenced by the California numbers to a lesser degree.
1 This exhibit draws on data from 373 open charter schools which report information on Limited English Proficiency. The percentage of Limited English Proficient charter school students is computed by dividing the count of LEP students by the total number of enrolled students.
On a state-by-state basis, there is a wide variation in the percentage of LEP students in charter schools compared to public schools. Of the states with a significant number of charter school students, California has the highest percentage of LEP students (20.7 percent), which is about the same percentage of LEP students as in the state's public schools (23.6 percent). The overall averages reflect the large number of charter and other public school LEP students in California, which boost the national totals significantly. Sixty percent of charter schools have no LEP students, and about 85 percent have less than ten percent LEP students. But six percent of charter schools serve more than one-half LEP students. Many such schools are in districts that have a large non-White population that is not necessarily the same as the racial/ethnic group represented among LEP students. In the example below, the charter school enrollment is predominantly LEP students; the surrounding district has an average of over 80 percent non-White students, the large majority of whom are African American:
This newly created urban middle school, located in an industrial area of a large city, enrolls students in grades 6, 7, and 8. The school attracts students from a nearby elementary school that serves a primarily Hispanic population. One-hundred percent of the school's students are Hispanic, 70 percent have limited proficiency in English, and all of the students are eligible for free and reduced-price lunch. The school was founded out of the concern of parents and teachers at the elementary school who were worried about children transitioning between the small, familial elementary school to one of two large junior high schools in the district that offered few services to language minority students. Some of the school's approximately 150 students are recent immigrants to the United States and need further support in their native Spanish language.
The smallness of the school creates a sense of community that binds teachers, students, and parents at this charter school. There is general agreement among all that the school has succeeded in creating an environment where students feel safe, faculty and staff care, and parents are welcomed. The school provides opportunities for parent involvement both in support of student learning and in school governance. Parents report that the school's small class sizes help students master English as well as academic content. The school's educational program draws on the district curriculum, builds on the cultures of the students and families, and seeks to prepare students for high school. The school has experienced turnover in both its teaching staff and its administrative staff but maintains its vision of providing an appropriate educational experience for its students.
EXHIBIT 4-14
ESTIMATED PERCENTAGE OF STUDENTS WITH DISABILITIES IN CHARTER SCHOOLS (1996-97)[1] AND ALL PUBLIC SCHOOLS IN THE SIXTEEN CHARTER STATES (1995-96)[2]
|
State |
Students with disabilities |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
Charter schools |
All public schools |
||||||||||
|
# of students |
% of students |
# of students |
% of students |
||||||||
|
California |
2,680 |
6.4% |
510,875 |
9.9% |
|||||||
|
Arizona |
1,428 |
8.8% |
68,228 |
9.6% |
|||||||
|
Michigan |
530 |
5.2% |
170,527 |
10.8% |
|||||||
|
Colorado |
497 |
7.7% |
62,697 |
9.9% |
|||||||
|
Minnesota[3] |
590 |
27.5% |
87,530 |
11.0% |
|||||||
|
Massachusetts |
366 |
9.2% |
142,955 |
16.6% |
|||||||
|
Texas |
85 |
3.8% |
409,281 |
11.3% |
|||||||
|
Wisconsin |
90 |
8.6% |
92,868 |
10.9% |
|||||||
|
Georgia |
398 |
5.9% |
121,728 |
9.6% |
|||||||
|
Florida |
49 |
11.0% |
283,104 |
13.7% |
|||||||
|
New Mexico[4] |
1,056 |
28.5% |
43,015 |
13.4% |
|||||||
|
Louisiana |
10 |
3.6% |
81,471 |
10.5% |
|||||||
|
Alaska |
6 |
6.6% |
15,589 |
12.8% |
|||||||
|
Delaware |
18 |
6.2% |
13,719 |
6.8% |
|||||||
|
District of Columbia[5] |
53 |
30.5% |
6,671 |
17.7% |
|||||||
|
Hawaii |
60 |
7.2% |
14,723 |
8.0% |
|||||||
|
Total[6] |
7,916 |
8.1% |
2,124,981 |
11.1% |
|||||||
1 This exhibit draws on data from 357 open charter schools which reported information on students with disabilities.
2 The percent of students served in all public schools in the sixteen charter states was computed using two sources. The source for the number of students served under IDEA was: Nineteenth Annual Report to Congress of the Implementation of the Individuals with Disabilities Education Act, U.S. Department of Education, October 1997. The source for the denominator (enrollment in the sixteen charter states) was U.S. Department of Education, National Center for Education Statistics, Common Core of Data Survey, 1994-95. Total student enrollment included students in open public schools which reported student membership.
3 Many Minnesota charter schools were designed to serve at-risk students and students with disabilities. Their founders sought to provide opportunities to students who were not necessarily being well-served by the traditional public schools. Two of the largest Minnesota schools were specifically designed to serve students with disabilities.
4 New Mexico data reflect only four charter schools; the number of schools is too small to draw any conclusions from the higher percentage of students with disabilities in charter schools compared to other public schools.
5 The District of Columbia data reflect only two charter schools; the number of schools is too small to draw any conclusions from the higher percentage of students with disabilities in charter schools compared to other public schools.
6 In interpreting the meaning of the figures in the Total row, it is important to realize that California has the largest number of students in charter schools, representing almost half (46 percent) of all charter school students for the time period shown in the table. California also accounts for one-third of all charter school students with disabilities, but the state represents less than one-quarter of the students in public schools in the 16 charter states. These figures imply that the total percent of charter school students with disabilities is negatively affected by the low percentage in California of charter school students with disabilities.
Eleven schools from the telephone survey reported that over one-half of their students have disabilities. The following two fieldwork sites illustrate this type of charter school. One school aims to serve students who have weak reading skills, many of whom have been identified as students with disabilities.
One formerly public kindergarten through eighth grade charter school targets students who are weak readers. Slightly more than one-half of the school's 175 students qualify for special education services and have current Individualized Educational Plans (IEPs). (The district average of students eligible for special education services is 13 percent.) The majority of the charter school students also are eligible for free and reduced-priced lunch and about 60 percent of the students are African American. Parents reported enrolling their children in the school because of their reading deficits. This is consistent with the school's vision, which is dedicated to the education of children with readiness skill deficits, including: developmental delays, reading difficulties, and various forms of learning disabilities. Staff at the school believe that one should first try to assist a person to accommodate to the environment, but if that should fail, the environment should be changed to meet the individual's needs. This philosophy has encouraged teachers at other schools in the surrounding district to refer parents whose children have similar reading disabilities.
The school's educational program focuses on the neurological aspects of reading disabilities, provides small-group instruction, and has a combined phonics and whole-language approach. Teachers report that the curriculum is teacher-generated to meet the needs of the students. The average class size is 22 students, with two adults per classroom, but the flexible grouping at the school means that teachers may have the same students over a three-year period. Students appreciate the fact that teachers take time to explain things and that they are encouraged to help each other. The school strives to return students to the regular public school once they are reading at grade level.
Another school focuses on students with disabilities and students with attention deficit disorder:
This small, formerly public kindergarten through 12th grade charter school targets learning disabled and artistically creative students. These student characteristics are what distinguishes the school's student body from that of the surrounding district; the charter school's student body is similar to the district's in terms of ethnicity and income level. One-third of the fewer than 100 students have active IEPs and are eligible for special education services. The school also has a large population of students with attention deficit disorder and some do not qualify for special education services. In addition, the school attracts a considerable number of gifted students who were bored with their previous schools and seek the more challenging and flexible environment offered by the charter school.
The school's program integrates the arts into the school's multigraded settings and provides ample opportunities for instructional flexibility. Parents, teachers, and students all described the school as a place where individuality is valued and differences in learning styles are seen as an asset rather than a handicap. Counseling, arts education, and academics are seen as equal parts of a holistic educational experience. The high school is broken into Advisory Groups, which meet every day for half an hour to work on a variety of projects and discuss issues or concerns. Advisory Groups are seen as an important part of the support network for students at the school; the small student-teacher ratio of the Advisory Groups (nine to ten students per staff member) allows the staff to be more familiar with students' needs.
Both of these charter schools have designed programs specifically to meet the unique needs of these groups of students. They are what might be called niche schools, providing a much-needed service for parents who feel their children are not adequately supported in regular public schools.
About 15 percent of the fieldwork schools specifically target at-risk students or dropouts. (Schools serving the broader population of students that are economically disadvantaged are described above.) The founders of these charter schools have identified a group of students whom the regular system has failed and designed programs to meet their needs. The following examples illustrate differences among the types of students enrolled in schools targeting at-risk students as well as some of the kinds of programs offered:
This newly created kindergarten through 12th grade charter enrolls about 750 students. Many of these students are not successful in the traditional high school, are students who need a flexible schedule in order to pursue artistic training, who want to accelerate their school program, or who are being home-schooled. In other respects, the student body is similar to that of surrounding district schools. (The school enrolls about three-quarters White students, about four percent students with disabilities, and no students who are Limited English Proficient.)
The school's individualized program suits the needs of this targeted group of students through a combination of classes delivered by teachers at the schools, an on-site interactive instructional television link, individual instruction from teachers, and home-schooling with required parent participation. Students have some classes on site, but spend most of their time working independently on a mixture of purchased and school-developed curriculum units. They are required to meet with their teachers at least once a week.
A second school illustrates how a similar population of students can be served by a different type of educational program:
This formerly public charter school, serving young people between the ages of 16 and 21, has an official policy: "In order to be admitted to [the school], a student must have withdrawn officially from the public schools without graduating or must be seriously at-risk of dropping out." Students at risk of dropping out must be referred officially to the school by another public school principal, guidance counselor, or other school official. In addition, students eligible to attend the school must be at least one grade level behind their graduating class cohort. The school enrolls about 100 students. As a result of the targeting of these groups of students, many are low-income (more than one-third of these students are eligible for free or reduced-priced lunch), 15 percent are Limited English Proficient, and more than half of the students are of either Hispanic or Asian decent. The small college preparatory school located on a community college campus is organized into multigraded small classes. The school operates from 11:00 AM to 8:00 PM to accommodate the needs of its students, many of whom are working or are parents. The school strives to meet individual learning needs and offers a competency-based, interdisciplinary curriculum.
These cases illustrate that many niche charter schools are similar to alternative public schools, except that they are not responsible to districts. All of the schools we described are in states that have established a central state authority to grant charters. The difficulty of overcoming bureaucratic impediments to starting alternative schools is a common complaint reported to us by charter developers of niche schools. Consequently, freedom from district bureaucracy may produce innovations that better serve youngsters. However, it is too early to tell whether they serve students as well as alternative schools that are accountable to their districts. The Study plans to conduct research in the future to address these questions.
2 The statistics in columns two and four are the result of pooling student enrollment data for different racial/ethnicity categories across the 89 percent of operational charter schools that responded to our survey and had valid racial data. (Columns three and five represent a similar computation for all public schools.) Since the number of charter schools is small and most charter schools have low enrollments, the larger schools have a disproportionate weight in the calculation. The last two columns show the result of computing similar numbers using the school as the unit of analysis rather than the student. Hereafter, we use the school as the unit for computation, unless otherwise noted.
3 The charter school average percentage of White students is higher than the pooled percentage of White students (compare data in column five to data in column three) because of the relationship between the racial/ethnic distribution and school size. Our data suggest that small (those with fewer than 200 students) charter schools are more likely to serve predominantly (greater than two-thirds) White students than are larger charter schools (slightly more than half of small charter schools serve predominantly White students, while fewer than 20 percent of other public schools do). Small charter schools are also more likely than other public schools to serve students of color: 27 percent of small charter schools serve predominantly students of color, while 14 percent of other public schools do. See Appendix D. The term "student of color" is used here to mean students who are not classified in the racial/ethnicity Census categories as White not Hispanic. This term is somewhat inaccurate for a variety of reasons, not the least of which is that the use of this term implies that all Hispanic students are non-White. Nonetheless, we will use "student of color" or "non-White" throughout the text.
4 Our analyses rely on data from schools in 16 charter states. Hereafter, the main body of the report provides data based on school averages or percentages. Exhibit D-3 in Appendix D shows data comparable to Exhibit 4-2 for the percentage of students in each state, regardless of the differences across schools in the enrollment.
5 Of the 102 charter schools in our sample that enroll between two-thirds and 100 percent students of color, about one-third serve predominantly (greater than or equal to 50 percent) African-American students, about one-third serve predominantly Hispanic students, and 14 charter schools serve primarily Native American students. Only two charter schools had students who were predominantly of Asian origin.
6 Exhibit D-4 in Appendix D compares alternative measures of racial enrollment patterns for charter schools on a state-by-state basis.
7 Exhibit D-5 in Appendix D shows a state by state distribution for all three levels of racial concentration for charter schools and all public schools in the 16 charter states.
8 First, charter schools in such states as Arizona, Massachusetts and Michigan may be authorized by state agencies and therefore may not be part of any local school district. Second, whether or not a charter school has been granted a charter by a local school district, some charter schools are not "within" the boundaries of any local districts. For example, home schools, distance learning, some special education, and a variety of other types of charter schools draw on students from several districts and locations. Third, some charter schools in Arizona have branch schools that are located in different locales. In total, we estimate that between 15 and 20 percent of charter schools do not necessarily draw students from within the boundaries of one district. In our fieldwork sample, about 15 percent of our field sites fit this description.
9 Comparison districts were identified in several different ways. Schools that were surveyed in the 1996 telephone survey were asked to identify the district in which they were physically located, and which we used as a district comparison. For schools not responding to the 1996 survey, we utilized information provided by the U.S. Department of Education where possible. These districts were selected by matching school and district zip codes. For several schools that we visited, we knew that the school enrolled students from multiple districts. For example, several Arizona schools enroll students from the entire Phoenix metropolitan area. In these cases, we used the Consolidated Metropolitan Statistical Area (CMSA) as a comparison rather than an individual district, and averaged student demographic information for schools in this CMSA. Similarly, for Massachusetts schools which pull students from multiple districts, we requested information from the Massachusetts Charter School office on numbers of students enrolled in given districts. For two Massachusetts schools, we used two districts as comparison sites and averaged student demographic information. Where district information was still missing, we telephoned the school to request their assistance with identifying a comparison district.
10 In statistical terms, the graph shows the results of a linear regression of the percentage of White students at charter schools regressed on the average percentage of White students enrolled in the surrounding district. The adjusted R Square is .50, which is highly significant. The regression coefficient is 1.08, which means that a one percent increase in the district average is likely to produce a one percent rise in the charter school percentage.
11 We also examined distinctiveness at the ten and 15 percent cutoffs. Though the exact percentage of charter schools that are or are not distinctive changes according to which definition we apply, the basic pattern discussed in this section is not sensitive to which of these definitions we use. We chose to use the 20 percent cutoff as the most conservative definition of distinctiveness.
12 For approximately five percent of the responding sample, we could not assign any single or multiple districts as a basis for charter school comparison.
13 The percentages in these tables are slightly different from those shown in Exhibit 4-5 because the number of charter schools used in the table is smaller, as explained in the text and the preceding footnote.
14 Children are eligible for free meals under the National School Lunch Program if their family income is at or below 130 percent of the official poverty level (in 1997, 130 percent of the poverty level is $20,865 for a family of four.) Children are eligible for reduced-price meals when their family incomes are between 130 percent and 185 percent of the poverty level (in 1997 185 percent of the poverty level for a family of four is $29,693).
15 For the majority of districts, we were able to obtain free and reduced lunch data from the National Center for Education Statistics' 1993-94 Schools and Staffing Survey, Teacher Demand and Shortage Questionnaire for Public School Districts. For two additional states (Arizona and Massachusetts), we were able to obtain some additional 1996-97 data from state sources.
16 The sources of the reduction in data are missing school data (36 schools), missing district data (95 schools) or missing data for both school and district (25 schools).
17 To explore the possibility that the missing data causes a bias, we conducted response bias analyses to determine whether there was systematic bias between those schools for which we were able to make district comparisons and those for which we were unable to make comparisons. Our analyses revealed that the two groups were similar with respect to distribution by state, school size, or school type (new versus pre-existing schools). In other words, to the degree that we were able to determine, we found no evidence of bias. Nevertheless, these comparative data should be treated as exploratory and subject to change given more complete data.
18 The 93 schools in our fieldwork sample were selected in an unbiased way from states with charter schools in operation for at least one year at the time of selection. The selection process aimed to generate a diverse group of schools to study more intensively, so schools were chosen within each state to include schools that are newly created, schools that converted to charter status, schools that are elementary, middle, and high schools, and schools that are distributed geographically within the state. Though the sample is not strictly representative of all charter schools, the 93 fieldwork sites that make up the sample have demographic characteristics very similar to the universe of charter schools. Two of the 93 original charter schools are no longer part of our fieldwork sample. One is no longer a charter school, and the other school has decided not to continue to participate in the Study.
19 We categorized the sample of 91 fieldwork schools into the same three categories as we did in the analysis of all schools: schools enrolling fewer than one-third White students, schools enrolling between one-third and two-thirds White students and schools enrolling more than two-thirds White students. When grouped this way, the fieldwork sites and all charter schools show a nearly identical distribution: 23 percent of the fieldwork sample and 28 percent of all charter schools enroll fewer than one-third White students; 26 percent of the fieldwork sample and 21 percent of all charter schools enroll between one-third and two-thirds students of color; and 51 percent of the fieldwork sample and 51 percent of all charter schools enroll more than one-third White students. Although the numbers become too small to reliably compute percentages, within the three categories, the proportion of fieldwork schools that are distinctive from their districts is roughly the same as for all charter schools.
[V. Why Charter Schools Are Started and What Attracts Parents to Them]