No Child Left Behind Act: Education Needs to Provide Additional  
Technical Assistance and Conduct Implementation Studies for	 
School Choice Provision (10-DEC-04, GAO-05-7).			 
                                                                 
The school choice provision of the No Child Left Behind Act	 
(NCLBA) of 2001 applies to schools that receive Title I funds and
that have not met state performance goals for 2 consecutive	 
years, including goals set before the enactment of NCLBA.	 
Students in such schools must be offered the choice to transfer  
to another school in the district. GAO undertook this review to  
provide the Congress a report on the first 2 years of the	 
implementation of NCLBA school choice. GAO reviewed (1) the	 
number of Title I schools and students that have been affected	 
nationally, (2) the experiences of selected school districts in  
implementing choice, and (3) the guidance and technical 	 
assistance that Education provided. GAO collected school	 
performance
data from all states, interviewed Education officials, and	 
visited 8 school districts in California, Illinois, Ohio,	 
Mississippi, Pennsylvania, Tennessee, and Washington.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-7						        
    ACCNO:   A14336						        
  TITLE:     No Child Left Behind Act: Education Needs to Provide     
Additional Technical Assistance and Conduct Implementation	 
Studies for School Choice Provision				 
     DATE:   12/10/2004 
  SUBJECT:   Elementary education				 
	     Elementary school students 			 
	     Performance measures				 
	     Public schools					 
	     School districts					 
	     Academic achievement				 
	     Secondary education				 
	     Secondary school students				 
	     Disadvantaged persons				 
	     School choice					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-05-7

     

     * Results in Brief
     * Background
          * Accountability for Results in All Public Schools under NCLBA and
            Prior Law
          * School Choice under NCLBA
     * About 1 in 10 Schools Identified for Choice in the First 2 Years of
       NCLBA, and 1 Percent of Eligible Students Transferred in 2003-2004
          * From 10 to 12 Percent of Public Schools Nationwide That Received
            Title I Funds Were Identified for Choice in Each of the First 2
            Years of NCLBA
          * About 1 Percent of Eligible Students Transferred in the Second
            Year; Little Is Known about Their Characteristics
     * Implementation of Choice Was Challenging in Selected Districts
          * Short Timeframes Created Challenges
          * Most Districts Offered Parents a Number of School
          * Several Districts We Visited Cited Lack of Sufficient Capacity as
            Limiting the Number of Transfers
          * Districts Used Different Strategies to Meet Transportation
            Requirement
     * Education Issued Extensive Guidance on Choice, but Questions Remain as
       Implementation Proceeds
          * In the First Year, Districts Implemented Choice Using Preliminary
            Guidance Released through Letters to Superintendents
          * Additional Guidance and Technical Assistance Provided in the
            Second Year of Implementation
          * Complexity of Implementing Choice Raises Issues Not Addressed in
            Guidance
     * Conclusions
     * Recommendations for Executive Action
     * Agency Comments and Our Evaluation
     * Appendix I: Scope and Methodology
     * Appendix II: NCLBA Interventions for Schools Not Meeting Yearly
       Performance Goals Over Time
     * Appendix III: Number of Title I Schools in Each State Identified for
       Choice in School Year 2002-2003
     * Appendix IV: Number of Title I Schools in Each State Identified for
       Choice in School Year 2003-2004
     * Appendix V: Number of Students in Each State Transferring under Choice
       Option in First 2 Years of NCLBA
     * Appendix VI: Poverty and Minority Rates of Schools Required to Offer
       Choice and Schools Offered as Transfer Options
     * Appendix VII: Comments from the Department of Education
     * Appendix VIII: GAO Contacts and Staff Acknowledgments
     * GAO Contacts
     * Staff Acknowledgments
     * Related GAO Products
          * Order by Mail or Phone

                 United States Government Accountability Office

Report to the Secretary of Education

GAO

December 2004

NO CHILD LEFT BEHIND ACT

     Education Needs to Provide Additional Technical Assistance and Conduct
               Implementation Studies for School Choice Provision

GAO-05-7

NO CHILD LEFT BEHIND ACT

Education Needs to Provide Additional Technical Assistance and Conduct
Implementation Studies for School Choice Provision

  What GAO Found

About 1 in 10 of the nation's 50,000 Title I schools were identified for
school choice in each of the first 2 years since enactment of the No Child
Left Behind Act (NCLBA) of 2001. The proportion of schools identified for
choice varied by state. About 1 percent of eligible children, or 31,000
students, transferred in school year 2003-2004. However, little is known
about the students who did and did not transfer or factors affecting
parents' transfer decisions. Education has launched a study that will
yield some information on these topics.

  Percentages of Title I Schools Identified for School Choice in 2003-2004 by
                                     State

Source: GAO analysis of state data.

Officials in most of the 8 districts GAO visited said they welcomed
NCLBA's emphasis on improved performance, but had difficulties providing
choice because of tight timeframes and insufficient classroom capacity.
Final state determinations of the schools that met state yearly
performance goals were not generally available before the school year
started, so offers of transfers were based on preliminary determinations.
District officials expressed concern that parents had inadequate time and
information to make an informed decision. Parents were offered at least 2
possible schools as transfer options, but many of these schools had not
met state performance goals in the most recent year. Because of limited
classroom capacity in 4 of the districts, some students did not receive
the opportunity to transfer. For students who transferred, transportation
was provided on school buses, public transit or personal cars, and most
districts spent less than 7 percent of the pool of funds that NCLBA
required be set aside for that purpose in school year 2003-2004.

Education issued extensive guidance on choice. However, the complexity of
providing school choice raises a number of issues that have not been
addressed in guidance available through October 2004, such as how to
handle cases where schools receiving transfers later are identified for
choice and how to expand capacity in the short-term within budgetary
constraints.

                 United States Government Accountability Office

Contents

Letter 1
Results in Brief 2
Background 4
About 1 in 10 Schools Identified for Choice in the First 2 Years of
NCLBA, and 1 Percent of Eligible Students Transferred in 2003
2004 8
Implementation of Choice Was Challenging in Selected Districts 18
Education Issued Extensive Guidance on Choice, but Questions
Remain as Implementation Proceeds 28
Conclusions 34
Recommendations for Executive Action 35
Agency Comments and Our Evaluation 36
Appendix III Number of Title I Schools in Each State Identified
for Choice in School Year 2002-2003
Appendix IV Number of Title I Schools in Each State Identified
for Choice in School Year 2003-2004

Appendix I Scope and Methodology

Appendix II NCLBA Interventions for Schools Not Meeting Yearly Performance
Goals Over Time

Appendix V Number of Students in Each State Transferring under Choice
Option in First 2 Years of NCLBA

Appendix VI Poverty and Minority Rates of Schools Required to

Offer Choice and Schools Offered as Transfer

Options

Page i GAO-05-7 No Child Left Behind Act

Appendix VII         Comments from the Department of Education          51 
Appendix VIII        GAO Contacts and Staff Acknowledgments GAO      54 54 
                        ContactsStaff Acknowledgments                   54    
Related GAO Products                                                    55 
Tables               Table 1: Students Eligible but Not Transferring       
                          from Elementary and Middle Schools Identified       
                         for Choice Compared with Students Transferring       
                           in 2003-2004 in One District VisitedTable 2:       
                           Milestone Dates for Implementation of Choice       
                          for School Year 2003-2004 in Six Districts We 
                               VisitedTable 3: Transfer Options Offered 
                         Parents of Elementary School Students in Seven 
                        Districts We VisitedTable 4: Title I Status and 
                             Performance of Schools Offered as Transfer 
                             Options for 2003-2004, by DistrictTable 5: 18 20
                         Poverty and Minority Rates of Schools Required 23 24
                        to Offer Choice and Schools Offered as Transfer 25 25
                         Options in FresnoTable 6: Poverty and Minority 26 28
                          Rates of Schools Required to Offer Choice and 30 40
                                 Schools Offered as Transfer Options in 
                             AkronTable 7: Number of Students Unable to 
                                Transfer Due to Capacity Limitations in 
                           2003-2004Table 8: Transportation Allocations 
                                 and Projected Expenses for School Year 
                        2003-2004, by DistrictTable 9: Dates, Type, and 
                              Title of Guidance or Technical Assistance 
                           Provided by Education on School Choice under 
                        NCLBATable 10: Characteristics for 2003-2004 of 
                                             Eight Districts We Visited 
Figures              Figure 1: Number of Title I Schools and Number        
                        Identified for School Choice in 50 States, the  
                        District of Columbia, and Puerto Rico in School 
                        Years 2002-2003 and 2003-2004                      10

Figure 2: Demographics of Students Enrolled at Title I Schools        
            Identified and Not Identified for Choice in School Year      
2003-2004                                                               11 
Figure 3: Proportion of Schools Identified for Choice at Various      
                 Grade Spans Compared with All Title I Schools in School 
Year 2003-2004                                                          12 
Figure 4: Percentage of Title I Schools Identified for School Choice, 
by State, in School Year 2003-2004                                      13 
Figure 5: Number of Title I Students Eligible for Choice and          
Number That Transferred in 42 States, the District of                 
Columbia, and Puerto Rico in School Year 2003-2004                      15 
Figure 6: Timeline for Tacoma Public Schools, 2003                      21 

        Abbreviations        
        NCLBA                No Child Left Behind Act                   
        NCES                   National Center for Education Statistics 

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office Washington, DC 20548

December 10, 2004

The Honorable Rod Paige The Secretary of Education

Dear Mr. Secretary:

Increased school choice is one of the cornerstones of The No Child Left
Behind Act (NCLBA) of 2001. Signed into law in January 2002, this major
legislation is designed to improve the academic achievement of all
students by requiring states to set annual performance goals for schools
that would result in all students being proficient in reading and math by
school year 2013-2014. Title I of the act provides financial support to
improve the education of low-income students in high-poverty schools. In
2004, about 50,000 public schools-over half of all public schools
nationwide-received a share of $12 billion in Title I funds. While all
public schools are expected to meet state performance goals, some of the
law's provisions apply only to Title I schools, for instance, the school
choice requirement. Title I schools are identified for choice when
performance goals, including goals set before the enactment of NCLBA, are
not met for 2 or more years in a row. In general, all students enrolled in
such schools must be offered the option of transferring to other schools
in the district that have not been identified for choice and
transportation must be provided. Transfer requests cannot be denied for
lack of classroom capacity, according to federal regulations. While states
set the performance goals and schools are judged on the performance of
their students, the nation's 15,000 school districts are held responsible
for implementing the school choice requirements of NCLBA, starting in
school year 2002-2003.

The school choice provision of NCLBA is intended to provide a unique
opportunity for students in schools not meeting state goals to attend
schools that have had greater success meeting these goals. However, little
is known about how states and districts are implementing the provision and
how many schools and students are participating. We undertook this review
to provide the Congress an early report on the first 2 years of NCLBA
implementation, with a more detailed look at the experiences of 8 school
districts. Specifically, we address:

                                Results in Brief

 1. To what extent have Title I schools been affected by the school choice
       provision of NCLBA in terms of the number of schools identified for
       choice and the number of students transferring?
 2. What were the experiences of selected school districts in implementing
       the choice provision?
 3. What guidance and technical assistance did the Department of Education
       provide states and districts as they implemented public school choice?

To determine the numbers of schools and students affected by choice
nationwide during the first 2 years of NCLBA, we compiled and analyzed
data from a variety of sources, including our surveys of state education
agencies and school districts, Department of Education (Education) reports
of state agencies' data, and other Education databases. We assessed the
reliability of these data and determined that they were adequate for our
purposes. To obtain information on district implementation, we visited 8
districts that had schools identified for choice and interviewed district
and school officials about their experiences. Located in seven
states-California, Illinois, Ohio, Mississippi, Pennsylvania, Tennessee,
and Washington-these districts were selected to achieve variation in
geographic, demographic, and program characteristics. All but one district
we visited had students transferring under choice. One rural district
-North Panola, Mississippi- offered supplemental educational services
instead of transfers under choice, as permitted, because the middle school
identified for choice was the only middle school in the district. To
ascertain the role of Education in providing guidance and technical
assistance, we reviewed regulations and other forms of guidance available
from Education and interviewed officials at the federal, state, and
district levels, as well as representatives of national educational
organizations and other experts in the area. See appendix I for a more
detailed description of our methodology. We conducted our work from
October 2003 through November 2004 in accordance with generally accepted
government auditing standards.

Since passage of NCLBA, about 1 in 10 schools receiving federal funds for
low-income students under Title I-approximately 1 in 20 public schools
nationwide-has been identified for school choice. In school year 20022003,
the first year under NCLBA, about 5,300 schools attended by 3 million
children were identified for choice; in the following school year, this
number rose to about 6,200 schools. These schools had larger

    Page 2 GAO-05-7 No Child Left Behind Act

proportions of minority and low-income students enrolled than Title I
schools that were not identified for choice. Across states, the percentage
of Title I schools identified for choice varied considerably. A number of
factors may contribute to this variation, including differences in the
diversity and size of school populations, as well as differences in
states' approaches to defining their yearly performance goals. Overall,
about 19,000 students transferred under the NCLBA choice option in the
first year and an additional 31,000 in the second year, representing about
1 percent of those eligible. While little is known about the factors that
affect transfer decisions or about the students involved, Education has
contracted for a study that will examine the reasons that parents give for
their transfer decisions and identify students' demographic
characteristics. Part of this study, still under design, is intended to
examine achievement outcomes of students who transfer. Our analysis of
data from one district showed that lower percentages of transferring
students were minorities and lower percentages were from low-income
families than was the case for the eligible students who did not transfer.

Officials at most of the 8 districts we visited said that they welcomed
NCLBA's increased emphasis on improved performance but had difficulties
providing school choice due to tight timeframes and insufficient classroom
capacity. One district received final data on schools' performance about a
month before school started in the fall; the other seven districts
received final data weeks or even several months after the school year had
begun. In order to notify parents of their options before the start of the
school year as required by law, most districts relied on preliminary data
to make school status determinations, putting the district at risk of
misidentifying schools for choice and misinforming parents about their
transfer options. District officials expressed concerns that parents may
not always fully understand their options or have adequate time or
information to make a fully informed decision. The total number of schools
offered as transfer options to any one parent ranged from 2 to over 30,
depending on the district, but many schools offered were similar to the
students' current schools. Limitations in available classroom capacity in
4 districts resulted in some eligible students not having an opportunity
to transfer. District officials provided various explanations for why this
occurred, including lack of feasible alternatives to increase capacity.
Districts employed a variety of strategies to provide transportation to
transferring students, including school buses, subsidized passes on public
transit, and stipends to parents, and most districts we visited spent less
than 7 percent of the pool of funds that NCLBA required to be set aside
for transportation and supplemental services costs.

                                   Background

Education issued guidance and final regulations on school choice in
December 2002, after the start of the first school year under NCBLA. Most
districts implemented school choice based on the preliminary guidance
contained in letters they received in June 2002, but some issues remained
unclear. Additional assistance was provided subsequently in the form of
policy letters, training tools, and presentations at conferences.
Education issued updated guidance in February 2004, after the start of the
second school year under NCLBA, and followed with a handbook on promising
practices in the implementation of school choice. While district officials
we visited generally had access to Education's guidance, questions
concerning the implementation of school choice remained, for example, how
to offer choice when physical capacity is limited. Other questions
involved issues that may arise as NCLBA implementation progresses, for
instance, districts' responsibilities if a school to which students have
transferred subsequently does not meet state goals and is itself required
to offer choice.

To help state and district officials in implementing choice, we are
recommending that Education monitor issues and promote promising practices
related to limited physical capacity and assist in developing strategies
to better inform parents about their transfer options. In addition, we are
recommending that Education undertake a rigorous study that will provide
more insight on the effects of school choice. In its comments on a draft
of this report, Education agreed with our recommendations on capacity and
on parental information. Education also agreed to consider including an
examination of academic outcomes and retention rates of transferring
students in its planned NCLBA implementation study. Education's comments
appear in appendix VII.

The Department of Education is responsible for overseeing state
implementation of NCLBA, which amended and reauthorized the Elementary and
Secondary Education Act. Title I of this act authorizes funds to states
for local school districts with high concentrations of children from
low-income families to improve the academic achievement of students
failing or at risk of failing to meet state standards. Title I is the
single largest federal program supporting education in kindergarten
through 12th grade, 1 supplying over $12 billion in federal funds in 2004.

1

In this report, we refer to Title I, Part A of the Elementary and
Secondary Education Act as "Title I." Other Parts of Title I (Parts B, C,
and D) are targeted at specific populations or purposes and are commonly
referred to by their program names, such as Even Start.

Page 4 GAO-05-7 No Child Left Behind Act

Accountability for Results in All Public Schools under NCLBA and Prior Law

These funds are designed to supplement the instructional services and
support that districts and schools provide. Title I and other federal
funding represent about 8 percent of total spending on elementary and
secondary public education, with the remaining 92 percent provided
primarily by states and localities. Title I funds are distributed by
formula to state education agencies, which retain a share but pass through
most of the funds to school districts. 2 Districts with at least a minimum
number and percentage of low-income students receive a share of Title I
funds. The districts are required to distribute Title I funds first to
schools with high poverty rates-over 75 percent-and then to eligible
schools in rank order of poverty either districtwide or within grade
spans. Because school enrollment numbers and demographics may vary from
one year to the next and because districts have some discretion in how
many and which schools receive Title I funds, the status of schools as
Title I or not Title I can vary from one year to the next. Approximately
25 million students were enrolled in schools eligible for Title I funds in
school year 2002-2003 out of a total of about 49 million students in all
schools nationwide, according to Education. 3

Stronger accountability for educational results is one of several
education reform principles embodied in NCLBA and it builds on
requirements in place under prior law. Prior to NCLBA, states were
expected to have accountability systems that included standards for what
students should learn and tests every year in certain grade levels to
measure their knowledge of reading and mathematics. 4 Each year,
increasing percentages of students were expected to demonstrate their
proficiency on these tests, and schools were judged on their ability to
make adequate yearly progress in educating students to the state's
standards-referred to in this report as meeting their yearly performance
goals. Title I schools that did not meet their goals for two consecutive
years were to be designated for improvement, provided technical
assistance, and required

2

State agencies may retain 1 percent for administration with the smallest
states permitted to retain a slightly higher amount, and, starting in
school year 2004-2005, must reserve 4 percent for school improvement
activities. In the 2 earlier years, state agencies were required to set
aside 2 percent for school improvement.

3Education's National Center for Education Statistics maintains
information on schools "eligible" for Title I funds, based on the data
used in the formula for distribution of funds. Given districts'
discretion, however, some eligible schools may not actually receive funds.

4

In this report, when we refer to states, we are including the
jurisdictions of the District of Columbia and Puerto Rico.

to implement improvement plans. States were at various stages of
implementation when NCLBA was enacted, so some states had been identifying
schools for improvement for several years while others were just beginning
the process.

Enactment of NCLBA strengthened accountability requirements by specifying
timetables for school improvement and by holding all public schools,
whether or not they receive Title I funds, accountable for the academic
performance of various subgroups of students. For example, schools must
reach yearly performance goals set by states that will result in 100
percent of students meeting state proficiency standards by school year
2013-2014. 5 In addition to meeting the state's performance goals in
general, schools are responsible for meeting those goals for specified
subgroups of students who (1) are economically disadvantaged,

(2)
           represent major racial and ethnic minorities, (3) have
           disabilities, or

(4)
           are limited in English proficiency. If any subgroup does not meet
           the target, the school is identified as not having made its yearly
           performance goal. 6

While NCLBA requires that all 94,000 public schools in the nation be held
accountable for their performance, it requires specific actions or
corrective interventions only for Title I schools that repeatedly miss
their yearly performance goals. 7 Two kinds of immediate interventions are
required for Title I schools that have not met their performance goals for
two consecutive years. On the one hand, plans are set in motion to improve
the school's performance. At the same time, students must be given the
opportunity to transfer to other schools under the school choice option.
Depending on how often schools continue to miss their goals, other
required actions range from offering students supplemental

5

For more information on state accountability systems, see GAO, No Child
Left Behind Act: Improvements Needed in Education's Process for Tracking
States' Implementation of Key Provisions, GAO-04-734 (Washington, D.C.:
Sept. 30, 2004).

6

The school can meet the state performance goal under the "safe harbor"
provision even when a student subgroup does not otherwise meet the goal.
To do so, the school must reduce the percentage of students in the
subgroup who are not proficient by at least 10 percent and that subgroup
must also show progress on another academic indicator that the state uses
to determine performance. For example, if a subgroup does not make the
state's goal for the percentage of students proficient in reading, but the
percentage in that subgroup who are not proficient declines by 10 percent
and the graduation rate for that subgroup improves, the school could make
its performance goal for that year.

7

States may, at their discretion, apply these interventions to other
schools as well.

School Choice under NCLBA

educational services, such as after-school tutoring, to completely
restructuring schools. See appendix II for further details on the specific
actions required in each year.

The first year that Title I schools do not meet performance goals, no
specific actions are required under NCLBA. However, if the goal is missed
the next year, districts generally must offer parents of students
attending these schools the choice to transfer their child to another
school. 8 The district must provide transportation to the new school,
within limits, and continue to pay for transportation until the school
from which the student transferred is no longer identified for choice. 9
Schools are no longer identified for choice when they have met their
yearly performance goals for at least 2 consecutive years.

Districts are required by federal regulations to offer parents at least
two schools from which to choose, if available, and these schools may be
any public school that is not itself currently identified for choice.
Thus, under NCLBA offered schools could include Title I schools that have
missed their yearly performance goals for a single year or any school that
does not receive Title I funds, regardless of its performance. However,
states could further limit the schools offered as transfer options, for
example, by prohibiting transfers to non-Title I schools that have not met
their yearly performance goals. Under circumstances where no viable
transfer options exist-as in districts with only one school serving
particular grade levels or where all schools in the district have
repeatedly missed their performance goals-districts are required, to the
extent practicable, to make arrangements with other districts to accept
their transfer students. 10

NCLBA requires that districts notify parents of the choice option by the
first day of the school year immediately following the test administration

8

NCLBA states that the opportunity to transfer must be offered unless such
transfers are prohibited by state law.

9

The student may remain in the new school until he or she completes the
highest grade level in the new school, but the district is no longer
required to pay for transportation when the school from which the student
transferred is no longer identified for choice.

10

In addition to making such arrangements for choice, districts have the
option of offering supplemental educational services to students attending
schools that have missed their yearly performance goals for two
consecutive years. If schools miss their yearly performance goals for
three or more years, districts are required to provide supplemental
educational services as well as choice.

  About 1 in 10 Schools Identified for Choice in the First 2 Years of NCLBA, and
  1 Percent of Eligible Students Transferred in 2003

that resulted in the school being identified for choice. For example, if
tests given in spring 2003 resulted in the school being identified for
choice, then the option had to be offered parents by the first day of
school of the 20032004 school year. Notices to parents must be in an
understandable and uniform format and, to the extent practicable, in a
language that parents can understand. These notices must explain why the
school was identified for choice and how it compares with others in the
district and state. In addition, federal regulations require that the
notice include information on the academic achievement of the schools
offered as transfer options.

Districts are not required to give parents their first choice among the
transfer options provided, but may not deny transfer requests based on
lack of physical capacity, such as lack of space within a building or
classroom, according to federal regulations. When deciding which schools
to offer as transfer options, districts can consider the amount of
available capacity, but they must offer options for all students enrolled
in schools identified for choice. When reviewing transfer applications,
making school assignments, and arranging for transportation, districts are
required to give priority to the lowest-achieving low-income students.

In each of the first 2 school years following enactment of NCLBA, from 10
to 12 percent of schools that received federal funds under Title I were
identified for school choice. Several million students were enrolled in
the schools identified for choice and were thus eligible to transfer.
About 31,000 students, representing 1 percent of those eligible,
transferred in the second year, school year 2003-2004. Although Education
has recently begun to collect information on the number of transferring
students, little is known about their demographic or academic
characteristics. Our analysis of data from one district showed that
proportionately fewer minority and low-income students transferred,
compared with students in the same schools who did not transfer.

From 10 to 12 Percent of Public Schools Nationwide That Received Title I
Funds Were Identified for Choice in Each of the First 2 Years of NCLBA

In each of the first 2 school years of NCLBA, about 1 in 10 Title I
schools- about 1 in 20 public schools nationwide-was identified for school
choice. About 5,300 schools 11 attended by 3 million children, were
identified for choice in the first year of NCLBA. As shown in figure 1,
the total number of schools identified for choice increased to about 6,200
in year two. Because schools must meet their performance goals for 2
consecutive years before they are no longer identified for choice, many of
the same schools may have been included in the total number for both the
first and second years.

11

The 5,300 schools shown for school year 2002-2003 are derived from state
data reported in 2004. Although Education released a list in July 2002
estimating that there were over 8,600 such schools for school year
2002-2003, many states later revised these numbers. For example, Education
estimated that 760 Ohio schools were identified for choice in 20022003,
but Ohio reported that only 161 schools were identified for choice in that
year.

Page 9 GAO-05-7 No Child Left Behind Act

Figure 1: Number of Title I Schools and Number Identified for School
Choice in 50 States, the District of Columbia, and Puerto Rico in School
Years 2002-2003 and

Number of Title I schools 55,000 52,100 52,500 50,000 45,000 40,000 35,000
30,000 25,000 20,000 15,000 10,000

6,200

5,300

    5,000

    2002-2003 2003-2004

Title I schools
Title I schools identified for choice

Source: GAO analysis of State data.

As figure 2 shows, Title I schools identified for choice enrolled larger
proportions of minority students and students from low-income families
than other Title I schools.

Figure 2: Demographics of Students Enrolled at Title I Schools Identified
and Not Identified for Choice in School Year 2003-2004

Percent of students 100

90

Percent minority Percent
low-income

Enrollment at Title I schools not identified for choice Enrollment at
Title I schools identified for choice

Source: GAO analysis of data from states and the National Center for
Education Statistics.

About 60 percent of all schools identified for choice were elementary
schools. However, this proportion is smaller than might be expected, given
that 71 percent of all Title I schools are elementary schools. As figure 3
shows, proportionately more middle and high schools were identified for
choice.

Figure 3: Proportion of Schools Identified for Choice at Various Grade
Spans Compared with All Title I Schools in School Year 2003-2004

    Title I schools identified for choice All Title I schools

4%

4%

Other, including ungraded High Middle Elementary

Source: GAO analysis of data from states and the National Center for
Education Statistics.

Most schools identified for choice in school year 2003-2004 were located
in urban and suburban areas. Although 15 percent of all Title I schools
were located in rural areas, only 11 percent of the Title I schools
identified for choice were in rural areas.

Figure 4 shows state variation in the proportion of schools identified for
choice. In the majority of states, 10 percent or fewer of Title I schools
were identified for choice in 2003-2004, but in some states a much larger
percentage was identified. One state, Wyoming, had no schools identified
for choice in 2003-2004. Among the states with relatively few schools
identified for choice were some of the nation's most rural states,
including Maine, Mississippi, and Nebraska, but also some more populous
states such as Florida and Texas. By contrast, in 22 states the percentage
of Title I schools required to offer choice ranged from 11 percent to 48
percent. Among these were several of the nation's most populous states,
including California, Illinois, and New York, but also one of the most
rural-Alaska. Georgia and Hawaii each had 40 percent or more of their
schools identified for choice, higher than any other state. See appendixes
III and IV for state details for each year.

Figure 4: Percentage of Title I Schools Identified for School Choice, by State,
                            in School Year 2003-2004

Source: GAO analysis of state data.

A number of factors contribute to state variations in the proportion of
schools identified for choice, including differences in school populations
and state accountability systems. Under NCLBA, if a school contains a

About 1 Percent of Eligible Students Transferred in the Second Year;
Little Is Known about Their Characteristics

minimum number of students in specific groups-low-income, major racial and
ethnic minorities, students with disabilities, and limited English
proficient-schools are held accountable for the academic outcomes of those
groups, in addition to academic outcomes of the entire school. Large or
diverse schools are likely to have more student groups containing the
state-defined minimum number of students, and consequently have more
performance targets. Because it is harder for schools with many targets to
meet their overall performance goals, states with larger or more diverse
schools may be more likely to have a higher percentage of schools miss
their targets and be identified for choice.

Characteristics of states' accountability systems also contribute to the
variation among states. For example, states use different standards and
set different annual progress rates for reaching 100 percent proficiency.
In addition, some states use smaller minimum student group sizes than
other states. The smaller the size of the group used, the more likely a
school will include additional student groups in accountability,
increasing the number of performance targets the school must meet. 12

About 19,000 students transferred under the NCLBA school choice option in
school year 2002-2003, the first year, and an additional 31,000 students
transferred in the second year. 13 As illustrated in figure 5, this number
transferring in the second year represented about 1 percent of the
students who were eligible. Across states, the number of eligible students
who transferred under NCLBA in 2003-2004 ranged from zero in 6 states to
over 7,000 in one state. States also varied in the extent to which
eligible students exercised the option and transferred. Oregon reported
the highest proportion of eligible students transferring at 17 percent,
followed by Florida with 6 percent. The remaining states had less than 5
percent of eligible students transfer. Further, states with more students
eligible for

12

For more information on differences in state accountability systems, see
GAO-04-734.

13

For the first year, Education reported 45,000 actual transfers based on
data from 48 states, but this figure included one state that subsequently
determined it had overstated its number by about 28,000 students. The
19,000 number we report is based on revised data for 50 states. For the
second year, Cynthia G. Brown, et al, estimated that about 65,000 students
requested transfers and 20,000 actually transferred based on data from 10
states and 68 districts (Choosing Better Schools: A Report on Student
Transfers under the No Child Left Behind Act, The Citizens' Commission on
Civil Rights, (Washington, D.C: May 2004)). Although higher, our 31,000
figure may still understate the actual numbers transferring because we
could not obtain any transfer data for 7 states. See app.V for more
information.

choice under NCLBA did not necessarily have more students use the transfer
option. For example, although Hawaii had more students eligible for
choice, Colorado had about twice as many students transfer. The number of
eligible students transferring in each year for each state is detailed in
appendix V.

Figure 5: Number of Title I Students Eligible for Choice and Number That
Transferred in 42 States, the District of Columbia, and Puerto Rico in
School Year 2003-2004

31,500

Students who transferred in the second year under NCLBA

Students enrolled in Title I schools identified for choice Source: GAO
analysis of state data.

Overall, the proportion of eligible students transferring in the most
rural states 14 was about the same as in other states; however, statewide
data may mask differences within states between rural and nonrural
districts. For example, Kansas, the rural state with the most students
eligible for choice, provided detailed data showing how many student
transfers occurred in each district. About 70 percent of Kansas transfers
were in the

The states are Alaska, Iowa, Kansas, Maine, Mississippi, Montana,
Nebraska, North Dakota, South Dakota, and Vermont, although Montana was
unable to provide data for the second year and was not included in the
analysis. These 10 states represent the most rural states in the country
based on the percentage of their school districts in rural communities,
the percentage of their students attending schools in rural communities,
and the average distance between the states' school districts and the
nearest metropolitan statistical area. See GAO No Child Left Behind Act:
Research on Effective Strategies Need for Small and Isolated Rural
Districts, GAO-04-909 (Washington, D.C: Sept. 23, 2004).

state's three largest districts-Wichita, Shawnee Mission, and Kansas City
-although about half of the students eligible for NCLBA transfers were
located in those districts. Officials in several rural states reported
that rural districts faced unique challenges implementing NCLBA choice. In
some rural districts, although students were eligible for choice, no
transfers took place because there were no other schools in the districts
that could be offered as transfer options. 15 Where transfer options were
available, sometimes the distances between schools made transfers
difficult.

In the 41 states that could provide student transfer data and had schools
identified for choice in both years, the total number of transferring
students rose by about 85 percent. 16 This increase was driven by several
states that had substantial increases, such as New York, New Jersey, and
South Carolina. However, 8 states reported declining numbers of transfers,
and 6 of these states also reported fewer schools identified for choice in
the second year, while 2 reported increases. 17

Little is known about the demographic characteristics and academic
performance of students who transferred under NCLBA school choice in
either year or reasons why parents accept or do not accept transfer
opportunities. Although Education has requested state data on the number
of students transferring each year, it has not collected data on the
characteristics or academic performance of transferring students.
Education officials told us that they have contracted for a major,
multifaceted study of NCLBA that will examine key areas of implementation,

15

Where districts lack other schools to offer as transfer options, districts
are required, to the extent practicable, to make transfer agreements with
neighboring districts, but we are not aware of any locations where such
interdistrict transfers have occurred.

16

The increase may be understated, because second year data for some states
were not available statewide and instead represents numbers for large
districts in these states. Florida, Vermont, and Wyoming were not included
among the 41 states, because they had no schools identified for choice the
first year, although in the second year, Florida accounted for about 1,800
transfers. See app. V for details.

17

The 8 states were Alabama, Illinois, Indiana, Iowa, North Dakota, Texas,
Utah and Missouri. Illinois and Missouri reported increases in the number
of schools identified the second year. Not included here are Massachusetts
and Washington because the transfers for the second year are based only on
the numbers in large districts, and thus may be understated. See app. III
and IV for the number of schools identified in each year and app. V for
the students transferring.

including school choice. 18 Two parts of the study relating to school
choice are descriptive: one is a descriptive comparison of the
demographics of students who choose to transfer and those who do not. A
second part examines reasons that parents give for their decisions about
whether or not to apply for transfers. A third part of the study, still
under design, will examine student achievement outcomes. This effort would
examine the academic outcomes over time of transferring students in a
sample of districts, but this portion of the study is not fully developed.
For instance, officials said they are still exploring several
possibilities for study methodology and whether demographic
characteristics of these students will be included in the achievement
analysis.

Our analyses of 2003-2004 demographic and academic data that we were able
to obtain from one district we visited showed diversity in transferring
students. Of the students who transferred, 53 percent were male, 62
percent were minorities representing all the major racial and ethnic
groups, and 82 percent were from low-income families as measured by their
eligibility for the free or reduced-price school lunch program. In
addition, 10 percent of these transferring students were English language
learners and 14 percent were enrolled in special education. 19 In general,
proportionately fewer minority and low-income students transferred,
compared with students who were eligible but did not transfer, as shown in
table 1. Our analysis of available student performance data from state
reading and math assessments showed little difference between transferring
students and those not transferring. The proportion of students who met
the standards was about the same for each group. 20

18

Education refers to this study as the National Longitudinal Study of No
Child Left Behind. Data will be collected for school years 2004-05 and
2006-07, with a report expected in September 2007.

19

Demographic categories used are not mutually exclusive.

20

Assessment data were available for about one-quarter of the students in
each group, because assessments were administered to students in only one
grade at each school.

Page 17 GAO-05-7 No Child Left Behind Act

Table 1: Students Eligible but Not Transferring from Elementary and Middle
Schools Identified for Choice Compared with Students Transferring in
2003-2004 in One District Visited

                      Percent minority Percent low income

             School 1 Students eligible but not transferring 76 93

                       Transferring students (N=15) 47 87

             School 2 Students eligible but not transferring 72 92

                       Transferring students (N=5) 40 100

             School 3 Students eligible but not transferring 63 83

                       Transferring students (N=75) 57 80

             School 4 Students eligible but not transferring 73 84

                       Transferring students (N=52) 63 81

             School 5 Students eligible but not transferring 59 78

                       Transferring students (N=58) 69 83

               Total Students eligible but not transferring 68 85

                      Transferring students (N=220) 62 82

  Implementation of Choice Was Challenging in Selected Districts

Source: School district officials.

Note: We were unable to associate 15 of the 220 transferring students with
the school from which they transferred.

Compared with students in the schools into which they transferred,
however, transferring students were somewhat lower performing on state
assessments. 21 About 33 percent of transferring students met state
reading standards, while 43 percent of the other students in the receiving
schools met these standards. Similarly, about 20 percent of transferring
students met state math standards, while 34 percent of the other students
in receiving schools met state math standards. Transfer students were also
more often from a minority background. About 62 percent of the
transferring students were minorities, but about 52 percent of the
students in receiving schools were minorities.

Officials in most of the 8 districts we visited mentioned that they
supported the NCLBA focus on improved student performance and
accountability; however, they had difficulties providing school choice,
primarily because of tight timeframes and insufficient capacity. To try to
get notices out to parents before school started, most districts took a
risk and acted on preliminary data on school performance that they
received

21

Based on 2002-03 reading and math assessments.

Short Timeframes Created Challenges

from the state in late summer because final data were not available.
Parents of eligible students were presented at least two schools as
transfer options, but many of these alternatives were similar to the
schools students were currently attending. Some districts were not able to
accommodate all transfer requests because the demand for some schools
exceeded their capacity. Districts employed a variety of strategies to
provide transportation to transferring students, including school buses,
public transportation, and cash stipends.

Although the law requires districts to notify parents of the choice option
by the start of the school year, 7 of 8 districts we visited did not
receive final results of school performance for the most recent year from
the state in time to meet the requirement. Consequently, many used
preliminary data to identify schools for choice. Akron was the only
district that had final results from the state when notices were sent to
parents. Four districts used preliminary data to identify which schools
had to offer choice and notified parents before school started. A fifth
district also used preliminary data but did not receive the data until
after school started. Using preliminary data can put districts at risk of
incorrectly identifying schools as having to offer choice and consequently
misinforming parents. 22 One district included language in the
notification letter to parents explaining that the transfer offer could be
revoked if final determinations by the state were different. Table 2 shows
key testing and notification dates in 6 of the districts we visited.

22

For example, Mississippi state officials told us that 3 of 10 schools
identified for choice in 2003-04 based on preliminary data were later
found to have met their goals. Mississippi requires districts to provide
choice based on preliminary results and to honor its commitment to choice
even when the final data yield different results. However, the school in
N. Panola was not among the three incorrectly identified.

Page 19 GAO-05-7 No Child Left Behind Act

Table 2: Milestone Dates for Implementation of Choice for School Year 2003-2004
                          in Six Districts We Visited

Preliminary results Final results from the in from the state on District
notices state on schools' yearly Month tests schools' yearly sent to
parents Start of school performance (2003 or School district given (2003)
performance (2003) (2003) year 2003-2004 later as noted)

            Akron, Ohio March June (early) July 25 August 27 July 21

       Chicago, Ill. March/April July 28 August 9 September 2 December 19

          Elgin, Ill. March/April July 30 August 6 August 25 June 2004

    Pittsburgh, Pa. March/April August 5 August 28 September 7 September 23

       Tacoma, Wash. April/May July 28 August 12 September 3 September 17

    North Panola, Miss.a April/May August 8 September 3 August 5 November 21

Source: School district officials.

a

North Panola, a rural district in Mississippi, did not offer school
choice, because the middle school identified was the only one in the
district. Instead, the district notified parents and offered supplemental
educational services.

The remaining 2 districts we visited, Memphis and Fresno, did not use
preliminary data from the most recent testing period, but rather used data
from the previous year to determine the schools that would have to offer
choice. 23 Officials said they were aware this delay was not in accord
with Education guidance but took this action to combine NCLBA choice with
their voluntary choice programs, which permit all students to request
transfers in the spring. Memphis officials said that they planned to
change their procedures and offer school choice twice in 2004-first in the
spring, during the open enrollment process, for schools that they already
know must offer choice and again in the fall when they receive the results
of the spring 2004 assessments. Fresno officials did not indicate that
they would be changing their procedures.

Given the tasks that districts must complete to offer school choice before
school starts, officials expressed concerns that little could be done to
mitigate these timeframe problems. Districts must first administer state
tests in the spring, which are sometimes sent to contractors to be scored.
Next, after receiving the preliminary test results from the state,
districts assess the scores to verify the accuracy of the data, use these
data to identify schools likely to be required to offer choice, and notify
schools. Schools may appeal this decision to the state. Only after
reviewing such

23

Tennessee did not provide Memphis with preliminary data prior to the start
of school. California provided incomplete data prior to the start of
school and final results in February 2004.

Page 20 GAO-05-7 No Child Left Behind Act

appeals do states release final determinations of which schools are
required to offer choice. Most districts we visited did not have the final
performance data before school started in the fall. Figure 6 shows the
timeline of events in one school district we visited; similar patterns
occurred in most others.

               Figure 6: Timeline for Tacoma Public Schools, 2003

    April

4/21-5/9

Testing period for school year 2002-2003.

May

5/15

Test booklets transported from district to vendor in Iowa for scoring.

    July

7/28

District receives preliminary yearly performance results for each school
from the state.

7/28-8/8

District reviews preliminary results, verifies data, and determines yearly
performance for each school.

August

8/4-8/8

District notifies schools that must offer choice and schools that will
receive transfer students.

8/12

Notice to parents about school choice and the schools that they may
select.

8/18-9/12

District notifies schools about transfers, sets final staffing for school,
arranges transportation, and finalizes student transfers.

September

9/3

School year starts.

9/12

Deadline for parents to notify district about school choice.

9/27

District receives notice of final yearly performance results for each
school from the state.

Source: Tacoma Public Schools.

The compressed timeframe for making school status determinations and
implementing the choice option left parents little time to make transfer
decisions, and district and school officials expressed concerns that
parents did not have adequate time to make an informed decision. In most
of the districts we visited, parents had 3 weeks or less to make their
transfer decisions. In addition, in districts and schools with highly
mobile populations, reaching parents can be time-consuming. Akron and
Memphis officials told us that many letters notifying parents of the
transfer option were sent to addresses found to be incorrect. To ensure
that parents had a greater chance of learning about school choice, some
districts used a variety of additional notification strategies-fliers,
newspaper articles, postings to the district Web site, and public
meetings. In addition, they provided parents several ways to communicate
their desires such as through the mail, by telephone, or going to the
district in person.

Officials in some districts also expressed concern that the information
provided to parents was not always clear and that it may not have been
sufficient for them to base their decisions. 24 Letters sent to parents
generally explained what it meant to be identified for choice, gave the
reasons for the identification, described the process for applying for
transfer, and listed the transfer school options. However, little
information was provided about the transfer schools. In some districts,
school officials were concerned that the wording of the letters may have
been confusing. They said that parents did not always understand the
meaning of the school choice option as explained in the letter and needed
more time to consult with district or school officials. For example,
officials in 2 districts told us that some parents misunderstood the
letter and believed that they were required to transfer their child to
another school. Other school officials talked about the need for parents
to have additional information about specialized services and
instructional support that certain schools provide in order to understand
the educational implications of their decisions. Officials in one district
told us that some parents who chose a transfer school later changed their
minds when they found that student support services their child had
received at their Title I school, such as extended day programs and after
school tutoring made possible by Title I funds, were not available at the
non-Title I transfer school.

Schools also faced challenges in implementing choice within the
timeframes, particularly in adjusting staffing and scheduling, when they
learned shortly before the start of school that they would be receiving
students under the NCLBA school choice program. For example, a Tacoma
middle school principal said that she faced a variety of challenges when
she was notified a month before school started that the school was to
receive NCLBA transfer students. Based on spring predictions of the
school's student population and student needs, she had released six
teachers. However, when notified the school was receiving 57 NCLBA
transfers in the fall, she had to quickly hire two new teachers and
reconfigure the schedule to include more remedial classes to accommodate
the learning needs of the transferring students. In addition, school
officials did not receive records for some students from the schools

24Researchers in Colorado and Massachusetts echoed the districts'
concerns. See Pamela Benigno, No Child Left Behind Mandates School Choice:
Colorado's First Year (Independence Institute: June 2003)
http://www.independenceinstitute.org/edpublications.aspx downloaded
November 5, 2004. See also William Howell, "Fumbling for an Exit Key:
Parents, Choice, and the Future of NCLB" in Frederick M. Hess and Chester
E. Finn, Jr, editors, Leaving No Child Behind? Options for Kids in Failing
Schools (New York, N.Y.: Palgrave Macmillan, Oct. 2004) 161190.

Most Districts Offered Parents a Number of Schools as Choices, but Many of
These Alternatives Were Similar to the Student's Current School

they left until after school started, and some students were initially
placed in the wrong classes.

Whenever possible, districts offered each parent at least 2 schools as
transfer options, as required by federal regulations, but some districts
offered more than 30 schools. The locations varied by district. Table 3
shows the number and location of elementary schools offered in the
districts we visited. Some districts offered schools based on geographic
location within the district and some offered schools districtwide. For
example, as table 3 shows, students in Memphis attending one of the 40
schools identified for choice selected from among 3-10 transfer schools
that were in the same general area of the city, while students in each of
the 6 schools identified in Akron selected from a group of 33 schools
spread across the district. Elementary schools offered as transfer options
were more commonly selected for their proximity to sending schools than
middle and high schools, which were generally offered districtwide.
Although not shown in table 3, parents generally were offered fewer
transfer options for middle and high school students, because districts
tend to have fewer middle and high schools than elementary schools.

Table 3: Transfer Options Offered Parents of Elementary School Students in Seven
                              Districts We Visited

Number of elementary schools offered as Location: Number of elementary
transfer options to Number of schools Districtwide or schools required to
offer parents for each school parents instructed to geographic area
District choice required to offer choice select within district

                        Akron, Ohio 6 33 1 Districtwide

                      Chicago, Ill. 305 37 2 Districtwide

                        Elgin, Ill. 6 16 3 Districtwide

                        Fresno, Calif. 23 2 2 Geographic

                      Memphis, Tenn. 40a 3-10 2 Geographic

                     Pittsburgh, Pa. 11 11-19 2 Geographic

                         Tacoma, Wash. 2 2 2 Geographic

Source: School district officials.

a

Memphis schools are those the state identified for choice in school year
2002-03.

Many schools that districts offered as transfer options had not met state
performance goals in the prior year, and some were at risk themselves of
having to offer choice in the following year. Among the seven districts
that offered transfers, all had some schools offered as choices that had
not met the state's yearly performance goals, based on the spring 2003
assessments. Table 4 provides more detail on the status of schools offered
as transfer options by district. Because many of these schools were Title
I schools and, therefore, subject to NCLBA requirements, those that did
not meet their yearly performance goals for a second consecutive year
would have to offer school choice the following year. For example, in
Memphis 37 Title 1 schools were offered as transfer options, and 29 of
these had not met yearly performance goals based on spring 2003 tests.
Some schools offered were not Title I schools and, therefore, were not
required to offer transfers, regardless of the performance of the school.
Overall, as shown in Table 4 for the districts we visited, from 21 to 73
percent of all schools offered, Title I and non-Title I, had met yearly
performance goals.

 Table 4: Title I Status and Performance of Schools Offered as Transfer Options
                           for 2003-2004, by District

Number of schools offered as Number and percentage of all transfer options
that did not meet schools offered as transfer options Number of schools
offered yearly performance goals based on that met yearly performance
goals as transfer options spring 2003 tests based on spring 2003 tests

Other Number Title I Percentage Title I District Title I schools Title I
Other schools and other and other

                          Akron, Ohio 32 5 11 3 23 62

                        Chicago, Ill. 21 19 12 11 16 40

                           Elgin, Ill. 0 16 0 6 10 62

                         Fresno, Calif. 15 3 10 1 7 39

                        Memphis, Tenn. 37 10 29 6 10a 21

                        Pittsburgh, Pa. 51 0 14 0 37 73

a

Tacoma, Wash. 4 4 1 3

Source: GAO analysis of school data from state or district education
agency Web sites.

a

In Memphis and Tacoma numbers do not add to totals because test data are
not available for schools that were new as of 2003-2004 and did not have a
testing history; in Tacoma there was one such school; in Memphis, there
were two. In addition, there was one school in Chicago that did not have
test data for 2002-2003.

Officials from large urban districts such as Fresno and Memphis said that
they would have few schools to offer as choices if they did not offer
Title I schools that had failed to meet the performance goals for only one
year. Officials in some districts expressed concerns that, as the bar for
meeting yearly performance goals is raised, more schools would fail and
few schools would be available as transfer options. In these districts,
over 80 percent of schools received Title I funds and many more students
could become eligible for transfer under NCLBA.

In districts such as Chicago, Fresno, and Memphis with high proportions of
Title I schools, the majority of the schools offered as transfer options
were often demographically similar to those attended by students eligible
for transfer. Specifically, the schools offered as transfer options served
many poor students and had high minority populations. As shown in table 5,
for example, 34 of Fresno's 39 schools required to offer choice-about
seven-eighths-had poverty rates that exceeded 75 percent, as did over half
of the 18 schools offered as transfer options.

  Table 5: Poverty and Minority Rates of Schools Required to Offer Choice and
                 Schools Offered as Transfer Options in Fresno

                               Poverty Minorities

Number of schools Number of schools School enrollment Number of schools
offered as transfer Number of schools offered as transfer percentage
required to offer choice options required to offer choice options

                                  0-25% 03 0 0

                                 26-50% 14 0 3

                                 51-75% 41 1 7

                               76-100% 34 10 38 8

Source: GAO analysis of school data from state or district education
agency Web sites.

In contrast, other districts that we visited tended to offer more transfer
options that differed demographically from the schools required to offer
choice. As shown in table 6, for example, 7 of Akron's 8 schools required
to offer school choice had poverty rates that exceeded 75 percent, but
less than one-third of schools offered as transfer options had such rates.
See appendix VI for poverty and minority rates of schools in seven
districts that we visited.

  Table 6: Poverty and Minority Rates of Schools Required to Offer Choice and
                  Schools Offered as Transfer Options in Akron

                               Poverty Minorities

Number of schools School enrollment Number of schools offered as transfer
percentage required to offer choice options

    Number of schools Number of schools required to offer offered as transfer
                                                               choice options

0-25% 01 0

26-50% 09 0

51-75% 1 16 4

76-100% 7 11 4

Source: GAO analysis of school data from state or district education
agency websites.

Several Districts We Visited Cited Lack of Sufficient Capacity as Limiting
the Number of Transfers

Despite the fact that all districts offered parents a choice of schools,
officials in four districts told us that they were unable to accommodate
some requests for transfers because of constraints on classroom capacity,
as shown table 7. In two districts in Illinois-Elgin, and
Chicago-officials said that they believed that state law did not allow
their districts to offer choice under NCLBA if it led to overcrowding in
schools. 25 Akron officials told us that they were seeking clarification
from the state about whether any transfers in their district would be
prohibited by Ohio state law. Memphis officials told us that demand
exceeded the capacity at certain schools that were already overcrowded,
and use of portables to expand capacity was unrealistic because of the
expense and lack of sufficient space on school campuses.

Table 7: Number of Students Unable to Transfer Due to Capacity Limitations
in 2003-2004

    Number of Number of students that students that applied for a transferred
                                          to District transfer another school

Number of

students

        unable to Number of transfer students not because of transferring for
                                                       capacity other reasons

Akron, Ohio 114 76 19

Chicago, Ill. 19,000 548 17,903

Elgin, Ill. 113 60 14

                          Fresno, Calif 465 101 0 364

Memphis,
Tenn. 830 388a 442

Pittsburgh, Pa. 161 151 0

Tacoma, Wash. 253 253 0

Source: School district officials; Chicago data from The Center for
Education Policy, From the Capitol to the Classroom, January 2003.

a

Memphis approved 388 student transfers but did not collect data on the
number of students that actually transferred.

In some districts with capacity constraints, open enrollment programs
could limit the ability of students to transfer under NCLBA. In all but 2
of the districts we visited, school choice was available to all students
through open enrollment programs. These programs offered students the
chance to apply for transfers, typically during the winter and spring
months, and, in

25

Elgin and Chicago cited an Illinois law, effective July 1, 2002, that
limits transfer options when the receiving school would exceed capacity as
a result of the transfer.

Page 26 GAO-05-7 No Child Left Behind Act

Districts Used Different Strategies to Meet Transportation Requirement

several districts, allowed transferring students to learn which school
they would attend before the end of that school year. In contrast,
students in most districts transferring under NCLBA did not know about
their opportunity until just before the next school year started. Unless
these districts took special care, schools could be filled to capacity
with transfers approved under the open enrollment program before NCLBA
students had the opportunity to apply. To avoid this situation, Akron gave
NCLBA transfers priority and delayed decisions on requests for transfers
under its open enrollment program until the decisions on NCLBA transfers
had been made.

Officials in Fresno, Pittsburgh, and Tacoma reported that they had not yet
experienced problems with capacity because few students had transferred.
However, some officials expressed concern that capacity could pose a
challenge in their district in the future. Specifically, officials in
Fresno, Memphis, and Tacoma noted that if more schools were required to
offer choice in the future, the number of students eligible to transfer
could increase and capacity could become a problem.

The districts we visited arranged and paid for the transportation of
students who transferred, as required under NCLBA, but did so in a variety
of ways as allowed under the law. For instance, some districts provided
school buses, while others paid for public transportation or provided cash
stipends to cover public or private transportation. In 5 of the 7
districts, school buses picked up elementary students who lived more than
1-2 miles from their schools. For middle and high school students, some
districts paid for public transportation by giving students passes or
tokens. Finally, Akron gave parents a $170 transportation subsidy at the
end of the school year in which students transferred to subsidize the
costs of public transit or defray the gasoline costs of driving their
child to school.

In providing transportation, districts used relatively little of the
funding that was required to be set aside for school choice transportation
and for supplemental services because few students transferred. In
2003-2004, the estimated expenditures for transportation represented less
than 7 percent of the set-aside funds in all but one district we visited.
As shown in table 8, the proportions ranged from less than 1 percent in
Akron to about 25 percent in Elgin. Most district officials did not expect
to spend the full amount that had to be set-aside for the combined costs
of choice-related transportation and supplemental educational services. 26
However, some district officials said they anticipated that as more
schools have to offer school choice and more students become eligible to
transfer, it is likely that transportation expenditures will increase.

Table 8: Transportation Allocations and Projected Expenses for School Year 2003-2004,
by District 
                                Required Projected           Number of     Percent of 
                           set-aside for dollars                            set-aside 
            Total Title I transportation       spent on students              used to 
                                     and                provided              provide 
District       allocation   supplemental transportation transportation transportation 
                                services                               
Akron, Ohio   $12,463,428     $2,492,686        $13,416             76           0.54 
Chicago,    $240,829,945a   $48,165,989a     $1,000,000            548           2.08 
Ill.                                                                   
Elgin, Ill.    $2,323,966       $464,793       $117,934             51          25.37 
Fresno,       $40,225,930     $8,045,186       $385,900            101           4.80 
Calif.                                                                 
Memphis,      $38,653,432     $7,730,686       $526,070              b           6.80 
Tenn.                                                                  
Pittsburgh,   $17,014,258     $3,402,852       $103,600            136           3.04 
Pa.                                                                    
Tacoma,        $8,854,322     $1,770,864        $72,070            253           4.07 
Wash.                                                                  

  Education Issued Extensive Guidance on Choice, but Questions Remain as
  Implementation Proceeds

Source: School district officials except where otherwise shown.

a

Department of Education estimated allocation for fiscal year 2003.

b

Memphis provided information on the number of transfers approved but did
not collect data on the number of students that actually transferred.

Education issued final regulations and guidance on school choice within a
year of NCLBA enactment, but did so after districts had begun their first
year of implementation, and some issues remain unclear. Extensive
additional guidance and technical assistance in the form of policy
letters, training tools, presentations at conferences, and a handbook on
promising practices became available at various times throughout the first
and second years of implementation. While district officials we visited
generally had access to Education's guidance, questions concerning the
implementation of school choice remained, as might be expected in initial
years of implementation. For example, there were "how to" questions about
ways to offer choice when building capacity is limited. There were also
"what if" questions involving issues that may arise as NCLBA
implementation progresses, such as districts' use of Title I funds for
transportation when students choose to remain at a school to which they
have transferred if that school subsequently does not meet its yearly

26

Generally districts are permitted to carry over up to 15 percent of their
Title I funds for use in the following year.

Page 28 GAO-05-7 No Child Left Behind Act

In the First Year, Districts Implemented Choice Using Preliminary Guidance
Released through Letters to Superintendents

performance goals and is itself required to offer choice. Some of these
questions have been addressed in guidance but others remain.

Responding to the need to get information out quickly, Education issued
preliminary guidance in June 2002 before the start of the first school
year. The information provided, however, was not always clear or complete.
The preliminary guidance was sent out in the form of a "Dear Colleague"
letter directly to school district superintendents as well as state
education agency officials. In the letter, Education acknowledged that its
preliminary guidance was necessarily brief and not as comprehensive as
guidance that would be forthcoming. The letter highlighted key topics such
as notices to parents, designation of sending and receiving schools,
prioritization of students, capacity and transportation. The letter stated
that choice had to be provided, unless prohibited by state law, to all
eligible students, "subject to health and safety code requirements." Some
district officials believed this language allowed them to limit the number
of transfers based on state or local health and safety codes or classroom
size requirements.

Subsequent guidance provided additional information about Education's
position on capacity and other issues. Final regulations and draft
guidance on choice were issued in December 2002, after the start of the
first school year. The final regulations applied to all aspects of Title
I, while the draft guidance applied specifically to school choice and was
characterized as "non-regulatory" guidance. The final regulations
clarified some key information and the December guidance added extensively
to material in the June 2002 letter. For example, in response to numerous
requests for clarification of its language on capacity, Education's
regulations made it explicit that districts were required to accommodate
all transfer requests while complying with all applicable state and local
health and safety codes as well as classroom size requirements. Districts
had to offer all students at schools identified for choice the option of
transferring and could not use lack of capacity as a reason to deny
students this option. The regulations explained that state law exempts
districts from offering choice only if the state law prohibits choice
through restrictions on public school assignments or the transfer of
students from one public school to another public school. The December
guidance went further to help clarify Education's position by contrasting
its regulations on capacity before and after enactment of NCLBA and
providing an explanation for the differences. Because there had been no
mention of capacity in NCLBA and some district representatives were
uncertain about the meaning of the preliminary guidance in the June
letter, the final regulations and December guidance represented an
important clarification of Education's official

Additional Guidance and Technical Assistance Provided in the Second Year
of Implementation

position on the issue. In these December documents, Education also
suggested ways that districts might expand capacity, for example, by
adding classes and additional teachers, in order to be able to offer
choice to students while adhering to state classroom size requirements and
health and safety codes.

In the second year of NCLBA implementation, Education updated and expanded
its draft guidance and published a handbook on promising practices in the
provision of school choice. Education also provided additional assistance
in the form of training materials, presentations at various conferences
and a toll-free hotline for district superintendents in both the first and
second years. See table 9 for a chronology of the various types of
guidance on choice issued by Education.

Table 9: Dates, Type, and Title of Guidance or Technical Assistance Provided by
                     Education on School Choice under NCLBA

                   Year Date Guidance, information, or action

2002 February 15 Policy letter to Chief State School Officers on timetable
for required actions in school year 2002-2003, including provision of
choice. http://www.ed.gov/policy/elsec/guid/secletter/020215.html

June 14 Preliminary guidance: Letter to District Superintendents and Chief
State School Officers on school choice and supplemental educational
services. http://www.ed.gov/policy/elsec/guid/secletter/020614.html

September Summary of law: NCLB: A Desktop Reference.

http://www.ed.gov/admins/lead/account/nclbreference/index.html

December 2 Final regulations to implement Title I under NCLBA.

http://www.ed.gov/legislation/FedRegister/finrule/2003-4/120903a.html

    December 4 Draft non-regulatory guidance: Public School Choice. (23 pp.)

2003 February Training materials: Choice Provisions of NCLBA (power point
                                    slides).

 March 24 Draft non-regulatory guidance: Impact of the New Title I Requirements
                          on Charter Schools. (12 pp.)

 Summer Letters to some states on additional steps needed to comply with choice
                            and other requirements.

August 21 Non-regulatory guidance: Local Educational Agency (LEA)
Identification and Selection of School Attendance Areas and Schools and
Allocation of Title I Funds to Those Areas and Schools.

http://www.ed.gov/policy/elsec/guid/edpicks.jhtml?src=fp

August 28 Policy letter to Chief State School Officers allowing carry over
of Title I funds set aside for school choice transportation.
http://www.ed.gov/policy/elsec/guid/secletter/030828.html

December 22 Consolidated State Performance Reports due, including data on
                           implementation on choice.

2004 January 7 Non-regulatory guidance: LEA and School Improvement.

http://www.ed.gov/policy/elsec/guid/list.jhtml?page=12&size=10&sort=date&desc=show

January Letter to school superintendents regarding hotline.

http://www.ed.gov/policy/gen/guid/secletter/040101.html

    February 6 Draft non-regulatory guidance: Public School Choice. (29 pp.)

http://www.ed.gov/policy/elsec/guid/edpicks.jhtml?src=fp

Year Date    Guidance, information, or action                                          
     May     Technical Assistance Handbook: Innovations in Education: Creating Strong  
             District School Choice Programs.                                          
             http://ed.gov/admins/comm/choice/choiceprograms/index.html                
     July    Non-regulatory guidance: Impact of the New Title I Requirements on        
             Charter Schools. (16 pp.)                                                 
             http://www.ed.gov/policy/elsec/guid/edpicks.jhtml?src=fp                  
     August  Policy letter to Chief State School Officers on how to calculate the      
     18      portion of transportation expenses related to choice when no additional   
             costs are incurred.                                                       
             http://www.ed.gov/policy/elsec/guid/stateletters/choice/choice081804.html 
     October Policy letter to Chief State School Officers on reallocating reserved     
       12    Title I funds, use of set-asides, and handling of schools that change     
             Title I status.                                                           
             http://www.ed.gov/policy/elsec/guid/stateletters/uofcssos.html            

Source: Education official s and Education's Web site www.ed.gov.

The February 2004 guidance was developed in response to state questions,
often made at the request of districts, for further clarification of
several issues. Although Education's primary relationship was with state
agencies, Education officials also made appearances at conferences
attended by district and school officials and made a concerted effort to
alert local education officials and other interested parties when it
released its latest guidance through electronic mailing lists to
subscribers and through its Web site. In many of the districts that we
visited, officials told us that they had access to Education's guidance,
either directly from Education's Web site, from the state agency or from a
national organization representing their interests, such as the Council of
the Great City Schools.

One of the major changes in the February 2004 guidance was a list of 10
ways that districts might increase capacity in order to provide school
choice for all eligible students requesting transfers. The guidance
suggested that districts "employ creativity and ingenuity" in developing
ways to expand capacity, such as setting up "virtual" schools,
reallocating portable classrooms, or creating "schools within schools"
that would be new, distinct schools, with separate faculty, within the
physical sites of schools required to provide choice. Some district
officials we interviewed expressed their reservations about the
feasibility of Education's suggestions on how to develop the needed
capacity, in part because of concerns about the costs of implementing the
suggestions.

The May 2004 technical assistance handbook describes promising practices
in several key areas, including capacity, that have been employed to
implement choice in 5 school districts. The most detailed and thorough
description covers strategies that these districts have used to deal with
parental notification and decision-making, but other chapters deal with
capacity and transportation, support for sending and receiving schools,
use of databases and surveys for planning, and factors leading to

Complexity of Implementing Choice Raises Issues Not Addressed in Guidance

success. With respect to capacity, the handbook lists the actions that
certain districts have taken but does not describe them in detail. For
example, the handbook states that Milwaukee established a special team
that spent 2 months assessing available capacity; Miami-Dade used
portables; Denver used teacher lounges and resource rooms as classrooms,
and Mesa created new schools. Individuals interested in more details could
contact the districts involved to find out more about timetables and costs
of these various strategies at the postal or Web site addresses listed in
the handbook. Education officials told us that within the first 6 months
following its publication they had sent out over 16,000 copies of handbook
to state officials and to organizations representing local education
officials, such as the National School Boards Association and the National
Alliance of Black School Educators.

Numerous questions concerning current or future implementation issues were
raised during our visits with district and school officials that were not
answered clearly in Education's February 2004 guidance on choice. The
issues involved how best to handle, within the context of federal
regulations and guidance, certain complex situations involving timetables,
schools receiving transfers, transportation, and capacity. With respect to
timetables for parental notification, district officials we visited in two
states were concerned about the accuracy of preliminary state
determinations of the schools that made or did not make yearly performance
goals. Because NCLBA required that they offer choice by the start of
school, the districts were acting on preliminary but possibly inaccurate
determinations made by states and were uncertain if there were any
circumstances that would permit them to delay choice until they received
final determinations. Basically, the questions involved how best to
mitigate the risks for all involved-districts, schools, parents and
students. Notices sent to parents had included warnings-either that the
school status might change within a month or that the offer of choice
might be withdrawn. However, there was interest in finding better ways to
deal with the uncertainty involved, including what steps should be taken
on behalf of parents and students when transfers have occurred either into
or out of schools that were designated incorrectly.

Even where school designations were known, planning for future
contingencies raised a number of questions about schools offered as
transfer options and about transportation arrangements. District officials
explained that they are operating in a dynamic environment where school
performance can change from one year to the next and their status as Title
I or not Title I schools can also change. 27 Officials in one district we
visited asked for confirmation that, if they could not find reasonable
alternatives, they would be permitted to offer as transfer options those
schools that had missed their performance goals for one or more years, as
long as they were not Title I schools. Confirmation could be inferred from
the February 2004 non-regulatory guidance on choice, but was not
clear-cut. Considerations of schools offered as transfer options led into
further questions about transportation provided for students who
transferred, for example if students could continue to receive Title
I-funded transportation if they had transferred into a Title I school
offered as a transfer option that later missed its yearly performance
goals for 2 consecutive years.

In several districts we visited, we found that officials were struggling
to find practical and realistic ways to offer choice when building
capacity, budgets and timeframes were limited. Some of these officials had
studied the suggestions offered in Education's February guidance but
considered creation of virtual or charter schools to be long-term projects
that could not provide capacity in time to meet short deadlines. Other
officials commented that they did not know what steps to take to create
"schools within schools," as suggested, or how to estimate the costs. Cost
considerations were a major issue in several districts where capacity
constraints had limited the number of transfers under NCLBA.

Education officials told us in November 2004 that they believed that the
guidance and technical assistance that they had provided thus far was
sufficient to meet the needs identified by states and district officials
with whom they were in contact. At that time, they had no specific plans
to issue further guidance or provide additional technical assistance on
these issues. However officials added that policy letters will continue to
be issued as needed in response to questions raised by states that have
not been addressed elsewhere.

Title I funds must be provided to schools with high poverty rates. Schools
that previously received no Title I funds may become eligible with changes
in their student enrollment or may lose their eligibility. Also, districts
have the option to extend Title I funding to other eligible schools with
lower poverty rates, or having done so in the past, to alter their
decision. For example, Memphis officials told us that they expanded the
number of schools receiving Title I funds in school year 2004-2005.

Page 33 GAO-05-7 No Child Left Behind Act

                                  Conclusions

Finally, some issues that district officials raised during our site visits
were not ones for which Education could provide guidance. These issues
involved distinctions between federal and state requirements that
individual states would be expected to resolve for their districts. For
example officials in one district sought clarification as to whether or
not their state applied the NCLBA interventions both to schools not
receiving Title I funds and to Title I schools. Officials in another
district were unsure about whether their state exempted schools that did
not receive Title I from some NCLBA interventions, such as school choice
and supplemental services, but not from other interventions, such as
corrective action and restructuring after repeatedly missing yearly
performance goals.

NCLBA is an important and complex piece of legislation, and as
implementation proceeds, Education will need to continue to help states
and districts address the many issues they face in providing school
choice. State and districts officials, although positive about the intent
of NCLBA, nevertheless identified a variety of challenges in implementing
the law. Half of the districts we visited did not grant as many transfers
as were requested because of constraints on the building capacities at
many of their schools. Difficulties related to building capacity are
unlikely to diminish in the future, and could become more pronounced if
the number of students eligible to transfer increases and the number of
schools available as potential transfer options decreases. In the first 2
years under NCLBA, Education data show that the number of schools not
making their yearly performance goals increased. Several state officials
suggest that this trend will continue. Consequently, it is likely that
more schools will be identified for choice, which would increase the
number of students eligible for transfer while decreasing the pool of
possible transfer schools. Further, new challenges may arise if the
schools to which students have transferred in the early years of NCLBA do
not themselves make their yearly performance goals.

In addition, our work raised questions about how well-informed parents are
about the school choice option. In the second year of NCLBA, about 1
percent of eligible students transferred, and without more information on
the reasons parents do or do not take advantage of the transfer option,
policy makers and school officials may miss opportunities to better serve
parents and students through the choice option. In addition, it is unclear
whether or not parents are receiving adequate information to make fully
informed transfer decisions. It may be that parents do not fully
understand why their child's school was identified for choice or the
educational services available in the transfer school. Education's
longitudinal study of

  Recommendations for Executive Action

o

NCLBA will address some of these questions. The parental survey will
explore the reasons parents do or do not exercise the transfer option and
the circumstances that facilitate or hinder their decisions. The result of
the survey may give Education and policymakers insight into the reasons
behind the numbers of students who have transferred, as well as assist
school officials in assuring that parents are aware of the option. In
addition, the technical assistance handbook that Education issued in May
2004 provides some suggestions that may help districts improve their
communications with parents, but this information is based on the
experiences of only a small number of school districts. Districts may need
additional help in various ways, including how to provide information on
choice options that can be easily understood by parents and how to provide
additional information parents need to make an informed decision.

Finally, little is known about transferring students or the effects of
transfers, but Education's plans for its major study of NCLBA are
promising. As planned, the study should provide insight into the
demographic characteristics of students transferring under the school
choice provision and the extent to which the lowest achieving students
from low-income families, identified for priority consideration under the
law, are exercising the transfer option. Equally important is Education's
proposed analysis of how transfers may affect the subsequent academic
performance of students who change schools under the choice provision of
NCLBA. This portion of Education's proposed study is critical to informing
policy makers and school officials about whether or not the school choice
option is achieving its intended outcome of improving student achievement;
however, this part of the study is still in the design phase.

To help states and districts implement choice and to gain a better
understanding of its impact, we recommend that the Secretary of Education:

Monitor issues related to limited classroom capacity that may arise as
implementation proceeds, in particular, the extent to which capacity
constraints hinder or prevent transfers. Based on this monitoring,
Education should consider whether or not additional flexibility or
guidance addressing capacity might be warranted.

     o Collect and disseminate additional examples of successful strategies
       that districts employ to address capacity limitations and information
       on the costs of these strategies.
     o Assist states in developing strategies for better informing parents
       about the school choice option by collecting and disseminating
       promising practices identified in the course of working with states
       and districts. For instance, Education might collect and share
       examples of clear, wellwritten, and particularly informative notices.
       In addition, Education should make the results of its parental
       surveys, conducted as part of its national study, widely available for
       use by states and districts to help them better refine their
       communications with parents regarding school choice.
     o For its student outcomes study, Education should use the methodology
       with the greatest potential to identify the effects of the school
       choice transfer on students' academic achievement. The methodology
       selected should allow it to compare academic outcomes for transferring
       students over several years with outcomes for similar students not
       transferring, while accounting for differences in student
       demographics. The study should also examine the extent to which
       transferring students remain in the schools to which they transfer.

                                Agency Comments
                               and Our Evaluation

We provided a draft of this report to the Department of Education for
review and comment. Education's written comments appear in appendix

VII. Recommended technical changes have been incorporated in the report as
appropriate.

Education said that the report would be a useful addition to the
literature on the public school choice provision and indicated its intent
to use the findings and recommendations in the report to improve
Education's technical assistance to states and districts and to strengthen
its implementation studies. Specifically, Education agreed with our
recommendations concerning monitoring capacity and disseminating
successful strategies to meet capacity challenges, noting several projects
under development that might assist in carrying out these recommendations.
Education also strongly supported our recommendation that it assist states
in better informing parents about the school choice option and related
some of its plans for doing so. Regarding our recommendation concerning
the department's study of choice implementation, Education said that it is
working to design a rigorous analysis of student outcomes and will take
our recommendation into consideration as it refines the design for the
study.

We are sending copies of this report to appropriate congressional
committees, the Secretary of Education, and other interested parties.
Copies will be made available to other interested parties upon request. In
addition, the report will be available at no charge on GAO's Web site at
http://www.gao.gov. If you or your staff have any question about this
report, please call me at (202) 512-7215. Key contributors are listed in
appendix VIII.

Marnie S. Shaul

Director, Education, Workforce, and Income Security Issues

                       Appendix I: Scope and Methodology

The objectives of this report were to determine (1) the extent to which
Title I schools have been affected by the school choice provision of The
No Child Left Behind Act (NCLBA) of 2001 in terms of the number of schools
identified for choice and the number of students exercising the option;
(2) the experiences of selected school districts in implementing the
choice provision; and (3) the kinds of guidance and technical assistance
that the Department of Education provided states and districts as they
implemented public school choice.

To determine the extent that schools have been affected by school choice
in terms of the number of schools required to offer choice and the number
of students exercising the option, we analyzed data for school years
20022003 and 2003-2004 using two sources: our survey of state education
agencies and state reports to Education. To obtain data on the number of
schools that had to offer choice, we used a different source for each
school year. For 2002-2003, we surveyed state education agencies in all 50
states, the District of Columbia, and Puerto Rico; for 2003-2004, we
obtained data from Education that had been reported by each state in its
Consolidated State Performance Report: Part I. Information on the number
of students that chose to transfer to another school for school year
2002-2003 was obtained from the Consolidated State Performance Report:
Part I; for 2003-2004, the data were obtained from our survey of the state
education agencies, the District of Columbia, and Puerto Rico. Although
there was a 100 percent response rate to our survey and to Education's
report, not all states provided complete information. Seven states did not
provide any transfer information for 2003-2004 because they did not plan
to collect this information until later in school year 2004-2005.

To test the reliability of these data, we performed a series of tests,
which included checking that data were consistent, that subtotals added to
totals and that data provided for one year bore a reasonable relationship
to the next year's data and to data reported elsewhere, including state
education websites. Where we found discrepancies or sought clarification,
we followed up with state officials. In several states, officials revised
the numbers that they had initially reported to us or to Education. We
determined these data to be sufficiently reliable for our purposes.

In addition, we sought information on schools and students from several
sources. The grade span and location of schools (urban or rural)
identified for choice and the demographics of their students was available
from the National Center for Education Statistics (NCES). We were not able
to describe the characteristics of the schools required to offer choice in
2002-

Appendix I: Scope and Methodology

2003 because the list of schools was not available. We analyzed data for
the nation as a whole and by state, expressing the results in relation to
the universe of all Title I schools or all public K-12 schools. When we
compared results in the first school year with results in the second, we
compared only states that provided information for both years and
eliminated any states that provided data only for a single year. Because
NCES data were not available for 2003-2004, the year for which we obtained
lists of schools identified for choice, we used as a proxy the 20022003
enrollment data for these schools including student numbers, minority
status and eligibility for the free or reduced price school lunch program
as a measure of the family income. Because these were the only data
available, and because we considered them adequate for our purposes, we
used 2002-2003 enrollment data to characterize schools in 2003-2004, based
on an assumption that at the aggregate levels the numbers and
characteristics did not differ significantly from one year to the next. We
discussed this assumption with education officials at NCES, and for a
sample of states, tested it by checking the changes from 20012002 to the
following year for schools identified. We also tested the reliability of
the NCES data by comparing our numbers to published totals and by
reviewing documentation. We considered these data to be sufficiently
reliable for our purposes.

To determine the experiences of selected school districts' implementation
of NCLBA school choice, we visited eight districts that had schools
required to offer choice. On the basis of our discussions with state
officials and our own research, we selected districts located in seven
states- California, Illinois, Ohio, Mississippi, Pennsylvania, Tennessee,
and Washington. Districts selected were based on geographic location and
district profile in terms of the number of schools required to offer
school choice, student population, and demographic profile. (See table 10
for district characteristics.) During our visits, we interviewed officials
in school district offices and in most districts, also interviewed
principals of schools that were required to offer school choice as well as
principals of schools that received transferring students. In each of
these districts, we attempted to obtain data on the characteristics of
students-such as race, poverty, and academic achievement-that had
transferred to another school under NCLBA school choice in school year
2003-2004; we had limited success at obtaining such information from most
schools. We were able to obtain information on transferring students'
academic achievement from one district but most districts had not
collected this information.

Appendix I: Scope and Methodology

Table 10: Characteristics for 2003-2004 of Eight Districts We Visited
                                              Fresno          North     
                  Akron   Chicago    Elgin  Unified  Memphis     Panola              Tacoma 
                 Public    Public (School   School     Public    School Pittsburgh   Public 
District       Schools,  Schools, District District, Schools, District,     Public Schools, 
                                  U-                                               
Characteristic     Ohio      Ill.   46),    Calif.      Tenn.        MS Schools,      Wash. 
                                    Ill.                                Pa.        
Geographic      Midwest   Midwest Midwest       West    South     South  Northeast     West 
area                                                                               
Total schools in                                                                   
districta                     58   588      51    86      176         5         85       54 
Total enrollment in                                                                
district            28,100 426,000   39,500  80,300  114,800b    1,820c     33,800   31,700 
Number of schools                                                                  
required to offer                                                                           
choice                     8       368       6    39      71d         1         21        5
Number of students                                                                 
eligible for         2,960  250,000   2,460  34,005    52,600       390      6,800    2,480 
choice                                                                             
Number of students                                                                 
that transferred        76       548e      60    101     388f         0        151      253 
Percentage of eligible                                                             
students that                                                                      
transferred             2.6%   0.2%    2.4%    0.3%      0.7%        0%       2.2%    10.2% 
Percentage of                                                                      
students                                                                           
economically                                                                       
disadvantaged        56%        85%      32%    75%       65%       94%        63%      55% 
Voluntary school                                                                   
choice program         Yes        Yes     No    Yes       Yes        No        Yes      Yes 

Source: GAO analysis of school data from state or district education
agency Web sites and school district officials.

aIncludes district charter and magnet schools; excludes specialized and
alternative schools.

b

Memphis student enrollment is average daily membership.

cThe student count for North Panola is from the NCES Common Core of Data
for 2002.2003.

d

Memphis schools are those the state identified for choice in school year
2002-03.

e

Chicago approved 1,097 student transfers and about half of the students
actually transferred.

f

Memphis approved 388 student transfers but did not collect data on the
number of students that actually transferred.

To determine the kinds of guidance and technical assistance that Education
provided states and districts as they implemented NCLBA public school
choice, we reviewed regulations, policy letters, and nonregulatory
guidance provided to states and districts. We also interviewed Education
officials involved with developing the guidance and providing assistance
to states in implementing school choice. To obtain the perspective of
officials using the guidance provided by Education, we

Appendix I: Scope and Methodology

interviewed district officials at all eight sites and state agency
officials in 2 states. In addition, to obtain a national perspective on
the effectiveness of Education's guidance and assistance to the states and
districts, we interviewed officials at the Council of the Great City
Schools, the Council of Chief State School Officers, and the Center on
Education Policy.

Appendix II: NCLBA Interventions for Schools Not Meeting Yearly Performance
Goals Over Time

Number of years of missing
performance goals based NCLBA interventions for    NCLBA interventions for 
on tests                schools that do            schools that            
administered in prior   receive Title I funds      do not receive-Title I  
school year                                        funds                   
First year missed       None                       None; states may set    
                                                      their own               
Second year missed         In first year of school None; states may set    
                             improvement, required to their own               
                           offer choice               
Third year missed       In second year of school   None; states may set    
                           improvement,               their own               
                           required to offer choice   
                           and supplemental           
                           educational services       
Fourth year missed      In corrective action, a    None; states may set    
                           required to offer choice   their own               
                           and supplemental services  
                                      In planning for None; states may set    
Fifth year missed        restructuring, b required their own               
                                             to offer 
                           choice and supplemental    
                           services                   
Sixth year missed             In implementation of None; states may set    
                           restructuring, required to their own               
                           offer choice and           
                           supplemental services      

Source: GAO analysis of NCLBA.

aCorrective action is a significant intervention in a school that is
designed to remedy the school's persistent inability to make adequate
progress toward all students becoming proficient in reading and
mathematics.

b

Restructuring is a major reorganization of a school, involving fundamental
reforms, such as significant changes in the school's staffing and
governance.

Appendix III: Number of Title I Schools in Each State Identified for Choice in
School Year 2002-2003

                   Page 43 GAO-05-7 No Child Left Behind Act

                  Number of schools that  Number of schools  Percent of Title 
                                                                    I schools 
State                received Title I     identified for    identified for 
                             funds                   choice            choice 
Hawaii                            139                 82             58.99 
Georgia                         1,064                600             56.39 
Indiana                           791                236             29.84 
Maryland                          361                105             29.09 
New Mexico                        533                116             21.76 
Illinois                        2,433                526             21.62 
Arizona                         1,021                193             18.90 
Delaware                          106                 20             18.87 
New Jersey                      1,379                259             18.78 
Idaho                             403                 72             17.87 
Massachusetts                   1,129                195             17.27 
Arkansas                          825                123             14.91 
California                      5,467                814             14.89 
New York                        2,941                431             14.65 
Colorado                          582                 80             13.75 
Tennessee                         856                114             13.32 
Rhode Island                      149                 17             11.41 
Pennsylvania                    1,892                198             10.47 
Nevada                            117                 12             10.26 
Michigan                       2,196a                216              9.84 
District of Columbia              172                 15              8.72 
Utah                              216                 18              8.33 
Ohio                            2,052                161              7.85 
Kansas                            648                 50              7.72 
Montana                           643                 45              7.00 
North Dakota                      443                 29              6.55 
Wisconsin                       1,083                 68              6.28 
Alabama                           872                 52              5.96 
Minnesota                       1,007                 60              5.96 
Alaska                            299                 17              5.69 
Nebraska                          497                 27              5.43 
South Carolina                    534                 27              5.06 
Washington                        997                 48              4.81 
Virginia                          785                 34              4.33 
South Dakota                      345                 13              3.77 

Appendix III: Number of Title I Schools in Each State Identified for
Choice in School Year 2002-2003

            Number of schools that  Number of schools      Percent of Title I 
                                                                      schools 
State          received Title I     identified for   identified for choice 
                       funds                   choice 
Iowa                        712                 26                    3.65 
Puerto Rico               1,511                 48                    3.18 
Kentucky                    866                 25                    2.89 
Oklahoma                  1,188                 30                    2.53 
West Virginia               452                  8                    1.77 
Missouri                  1,277                 21                    1.64 
Connecticut                 528                  8                    1.52 
New Hampshire               269                  4                    1.49 
Mississippi                668a                  9                    1.35 
North Carolina            1,131                 15                    1.33 
Louisiana                   943                 11                    1.17 
Maine                       442                  4                    0.90 
Oregon                      582                  5                    0.86 
Texas                     4,823                 37                    0.77 
Florida                   1,374                  0                    0.00 
Vermont                     214                  0                    0.00 
Wyoming                     171                  0                    0.00 
Total                    52,128              5,324 

Source: GAO survey, Education's Consolidated State Performance Report.

aM

ichigan and Mississippi did not report the number of Title I schools for
school year 2002-2003. Michigan suggested and we agreed to use the number
in school year 2001-2002 as a proxy. For Mississippi, we used a proxy from
the National Center for Education Statistics.

Appendix IV: Number of Title I Schools in Each State Identified for Choice in
School Year 2003-2004

                   Page 45 GAO-05-7 No Child Left Behind Act

                  Number of schools that Number of schools   Percent of Title 
                                                                    I schools 
State                received Title I   identified for      identified for 
                             funds             choice                  choice 
Georgia                         1,115                533             47.80 
Hawaii                            204                 82             40.20 
Arkansas                          824                230             27.91 
Maryland                          380                102             26.84 
Illinois                        2,357                562             23.84 
Nevada                            117                 27             23.08 
New Mexico                        546                120             21.98 
California                      5,521               1205             21.83 
Alaska                            296                 64             21.62 
Arizona                         1,021                220             21.55 
New Jersey                     1,379a                262             19.00 
Massachusetts                   1,163                208             17.88 
New York                        3,006                528             17.56 
Pennsylvania                    1,724                298             17.29 
South Carolina                    549                 90             16.39 
Rhode Island                      147                 24             16.33 
Michigan                       2,196a                352             16.03 
Indiana                           786                 97             12.34 
Colorado                          659                 80             12.14 
Delaware                          104                 12             11.54 
District of Columbia              138                 15             10.87 
Idaho                             399                 43             10.78 
South Dakota                      336                 32              9.52 
Puerto Rico                     1,494                140              9.37 
Ohio                            2,116                191              9.03 
Louisiana                         945                 69              7.30 
Tennessee                         856                 56              6.54 
North Dakota                      365                 23              6.30 
Montana                           678                 40              5.90 
Virginia                          791                 44              5.56 
Alabama                           859                 46              5.36 
Wisconsin                       1,095                 52              4.75 
Kansas                            665                 30              4.51 
Washington                        995                 44              4.42 
Minnesota                       1,006                 38              3.78 

Appendix IV: Number of Title I Schools in Each State Identified for Choice
in School Year 2003-2004

            Number of schools that Number of schools       Percent of Title I 
                                                                      schools 
State          received Title I   identified for     identified for choice 
                       funds             choice       
Oklahoma                  1,238                 46                    3.72 
North Carolina            1,154                 35                    3.03 
Florida                   1,426                 43                    3.02 
Kentucky                    874                 25                    2.86 
Utah                        218                  6                    2.75 
Missouri                  1,275                 32                    2.51 
Connecticut                 499                 12                    2.40 
Vermont                     209                  4                    1.91 
West Virginia               409                  7                    1.71 
Iowa                        694                 11                    1.59 
Maine                       461                  6                    1.30 
Nebraska                    494                  6                    1.21 
Oregon                      595                  7                    1.18 
Mississippi                668a                  7                    1.05 
New Hampshire               248                  2                    0.81 
Texas                     5,061                  9                    0.17 
Wyoming                     184                  0                    0.00 
Total                    52,539              6,217 

Source: GAO survey, Education's Consolidated State Performance Report.

aM

ichigan, Mississippi, and New Jersey did not report the number of Title I
schools in school year 2003-2004. Michigan suggested and we agreed to use
the number in school year 2001-2002 as a proxy. For Mississippi, we used a
proxy from the National Center for Education Statistics for the 2002-2003
school year. For New Jersey, we used the number of Title I schools the
state reported for the 2002-2003 school year.

     Appendix V: Number of Students in Each State Transferring under Choice
                        Option in First 2 Years of NCLBA

                 Number of students who      Number of Percent of eligible    
                                          students who students a who         
State                transferred in  transferred in         transferred in 
                           2002-2003         2003-2004              2003-2004 
Oregon                           742            873                  17.40 
Florida               Not applicable          1,820                   6.28 
South Carolina                   519          1,772                   4.28 
West Virginia                     49             90                   4.14 
New Mexico                       529          1,699                   3.72 
Kentucky                         229            384                   3.66 
Alabama                          777            740                   3.50 
North Carolina                    93            337                   3.42 
Connecticut                        7            262                   3.32 
Delaware                           0            195                   2.92 
Utah                             204             58                   2.85 
Tennessee                        810           929b                   2.82 
Indiana                        1,301          1,199                   2.52 
Virginia                         277            432                   2.26 
Kansas                           202            212                   2.18 
District of Columbia             192            197                   2.18 
Maryland                         709          1,050                   2.01 
New Jersey                       257          2,738                   1.80 
Minnesota              Not available           270b                   1.68 
New York                       1,507          7,373                   1.51 
Iowa                             170             60                   1.48 
Ohio                             698          1,169                   1.24 
Washington                       620           270b                   1.21 
Nevada                           127            226                   1.20 
Louisiana                         18            371                   0.88 
Colorado                         194           299b                   0.86 
Pennsylvania                     110          1,126                   0.56 
Illinois                      1,418c         1,364c                   0.34 
Hawaii                            21            154                   0.31 
Alaska                             2             26                   0.31 
Massachusetts                    845           304b                   0.31 
California                     3,139          3,419                   0.28 
New Hampshire                      1              2                   0.23 
Mississippi                        4              7                   0.19 
Rhode Island                      17            27b                   0.19 

                   Page 47 GAO-05-7 No Child Left Behind Act

Appendix V: Number of Students in Each State Transferring under Choice
Option in First 2 Years of NCLBA

         Number of students who Number of students Percent of eligible        
                                               who students a who             
State        transferred in      transferred in   transferred in 2003-2004 
                   2002-2003             2003-2004 
Missouri                  91                 16                       0.13 
South Dakota               1                  5                       0.07 
Idaho                      0                  4                       0.03 
Maine                      0                  0                       0.00 
Nebraska                   0                  0                       0.00 
North Dakota              11                  0                       0.00 
Puerto Rico                0                  0                       0.00 
Texas                     59                  0                       0.00 
Vermont       Not applicable                  0                       0.00 
Arizona                   83      Not available              Not available 
Arkansas                 171      Not available              Not available 
Georgia                1,874      Not available              Not available 
Montana                   38      Not available              Not available 
Oklahoma                 549      Not available              Not available 
Wisconsin                111      Not available              Not available 
Michigan       Not available      Not available              Not available 
Wyoming       Not applicable     Not applicable             Not applicable 
Total                 18,732             31,479 

Source: GAO survey, Education's Consolidated State Performance Report.

aTo calculate percents, we used NCES 2002-2003 school year enrollment
figures for all schools identified for choice in the 2003-2004 school
year. See app. I.

bTransfer figures shown may be understated, because they represent
transfers in large districts in these 6 states: Colorado, Massachusetts,
Minnesota, Rhode Island, Tennessee, and Washington.

c

In both years, Illinois state officials reported transfers for most but
not all districts. In 2002-2003, the state report included Chicago; in
2003-2004, the state report excluded Chicago and we added the transfer
figure provided by Chicago.

Note: Michigan officials told us that they were not able to distinguish
the number of NCLBA transfers from the other transfers under their
statewide open enrollment programs.

Appendix VI: Poverty and Minority Rates of Schools Required to Offer Choice and
Schools Offered as Transfer Options

                               Poverty Minorities

School enrollment Number of schools percentage by required to offer
district choice

    Number of schools Number of schools offered as transfer Number of schools
                 offered as transfer options required to offer choice options

Akron, Ohio

                                 0-25% 0 1 011

                                 26-50% 0 9 011

                                51-75% 1 16 4 7

                                76-100% 7 11 4 8

                               Chicago, Illinois

                                  0-25% 1 1 02

                                 26-50% 0 5 02

                                51-75% 2 11 3 6

                             76-100% 365 23 365 29

Elgin, Illinois

                                 0-25% 0 13 0 4

                                 26-50% 0 3 09

                                 51-75% 3 0 13

                                 76-100% 2 0 40

                               Fresno, California

                                  0-25% 0 3 00

                                 26-50% 1 4 03

                                 51-75% 4 1 17

                               76-100% 34 10 38 8

Memphis, Tennesseeb

                                  0-25% 0 1 00

                                 26-50% 0 9 02

                                 51-75% 5 5 37

                              76-100% 66 30 68 36

Pittsburgh, Pennsylvania

                                  0-25% 0 0 06

                                26-50% 0 11 0 15

                                51-75% 2 17 8 17

                              76-100% 19 23 13 13

Appendix VI: Poverty and Minority Rates of Schools Required to Offer
Choice and Schools Offered as Transfer Options

                               Poverty Minorities

School enrollment             Number of  Number of    Number of  Number of 
percentage by district          schools    schools      schools    schools 
Tacoma, Washington 0-25%    required to offered as  required to offered as 
                            offer choice 0   transfer offer choice   transfer 
                                            options 0            0  options 0 
26-50%                                0          1            0          3 
51-75%                                0          3            4          5 
76-100%                               5          3            1          0 

Source: GAO analysis of school data from state or district education
agency Web sites.

a

For Chicago, minority data were not available for one of the 40 transfer
schools.

b

Two of the 47 transfer schools in Memphis were new in school year
2003-2004, and no data were available on the poverty or minority rates for
the student enrollment at these two schools.

cOne of the 8 transfer schools in Tacoma was new in school year 2003-2004,
and no data were available on the poverty rate for the student enrollment
at this one school.

Appendix VII: Comments from the Department of Education

Appendix VII: Comments from the Department of Education

Appendix VII: Comments from the Department of Education

Appendix VIII: GAO Contacts and Staff Acknowledgments

Deborah L. Edwards (202) 512-5416 or [email protected]

  GAO Contacts

Patricia L. Elston (202) 512-3016 or [email protected]

In addition to those named above, the following individuals made

  Staff

important contributions to this report: Nancy Purvine, Sara Margraf,
Acknowledgments Scott Spicer, John Mingus, Amy Buck, and Margaret Armen.

Related GAO Products

No Child Left Behind Act: Improvements Needed in Education's Process for
Tracking States' Implementation of Key Provisions. GAO-04-734. Washington,
D.C.: September 30, 2004.

No Child Left Behind Act: Additional Assistance and Research on Effective
Strategies Would Help Small Rural Districts. GAO-04-909. Washington, D.C:
September 23, 2004.

Special Education: Additional Assistance and Better Coordination Needed
among Education Offices to Help States Meet the NCLBA Teacher Requirements
. GAO-04-659. Washington, D.C.: July 15, 2004.

Student Mentoring Programs: Education's Monitoring and Information Sharing
Could Be Improved. GAO-04-581. Washington, D.C.: June 25, 2004.

No Child Left Behind Act: More Information Would Help States Determine
Which Teachers Are Highly Qualified. GAO-03-631. Washington, D.C.: July
17, 2003.

Title I: Characteristics of Tests Will Influence Expenses: Information
Sharing May Help States Realize Efficiencies . GAO-03-389. Washington,
D.C.: May 8, 2003.

Disadvantaged Students: Fiscal Oversight of Title I Could Be Improved.
GAO-03-377. Washington, D.C.: February 28, 2003.

Title I: Education Needs to Monitor States' Scoring of Assessments.
GAO-02-393. Washington, D.C.: April 1, 2002.

Title I Funding: Poor Children Benefit Though Funding Per Poor Child
Differs . GAO-02-242. Washington, D.C.: January 31, 2002.

Title I Program: Stronger Accountability Needed for Performance of
Disadvantaged Students . GAO/HEHS-00-89. Washington, D.C.: June 1, 2000.

(130321)

  GAO's Mission

The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site ( www.gao.gov) . Each weekday, GAO posts GAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To

have GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

                             Order by Mail or Phone

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

  To Report Fraud, Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm

  E-mail: [email protected]

Federal Programs Automated answering system: (800) 424-5454 or (202)
512-7470

Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Relations
Washington, D.C. 20548

Susan Becker, Acting Manager, [email protected] (202) 512-4800

  Public Affairs

U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

    PRINTED ON

RECYCLED PAPER
*** End of document. ***