BIA and DOD Schools: Student Achievement and Other		 
Characteristics Often Differ from Public Schools' (28-SEP-01,	 
GAO-01-934).							 
								 
Unlike public schools, where federal funding constitutes a small 
portion of total resources, the Bureau of Indian Affairs (BIA)	 
and the Department of Defense (DOD) school systems depend almost 
entirely on federal funds. Although the two school systems have  
this feature in common, their histories and settings are quite	 
different.  The academic achievement of many BIA students as	 
measured by their performance on standardized tests and other	 
measures is far below the performance of students in public	 
schools. BIA students also score considerably below national	 
averages on college admissions tests. BIA school officials	 
indicate that nearly all BIA teachers are fully certified for the
subjects or grade levels they teach, although officials said that
some schools experience considerable difficulties recruiting and 
retaining qualified staff. In terms of educational technology,	 
access levels to computers and the Internet reported by BIA	 
school administrators exceeded those of public schools, but	 
technical support to maintain computers and to assist teachers	 
with using technology in the classroom was more limited. Problems
with school facilities were reported by many school		 
administrators. Estimated per-pupil expenditures for BIA schools 
vary widely by school type (for example, day or boarding), but	 
are generally higher than for public schools nationally. The	 
academic achievement of DOD students, as measured by their	 
performance on standardized tests and their plans for enrolling  
in college, generally exceeds that of elementary and secondary	 
students nationwide. On college admission tests, DOD students	 
score at or near national averages. DOD school administrators	 
indicated that nearly all their teachers are fully certified for 
the subjects or grade levels they teach, and about two-thirds	 
have advanced degrees. Access levels to computers and the	 
Internet were better than public schools, and nearly all	 
administrators reported that technical support is available in	 
their school. Many DOD administrators reported some problems with
their school facilities, but overall the conditions of their	 
buildings did not vary greatly from those reported by public	 
schools in 1999. Estimated per-pupil expenditures at DOD schools 
located overseas were higher than expenditures for those located 
in the U.S.  DOD records show that a substantial part of the	 
difference is related to moving and housing costs for teachers	 
and staff in overseas schools. DOD's domestic schools, which	 
generally do not have these expenses, are much closer to national
per-pupil expenditures. 					 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-01-934 					        
    ACCNO:   A01447						        
  TITLE:     BIA and DOD Schools: Student Achievement and Other       
Characteristics Often Differ from Public Schools'		 
     DATE:   09/28/2001 
  SUBJECT:   Academic achievement				 
	     Comparative analysis				 
	     Computers						 
	     Educational facilities				 
	     Elementary schools 				 
	     Federal funds					 
	     Public schools					 
	     Secondary schools					 
	     Teachers						 
	     American College Test				 
	     Dept. of Education National Assessment		 
	     of Educational Progress				 
								 
	     Dept. of Education Safe and Drug-Free		 
	     Schools and Communities Program			 								 
	     Dept. of Education Title I Program 		 
	     National School Breakfast Program			 
	     National School Lunch Program			 
	     Scholastic Assessment Test 			 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** 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-01-934
     
BIA AND DOD SCHOOLS

Student Achievement and Other Characteristics Often Differ from Public
Schools? Report to Congressional Requesters

United States General Accounting Office

GAO

September 2001 GAO- 01- 934

Page i GAO- 01- 934 BIA and DOD Schools Letter 1

Results in Brief 2 Background 3 The BIA School System 12 The DOD School
System 34 Agency Comments 45

Appendix I Scope and Methodology 47 Student Achievement 48 Teacher Staffing
and Educational Technology 50 Facilities 51 Estimated Per Pupil Expenditures
53

Appendix II Comments from the Department of Defense 56

Appendix III Comments from the Department of the Interior 57 GAO Comments 68

Appendix IV GAO Contacts and Staff Acknowledgments 74 GAO Contacts 74 Staff
Acknowledgments 74

Tables

Table 1: Performance of Students in BIA and Public Schools on State
Assessments in School Year 1999- 2000 13 Table 2: 1999- 2000 BIA Graduates?
Performance on SAT and ACT

Compared With National Averages 14 Table 3: Technology Statistics for BIA
and Public Schools 20 Table 4: Reported Facility Deficiencies for BIA
Schools and

Dormitories and Public Schools 25 Table 5: Selected Reported Deficiencies
and Estimated Repair

Costs for BIA Schools and Dormitories 29 Table 6: Comparison of Per- Pupil
Expenditures at BIA and Public

Schools, 1997- 98 32 Table 7: National Percentile Rankings for Students in
DOD Schools

on the TerraNova Multiple Assessments Standardized Tests for School Year
1999- 2000 35 Contents

Page ii GAO- 01- 934 BIA and DOD Schools

Table 8: Average SAT and ACT Scores for DOD and Public School Students
Graduating in 2000 36 Table 9: Median Class Sizes at DOD Schools 38 Table
10: Technology Statistics for DOD and Public Schools 39 Table 11: Reported
Facilities Deficiencies for DOD and Public

Schools 42 Table 12: Selected Reported Deficiencies and Estimated Repair

Costs for DOD Schools 44 Table 13: Comparison of Per- Pupil Expenditures at
DOD and

Public Schools, 1997- 98 45

Figures

Figure 1: Locations of BIA Schools in School Year 1999- 2000 4 Figure 2:
Locations of DOD Europe Area Schools in School Year

1999- 2000 7 Figure 3: Locations of DOD Pacific Area Schools in School Year

1999- 2000 8 Figure 4: Locations of DOD Domestic and Cuba Schools in School

Year 1999- 2000 9 Figure 5: Percentage of BIA Schools Reporting That a
Majority of

Teachers Received Various Types of Training 17 Figure 6: Percentage of BIA
Schools Reporting That a Majority of

Teachers Assign Students Computer or Internet Work at Least Once Per Week
for Various Classroom Activities 22 Figure 7: Percentage of BIA Schools
Reporting That Various Work

Spaces and Equipment Are Insufficient 27 Figure 8: Percentage of DOD Schools
Reporting That a Majority of

Teachers Received Various Types of Training 37 Figure 9: Percentage of DOD
Schools Reporting That a Majority of

Teachers Assign Students Computer or Internet Work At Least Once Per Week
for Various Classroom Activities 40 Figure 10: Percentage of DOD Schools
Reporting That Various

Work Spaces and Equipment Are Insufficient 43

Abbreviations

ACT American College Test AIMS Arizona Instrument to Measure Standards AME
Applied Management Engineering, Inc. BIA Bureau of Indian Affairs

Page iii GAO- 01- 934 BIA and DOD Schools

DOD Department of Defense DODEA Department of Defense Education Activity
NAEP National Assessment of Educational Progress NCES National Center for
Education Statistics NRC National Research Council OIEP Office of Indian
Education Programs SAT Scholastic Assessment Test USDA U. S. Department of
Agriculture

Page 1 GAO- 01- 934 BIA and DOD Schools

September 28, 2001 The Honorable Robert C. Byrd The Honorable Ben Nighthorse
Campbell The Honorable Byron L. Dorgan The Honorable Daniel K. Inouye United
States Senate

The federal government has direct responsibility for two school systems
serving elementary and secondary students- the Bureau of Indian Affairs
(BIA) and the Department of Defense (DOD) school systems. Unlike public
schools, where federal funding constitutes a small portion of total
resources, the BIA and DOD school systems depend almost entirely on federal
funds. Although the two school systems have this feature in common, their
histories and settings are quite different. Because these school systems are
a federal responsibility, the Congress is interested in ensuring that
children attending BIA and DOD schools are receiving a quality education.

To better understand the quality of the educational environments in BIA and
DOD schools, you asked us to provide information about several aspects of
both school systems. Our study provides information on student academic
performance, teacher staffing, access to educational technology, the
condition of facilities, and expenditure levels for each system. This study
also provides comparative data for public schools when these data were
available.

The scope of our review included BIA day schools and boarding schools, and
DOD schools located both in the United States and overseas. To the extent
possible, we excluded from our analyses BIA peripheral dormitories on
reservations- which house Indian students who attend nearby public schools-
because they generally do not have academic programs. Information for this
review was gathered primarily from agency data obtained from BIA and DOD,
and from a mail survey of administrators at all BIA and DOD schools covering
the 1999- 2000 school year. Survey response rates were 81 percent and 92
percent for BIA and DOD schools, respectively. We also obtained data on
student performance on achievement tests and college entrance examinations
from states and college testing organizations for students in BIA schools.
Data needed to precisely compute per- pupil expenditures for both BIA and
DOD schools were not available. Therefore, our calculations are estimates
based on BIA and DOD agency data and federal expenditure information from
tribal

United States General Accounting Office Washington, DC 20548

Page 2 GAO- 01- 934 BIA and DOD Schools

audit reports submitted under the Single Audit Act. We also conducted site
visits to nine BIA schools located in South Dakota, Arizona, Minnesota, and
Washington, and seven DOD schools located at three military installations in
the United States- Maxwell Air Force Base, in Alabama; and Fort Benning and
Robins Air Force Base, in Georgia. For further details about our scope and
methodology, see appendix I. We did not verify information collected for
this study but obtained supporting documentation where appropriate. We
conducted our work between June 2000 and July 2001 in accordance with
generally accepted government auditing standards.

The academic achievement of many BIA students as measured by their
performance on standardized tests and other measures is far below the
performance of students in public schools. BIA students also score
considerably below national averages on college admission tests. Academic
performance has been strongly associated with educational and income levels
of parents, and students in BIA schools often come from family settings
where education, employment, and earning levels are lower than the national
average. BIA school administrators indicate that nearly all BIA teachers are
fully certified for the subjects or grade levels they teach, although
several officials said that some schools experience considerable
difficulties recruiting and retaining qualified staff. In terms of
educational technology, access levels to computers and the Internet reported
by BIA school administrators exceeded those of public schools, but technical
support to maintain computers and to assist teachers with using technology
in the classroom was more limited than in public schools. Problems with
school facilities were reported by many school administrators. For example,
administrators at more than 60 percent of responding BIA schools reported
having at least one building in inadequate condition compared with about a
quarter of administrators at public schools surveyed by the Department of
Education (Education) in 1999. Finally, our estimated per- pupil
expenditures for BIA schools vary widely by school type (for example, day or
boarding), but are generally higher than for public schools nationally. A
number of factors distinguish BIA schools from public schools, which may add
to their costs. For example, a high proportion of BIA students have special
needs, and BIA schools support a broader infrastructure (such as sewer and
water systems) than most public schools.

The academic achievement of DOD students, as measured by their performance
on standardized tests and their plans for enrolling in college, generally
exceeds that of elementary and secondary students nationwide. Results in
Brief

Page 3 GAO- 01- 934 BIA and DOD Schools

On college admission tests, DOD students score at or near national averages.
DOD school administrators indicated that nearly all their teachers are fully
certified for the subjects or grade levels they teach, and about two- thirds
have advanced degrees. Access levels to computers and the Internet reported
by these administrators were better than public schools, and nearly all
administrators reported that technical support is available in their school.
Many DOD school administrators reported some problems with their school
facilities, but overall the conditions of their buildings did not vary
greatly from those reported by public schools in 1999. Our estimated per-
pupil expenditures at DOD schools located overseas were higher than
expenditures for those located in the U. S. DOD records show that a
substantial part of the difference is related to moving and housing costs
for teachers and staff in overseas schools. DOD?s domestic schools, which
generally do not have these expenses, are much closer to national per- pupil
expenditures.

The BIA and DOD school systems are unlike public school systems in a number
of ways and are also distinct from each other. Information on the schools,
the students and families they serve, the settings in which they operate,
and other fundamental aspects of these systems helps put our findings into
context.

BIA schools serve less than 10 percent of all Indian students enrolled in
elementary and secondary schools in this country. 1 In school year 1999-
2000, the total enrollment was 47,080 students in the 171 2 schools funded
by BIA. BIA schools are located in 23 states; however, over 70 percent of
the schools are in four states- Arizona, New Mexico, North Dakota, and South
Dakota (see fig. 1). The schools are located primarily in rural areas and
small towns and serve Indian children living on or near reservations.
Individual school enrollments range in size from 14 to over 1,000 students,
but a little over half of the schools enroll fewer than 250 students; most
are combined schools spanning both elementary and secondary grades. A

1 According to the National Center for Education Statistics (NCES),
approximately 541,000 American Indian/ Alaska Native children were enrolled
in public elementary and secondary schools in the 50 states and the District
of Columbia in school year 1999- 2000.

2 In addition, BIA supports 14 peripheral dormitories which served 1,689
children in school year 1999- 2000. Two of the 14 peripheral dormitories
offered kindergarten- only programs in school year 1999- 2000. We considered
peripheral dormitories to be outside the scope of this report and excluded
them from our analyses where possible. Background

BIA Schools

Page 4 GAO- 01- 934 BIA and DOD Schools

unique feature of the BIA school system is that almost one- third of the
schools have a residential component; that is, they board at least a portion
of the students who attend the school. In total, about 17 percent of BIA
students reside in school dormitories. Some students board because their
homes are so far away or inaccessible that daily transportation is
impractical. However, an increasing proportion of students reside in the
dormitories for other reasons; for example, some students live in boarding
schools to separate them from dysfunctional or severely impoverished home
environments.

Figure 1: Locations of BIA Schools in School Year 1999- 2000

Source: Data are from BIA?s Office of Indian Education Programs. School
locations are indicated by stars, but some stars represent more than one
school.

Education programs and activities of BIA schools are administered by the
BIA?s Office of Indian Education Programs (OIEP); however, in general, the
organization of BIA schools is highly decentralized. Each school is

Page 5 GAO- 01- 934 BIA and DOD Schools

governed by its own independent school board, which has authority over
functions such as hiring personnel, adopting budgets, and setting policy.
Another important aspect of the BIA school system is the agency?s support
and encouragement of tribal control of school programs. In school year 1999-
2000, 108 schools, or about two- thirds of all BIA- funded schools, were
operated by tribes or tribal organizations under grants, contracts, or
compacts with BIA; the remainder were operated by BIA.

In fiscal year 2001, BIA received a total of $777.6 million to support the
operations of its schools and address educational facility needs. 3 This
represented a substantial (30 percent) increase over fiscal year 2000
funding levels. Nearly all of the increase was for repair or replacement of
school facilities; funding for school operations increased only moderately.
4 Like many public schools, BIA schools also receive funding from Education
programs, such as Title I and Safe and Drug Free Schools and Communities.
Funds from these programs are provided to BIA and passed through to schools.
5 Fiscal year 2000 funding from Education programs totaled $132 million. BIA
schools also participate in the U. S. Department of Agriculture?s (USDA)
child nutrition programs, such as the National School Lunch and Breakfast
programs, which provide free or reducedprice meals for children living in
families earning below certain income levels.

DOD operates schools for the children of military and civilian DOD personnel
overseas and on certain military bases in the United States. 6 In total, DOD
operated 224 elementary and secondary schools in school year 1999- 2000,
with an enrollment of almost 108,000 students. A little over

3 Funding for BIA schools comes primarily from two BIA appropriations:
Operation of Indian Programs and Construction. 4 Funding for school
facilities repair and replacement- which comprises a smaller portion of
BIA?s education budget- increased by 121 percent. Funding for school
operations, the majority of the budget, increased by about 5 percent. Taken
together these increases in fiscal year 2001 raised funding 30 percent over
the previous year.

5 Under an agreement with Education, BIA receives flow through funds for
several Education programs and is permitted to retain 1. 5 percent of these
funds to perform administrative responsibilities similar to those performed
by state education agencies.

6 The schools are organized into two separate but parallel systems:
Department of Defense Dependents Schools overseas, and the Department of
Defense Domestic Dependent Elementary and Secondary Schools in the United
States. We refer to schools in both systems as DOD schools. DOD Schools

Page 6 GAO- 01- 934 BIA and DOD Schools

two- thirds of DOD schools are located overseas, concentrated in several
countries in Europe and in Japan, Korea, and Cuba (see figs. 2 and 3). About
74,280 students were enrolled in DOD overseas schools in school year 1999-
2000. 7 The domestic schools are located on military bases in seven states
concentrated in the Southeast and in Guam and Puerto Rico (see fig. 4) and
enrolled about 33,690 students in school year 1999- 2000. 8 Individual
school enrollments range in size from 32 to over 1,300 in the two DOD
systems, with the median school size being about 450 students. According to
a DOD official, nearly all eligible children of military parents overseas
attend DOD schools. However, most children of military parents in the U. S.
attend public schools. DOD officials estimate that about 15 percent of
school age military dependents in total attend DOD schools abroad and in the
United States.

7 In addition, DOD pays tuition for about 2, 080 eligible students overseas
to attend non- DOD schools in countries where no DOD schools are available.
8 In addition, DOD pays tuition and/ or transportation costs for
approximately 2, 085 DOD students enrolled in public schools operated by the
local education agency in four states.

Page 7 GAO- 01- 934 BIA and DOD Schools

Figure 2: Locations of DOD Europe Area Schools in School Year 1999- 2000

Source: Data are from DODEA 1999- 2000 Accountability Report. (Arlington,
Va.: Department of Defense Education Activity (DODEA)). Some locations may
have more than one school.

AFRICA EUROPE

A T L A N T I C O C E A N

Page 8 GAO- 01- 934 BIA and DOD Schools

Figure 3: Locations of DOD Pacific Area Schools in School Year 1999- 2000

Source: Data are from DODEA 1999- 2000 Accountability Report. Some locations
may have more than one school.

A S I A

P A C I F I C O C E A N

Page 9 GAO- 01- 934 BIA and DOD Schools

Figure 4: Locations of DOD Domestic and Cuba Schools in School Year 1999-
2000

Source: Data are from DODEA 1999- 2000 Accountability Report. Some locations
may have more than one school.

Cuba Guam Puerto Rico

Page 10 GAO- 01- 934 BIA and DOD Schools

Education programs and activities of DOD schools are managed and directed by
the Department of Defense Education Activity (DODEA). The overseas and
domestic schools are organized into 24 districts, each managed by a
superintendent and other district staff, such as curriculum coordinators and
maintenance supervisors.

In fiscal year 2001, DOD schools received a total of $1.3 billion to support
school operations and address facility needs. 9 Overall, this represented
level funding since fiscal year 2000, although funding for school operations
and maintenance rose slightly while funding for school construction
decreased. DOD schools are not authorized to receive grants from federal
programs, including those administered by Education; however, DOD schools do
participate in the USDA child nutrition programs.

While the federal government has responsibility for both the BIA and DOD
school systems, the two systems are very distinct from each other,
particularly with respect to the types of students and families they serve.
The proportion of students who have special needs is much higher in BIA
schools than DOD schools. For example, according to agency records, about 1
in every 5 BIA students is enrolled in special education, compared with 1 in
every 12 DOD students; and nearly 60 percent of BIA students have limited
English proficiency compared with just 6 percent of DOD students. Students
in the two systems also differ in terms of their economic need. Eligibility
for the USDA?s free or reduced- price lunch programs is often used as an
indicator of poverty. While data on the proportion of children eligible for
free or reduced- price lunch were not complete for BIA schools, available
data for 123 of the 171 schools showed that over 80 percent of students were
eligible, compared with about onethird of students at DOD schools.

Specific information on the education levels and employment status of
parents of BIA and DOD students was not available, but other information
suggests significant differences between the two. For example, unemployment
on or near Indian reservations is very high- over 40 percent in 1999. In
contrast, students attending DOD schools are generally military dependents
and, by virtue of that fact, have at least one parent

9 Funding for DOD schools comes from Military Construction Appropriations,
DefenseWide; and Defense Appropriations- Operations and Maintenance,
Defense- Wide, and Procurement, Defense- Wide. Characteristics of BIA and

DOD Students and Their Families Differ

Page 11 GAO- 01- 934 BIA and DOD Schools

who is employed. With respect to education, about one- third of Indian
adults do not have high school diplomas, according to Census Bureau data for
1990; in contrast, nearly all DOD active military personnel have completed
high school. Moreover, a number of factors can affect parental involvement
in BIA schools, including long distances between home and school, language
barriers, and parents? poor associations with schools due to past federal
policies emphasizing the assimilation of Indian children. In comparison,
military communities have a tradition of supporting military families, and
this support extends to their schools. Parental involvement is highly
emphasized in DOD schools, with some base commands providing release time to
military personnel to volunteer in schools.

At various times, the Congress has considered the question of whether there
is a need for BIA and DOD schools in locations where public school systems
are close by. Since 1794, when the first treaty providing for any form of
Indian education was signed, the federal government has provided education
services for Indian children, and has carried it out almost exclusively
through the BIA. However, in the 20th century, the states began providing
for Indian education and currently provide schooling for the majority of
Indian elementary and secondary students. During the latter part of the
1970s, the Congress reviewed the need to construct certain BIA schools close
to public schools to determine if BIA was properly ranking schools on its
school construction list. Though Indian families can choose where to send
their children to school, BIA?s policy at that time was to have children
attend nearby public schools with adequate facilities instead of building
new BIA schools to serve these students. However, in 1988, the Congress
decided that proximity to a public school could not be the primary reason
for rejecting a BIA school construction project.

The need for DOD domestic schools also has been reviewed a number of times,
specifically to determine whether transferring schools to nearby public
school districts is feasible. DOD domestic schools were established to
provide education to military children in communities where the local
schools were deemed unable to provide a suitable education. After 1950,
schools were added to the system to allow military children to attend
integrated schools in locations where local schools remained segregated.
Since that time, however, more and more schools have been transferred to
public school districts, in part because of the integration of public
schools and Education?s emphasis on state and local responsibility for the
education of military children. Thus, while at one time there were about 100
installations with DOD- operated schools, now there are 14 located in Need
for BIA and DOD

Schools

Page 12 GAO- 01- 934 BIA and DOD Schools

the 50 states. According to a DOD official, most children of military
parents in the U. S. attend public schools.

The academic achievement of many BIA students falls far below that of public
school students. As in most public schools, nearly all BIA teachers are
certified and class sizes in BIA schools are smaller than national averages,
even though officials report some difficulty recruiting staff. BIA students
have greater access to computers and Internet connections than public school
students generally, though a smaller proportion of BIA schools than public
schools has technology support staff to maintain equipment and assist
teachers in integrating technology into classrooms. BIA administrators
report extensive facilities problems and agency records show a deferred
maintenance and repair backlog approaching $1 billion. Estimated per- pupil
expenditures vary widely among different types of BIA schools, such as
boarding schools and day schools, and are higher than per- pupil
expenditures at public schools. It is difficult to draw conclusions about
differences in expenditure levels between BIA and public schools, in part
because of cost factors that affect some BIA schools, such as higher
proportions of students who have special needs, broader infrastructure
responsibilities, longer distances to transport children, and the cost of
residential programs in boarding schools.

In school year 1999- 2000, BIA students scored far below public school
students on state assessments in North Dakota, South Dakota, and Arizona-
three states with large numbers of BIA schools. 10 As shown in table 1, the
average national percentile rankings for BIA students on North Dakota?s
statewide assessment, for all grades and subject areas tested, ranged from
the 25th to the 33rd percentiles in school year 1999- 2000. (By definition,
the national average on the norm- referenced tests used in North Dakota and
South Dakota is the 50th percentile.) The corresponding averages for public
school students in North Dakota ranged between the 64th and 71st
percentiles. Similarly, in South Dakota, the average national percentile
rankings for BIA students on South Dakota?s statewide assessment were much
lower than the averages for public school students

10 We were able to obtain data for all BIA schools in North Dakota and South
Dakota and for 19 of the 50 BIA schools in Arizona that participated in the
state?s assessment program. About 36 percent of all BIA students attend the
schools to which our data relate. Data were not available for BIA schools in
New Mexico, which has about one- quarter of all BIA students. The BIA School

System Academic Performance of Many BIA Students is Below That of Public
School Students

Page 13 GAO- 01- 934 BIA and DOD Schools

in the state, both overall and for specific skill areas such as language
arts, reading, and math. Finally, in Arizona, a smaller proportion of BIA
students met state standards than public school students. In school year
1999- 2000, in reading, writing, and math, at each grade level tested, the
proportion of BIA students who met or exceeded the standards was far lower
than the proportion of public school students. (See table 1.)

Table 1: Performance of Students in BIA and Public Schools on State
Assessments in School Year 1999- 2000

Average performance of BIA students Average performance of public

school students

North Dakota a 25th- 33rd percentile 64th- 71st percentile South Dakota b
25th- 28th percentile 60th- 67th percentile Arizona c 0- 27 percent met or
exceeded

standards 15- 71 percent met or exceeded standards

Note: The table shows the range of scores on these assessments for all grade
levels and subject areas tested. a North Dakota uses the TerraNova Basic
Multiple Assessments as an assessment tool at grade 4, and the TerraNova
Complete Battery Plus at grades 6, 8, and 10. Scores shown are total scores.
b South Dakota uses the Stanford Achievement Test as an assessment tool at
grades 2, 4, 8, and 11.

Scores shown are for the complete battery. c Arizona uses an assessment test
called the Arizona Instrument to Measure Standards (AIMS) in

grades 3, 5, 8, 10, and 11. AIMS classifies students into four categories in
relation to the standards- Falls Far Below, Approaches, Meets, or Exceeds.

For Arizona and North Dakota, data for public school students by ethnicity
were available. These data indicate that Indian students in public schools
in these two states score below state averages but higher than BIA students
in those states.

In addition to standardized testing, BIA schools assess students?
proficiency in language arts and math using a combination of other measures
including ?authentic assessment.? Authentic assessment involves evaluating
student achievement based on a review of multiple items such as portfolios
illustrating students? work, grades, and work samples as well as teacher
observations. Many school districts and states have been incorporating
aspects of authentic assessment into their student assessment programs.
Authentic assessment often requires students to demonstrate problem- solving
skills and is thought to reflect real- world situations. Some researchers
consider authentic assessment more appropriate for minority students than
standardized testing because standardized tests have been criticized as
being culturally biased. In particular, because of the diversity of
languages and cultures among

Page 14 GAO- 01- 934 BIA and DOD Schools

Indian students, some educators have found these authentic assessment
methods more useful. These assessments are reported annually by all
BIAfunded schools. Using authentic assessment approaches, about half of BIA
students have been assessed as proficient or advanced, in both math and
language arts, each year over the last 3 years.

Other measures often used as indicators of students? academic success
include dropout rate, college admissions test scores, and the proportion of
students planning to attend college. On these measures, BIA students perform
less well than public school students. For example, in their annual reports,
BIA high schools report dropout rates- the proportion of high school
students who leave school and do not enroll in another school during the
year- averaging about 10 percent. This is higher than the national average
dropout rate (about 5 percent) and slightly higher than dropout rates for
other ethnic minority groups. In addition, BIA students score significantly
below national averages on college admission tests. As shown in table 2, BIA
students who take the Scholastic Assessment Test (SAT) and the American
College Test (ACT)- two widely used college entrance examinations- score
below students nationally in both verbal/ English and math assessments. They
also score below students from low- income families on these tests. Finally,
in our survey, BIA school officials estimated that about 28 percent of their
graduates planned on enrolling in college after graduation, about evenly
split between 4- year and 2- year colleges. Nationally, in 1999, a little
less than two- thirds of high school graduates were enrolled in college the
year after they completed high school- about 40 percent in 4- year colleges
and 24 percent in 2- year colleges. For students from low- income families,
the percentage who were enrolled in college the year after they graduated
was around 50 percent.

Table 2: 1999- 2000 BIA Graduates? Performance on SAT and ACT Compared With
National Averages

SAT scores a ACT scores b Verbal Math English Math

BIA students 383 382 14.7 15.9 National average 505 514 20.5 20.7

Note: A small number of BIA seniors (fewer than 70) take the SAT. Far more
BIA students (about 800) take the ACT because BIA schools are located in
areas of the country where the ACT is the predominant college admission
test. a Data provided by the College Board.

b Data provided by the American College Testing Service.

Page 15 GAO- 01- 934 BIA and DOD Schools

While the academic achievement of BIA students is low relative to students
nationally, to some extent the performance of BIA students may be associated
with conditions that are beyond a school system?s control. Higher student
outcomes have been strongly associated with higher educational and income
levels of parents. BIA students come from environments where family
education levels are low and unemployment and poverty rates are high. For
example, 1990 Census Bureau data show that one- third of Indians age 25 or
older do not have high school diplomas compared with one- quarter of all
adults nationally; and for Indians living on reservations or trust lands,
the proportion of adults who are not high school graduates is over 45
percent. Indians living on or near reservations also experience high rates
of unemployment. Data from BIA?s Indian Labor Force Report show that 43
percent of the potential labor force living on or near reservations remained
unemployed in 1999. Poverty rates are also high among Indian families, with
27 percent having incomes below the poverty level, compared with 10 percent
of all families. Other data suggest that the proportion of families with
incomes below the poverty level is even higher for those living on
reservations and trust lands. Finally, the issue of alcohol and substance
abuse is significant for Indian communities; the death rate due to
alcoholism is about 7 times higher for Indians than for all races. Parental
substance abuse has been identified as an environmental risk factor
associated with learning problems, learning disabilities, and developmental
delays in children. In a previous Education survey, almost 60 percent of BIA
teachers considered parental alcohol or drug abuse a serious problem in
their schools, compared with about 13 percent of teachers in public schools
with low Indian student enrollment. 11

BIA officials noted that the agency has developed programs to begin to
moderate the influences of economic and family conditions on students?
academic performance. These include early childhood and family literacy
programs as well as a therapeutic residential model aimed at achieving
positive changes in attitude, behavior, and the academic performance of
students attending boarding schools. In addition, in its Annual Performance
Plan, BIA has established several goals that address student academic
performance. These goals concern student proficiency in language arts and
math, student attendance, teacher proficiency in new assessments and
technology, school accreditation, and number of degrees

11 NCES, Characteristics of American Indian and Alaska Native Education:
Results From the 1990- 91 and 1993- 94 Schools and Staffing Surveys, NCES
97- 451 (Washington D. C.: U. S. Department of Education, March 1997).

Page 16 GAO- 01- 934 BIA and DOD Schools

conferred on Indian students at tribally controlled community colleges and
at BIA?s two post- secondary schools. 12

According to the National Center for Education Statistics (NCES), one of the
major elements that characterizes teacher quality is teacher preparation and
qualifications. This refers to certification, education, and continuing
learning. Results from our survey indicate that nearly all teachers in BIA
schools were certified in school year 1999- 2000. Typically, about 95
percent of teachers in BIA schools were fully certified in the subjects or
grades they taught and another 3 percent had temporary or provisional
certification. About 92 to 93 percent of public school teachers in general
were fully certified in the subjects they taught. 13 Teachers in BIA-
operated schools were slightly more likely to be fully certified than
teachers in tribally operated schools. In addition to certification,
teachers are being encouraged to pursue advanced degrees in order to gain a
more advanced understanding of their discipline. About one- fourth of
teachers in BIA schools had advanced degrees, compared to about 46 percent
of public school teachers generally. 14 The proportion of teachers with
advanced degrees in BIA- operated schools was about 33 percent, while in
tribally operated schools it was about 21 percent.

BIA teachers have access to various kinds of professional development and
BIA schools support professional development in several ways. For example,
93 percent of the BIA schools that responded to our survey reported that the
majority of their teachers received in- service training provided by the
school during the 1999- 2000 school year. Other types of training widely
used included in- service training provided by the agency and workshops
provided by professional associations, as shown in figure 5.

12 BIA receives funds for two post- secondary schools: Haskell Indian
Nations University and the Southwestern Indian Polytechnic Institute. 13
1998 data as reported in The Condition of Education 2000, Table 47- 3, NCES
(Washington, D. C.: U. S. Department of Education, June 2000). 14 1998 data
from Fast Response Survey System as reported in The Condition of Education
2000, Table 47- 1, NCES (Washington, D. C.: U. S. Department of Education,
June 2000). Teacher Staffing at BIA

Schools Is Comparable to That of Public Schools Despite Some Recruiting
Difficulties

Page 17 GAO- 01- 934 BIA and DOD Schools

Figure 5: Percentage of BIA Schools Reporting That a Majority of Teachers
Received Various Types of Training

Source: GAO survey.

BIA schools facilitate their teachers? professional development in a variety
of ways. The following supports were each reported by nine out of ten survey
respondents: permitting time off to participate in training, setting days
aside for training, paying travel or per diem for training, and paying
tuition or fees. Such training and supports for professional development are
comparable to the situation for public school teachers generally, nearly all
of whom report having some professional development in the previous year
with similar kinds of support from their schools. 15

While nearly all teachers at BIA schools were certified, officials at some
schools we visited, particularly tribally operated schools, recounted some
difficulties recruiting and retaining qualified staff. At one tribally
operated school in Washington, the chief school administrator said it was
difficult to attract teachers because the school could not compete with

15 NCES, Schools and Staffing Survey, 1993- 94, a national sample survey
(Washington, D. C.: U. S. Department of Education).

010 20 30 40 50 60 70 80 90100 College Courses in Teaching Methods

or Assessment College Courses in Subject Matter

Professional Association Workshop In- Service Provided by Agency

In- Service Provided by School 93 66 38 11 13 Percentage of Schools Type of
Training

Page 18 GAO- 01- 934 BIA and DOD Schools

compensation packages offered by nearby urban districts. 16 The 2000- 2001
school year was the first year the school was able to pay teachers at the
state salary scale, but the school offered limited retirement benefits. As a
result, the administrator said the school tended to attract beginning
teachers or retired teachers who had pensions from their public school
careers. Other officials noted that the remote locations of some schools
hinder recruiting. Many BIA schools are located in settings with few
amenities to attract teachers and other staff. For example, at one school we
visited in Arizona, the closest town offering major shopping and banking
services was 96 miles away, and the principal said the community?s isolation
and lack of amenities contributed to teacher turnover. In many locations,
basic housing is not even available for staff to purchase or rent, so BIA
has constructed employee quarters for which staff pay rent. About one- third
of all BIA schools and 45 percent of tribally operated schools had teacher
turnover rates of 25 percent or higher in school year 1999- 2000; the
average turnover rate was 18 percent. Nationwide, about 87 percent of full-
time public school teachers remain at the same school from one year to the
next, implying a turnover rate of about 13 percent. 17

A teacher?s ability to effect student learning may be influenced by the
number of students in the classroom, and various studies have associated
smaller class sizes with higher student achievement. Class sizes reported by
BIA administrators are generally smaller than national averages, with
tribally operated schools reporting smaller classes than BIA- operated
schools. Nationally, in 1998 the average class size for general elementary
classrooms was 23 students and in departmentalized settings (those in which
a teacher?s main assignment is in one particular subject area, such as
English or social studies) it was 24 students. 18 The median number of
students per classroom reported by BIA- operated schools was about 20 to 22
in grades K through 8 and about 17 in grades 9 through 12. Tribally operated
schools had about 17 to 18 students per classroom in grades K

16 BIA- operated schools use a uniform salary schedule comparable to the
salary schedule used by the Department of Defense. This was implemented in
part in an effort to recruit and retain quality staff. However, tribally
operated schools determine salaries independently and there are no summary
data available about their teachers? salaries.

17 NCES, Schools and Staffing Survey, 1993- 94 and Teacher Follow- up
Survey, 1994- 95

(Washington, D. C.: U. S. Department of Education). 18 NCES, Teacher
Quality: A Report on the Preparation and Qualifications of Public School
Teachers, NCES 1999- 080 (Washington, D. C.: U. S. Department of Education,
Jan. 1999). Departmentalized settings are most often at the secondary level.

Page 19 GAO- 01- 934 BIA and DOD Schools

through 6 and about 16 students per classroom in grades 7 through 12. Our
survey did not ask specifically about multiple grades in a classroom.
However, some BIA schools are very small and have fewer full- time teachers
than grade levels taught in the school. For example, there are five BIA
schools with 51 or fewer students in grades ranging from K to 6 and K to 12.
Four of these five schools have fewer full- time teachers than the number of
grades in the school, suggesting that multiple grades are being taught in
some classrooms.

BIA schools appear to be in step with schools around the country in making
technology available as an integral tool for learning. Based on our survey
results, BIA students have greater access to computers than public school
students nationwide. BIA administrators reported having about 1
instructional computer for every 3.5 students overall, although access
levels varied somewhat from school to school, ranging from 1 computer for
every student to 1 computer for every 18 students. In comparison, the
national average for public schools is 1 computer for every 5 students, as
reported by NCES for 2000. 19 A major goal set by Education is that schools
have 1 modern multimedia computer for every 5 students. 20 We do not know
about the quality or capacity of the computers in BIA or public schools, but
46 percent of the computers in BIA schools were reported to be 3 years old
or older.

In some respects BIA schools provide their students with greater access to
the Internet than public schools. According to a BIA official, all 171 BIA
schools are connected to the Internet. Nationally, nearly all public schools
now have Internet access. However, BIA school officials reported that 1
computer with Internet access was available for every 4.3 students, a ratio
appreciably better than the 1 to 7 ratio for public schools nationwide. Some
policymakers have stressed the importance of not only connecting every
school to the Internet but eventually connecting every classroom. A little
over 80 percent of BIA instructional rooms have at least one computer with a
connection to the Internet. This number is also slightly above the national
average for public schools. However, as with some public schools, some BIA
schools still use the lowest- speed connections

19 NCES, Internet Access in U. S. Public Schools and Classrooms: 1994- 2000,
NCES 2001071 (Washington, DC: U. S. Department of Education, May 2001). 20
U. S. Department of Education, 1999 Performance Report and 2001 Annual Plan

(Washington, D. C.: 2000). Access to Computers and

the Internet Reported by BIA Schools Is Greater Than Public Schools, but
Technology Support Is More Limited

Page 20 GAO- 01- 934 BIA and DOD Schools

for their primary connection to the Internet, and thus cannot access complex
computer applications and resources. Table 3 summarizes the information on
students? access to technology at BIA schools based on our survey and for
public schools nationally.

Table 3: Technology Statistics for BIA and Public Schools BIA schools a
Public

schools b

Number of students per instructional computer 3.5 5 Number of students per
instructional computer with Internet access 4.3 7 Percentage of schools
connected to the Internet 100 98 Percentage of instructional rooms with
Internet access 81 77 Percentage of schools using dial- up connections to
the Internet c 10 11 a Data for BIA schools come from GAO?s survey, with the
exception of the percentage of schools connected to the Internet, which was
provided by a BIA official. Data represent technology access in BIA schools
at the time of our survey, which was administered in Nov. 2000. b Public
school data are for school year 1999- 2000 and come from NCES? Internet
Access in U. S.

Public Schools and Classrooms: 1994- 2000, NCES 2001- 071 (Washington, D.
C.: U. S. Department of Education, May 2001). c Other types of connections
schools may be using include dedicated line, cable modem, and wireless
connection.

While computers and the Internet are generally available at BIA schools, a
smaller proportion of the schools have paid (versus volunteer) technology
support staff than public schools. In a survey conducted by Education,
nearly all (95 percent) of public schools reported having some paid staff,
either full- or part- time, to support advanced telecommunications (for
example, networked computers and computer hardware and software) in their
school in 1999. In our survey, we asked whether schools had (1) staff who
provided technical support or performed maintenance on computers and (2)
staff who assisted teachers in using computer software or the Internet to
instruct students. The results indicated that more than onequarter of BIA
schools did not have paid technical support and about onethird did not have
paid staff to help teachers use computers for instruction in 2000. Evidence
suggests that technology support of this kind has a positive impact on
teacher use of technology in the classroom. 21 According to an NCES report,
research in general shows that public school teachers

21 A. Ronnkvist, S. Dexter, and R. Anderson, Technology Support: Its Depth,
Breadth, and Impact in America?s Schools, Teaching, Learning and Computing:
1998 National Survey

(Irvine, Calif.: June 2000).

Page 21 GAO- 01- 934 BIA and DOD Schools

view the lack of technology support for integrating telecommunications into
the classroom and the lack of technical assistance in schools to be
obstacles to their effective use of technology. 22 Our survey results show
that these areas of support are not available in many BIA schools. One
reason why technology staff is more limited in BIA schools, according to one
BIA official, is that the high turnover rate for technology coordinators is
a problem throughout the BIA school system. He indicated that these
technology coordinators often receive training for the work and then leave
the schools for higher- paying positions.

BIA school administrators reported that their computers are used regularly
for learning activities. Although there is evidence to suggest that the
effective use of technology in the classroom can improve student learning,
the research on what types of classroom activities are best served by
technology applications is not conclusive. 23 Nonetheless, studies on the
uses of technology in schools tend to distinguish more basic uses of
computers- such as for drill and practice, which the research shows can be
effective for learning- from more pedagogically complex uses of computers-
such as using the Internet in small groups for class projects, where the
research data are less extensive. In our survey, we asked school
administrators to estimate the portion of their teachers who routinely
engaged their students in the use of computers for certain activities. As
shown in figure 6, many of the teachers at BIA schools regularly engage
students in the use of computers for practice drills and for learning
subject matter. Fewer BIA teachers are using them on a regular basis for
student research or to develop student skills in problem solving.

22 NCES, Teachers? Tools for the 21st Century, A Report on Teachers? Use of
Technology,

NCES 2000- 102 (Washington, D. C.: U. S. Department of Education, Sept.
2000). 23 H. Wenglinsky, Does It Compute? The Relationship Between
Educational Technology and Student Achievement in Mathematics (Princeton, N.
J.: Educational Testing Service, Sept. 1998).

Page 22 GAO- 01- 934 BIA and DOD Schools

Figure 6: Percentage of BIA Schools Reporting That a Majority of Teachers
Assign Students Computer or Internet Work at Least Once Per Week for Various
Classroom Activities

Source: GAO survey.

BIA schools appear to be making substantial investments in school
technology. Administrators reported to us that in 1999- 2000 they spent an
average of $401 per student to support educational technology in their
schools. In comparison, according to a national survey, public school
districts planned to spend $113 per student on educational technology in the
2000- 2001 school year. 24 These expenditures covered all aspects of
supporting educational technology, such as purchasing computer hardware and
software, installing and upgrading local and wide area networks, paying for
technical support and training for teachers, and paying for
telecommunications access.

24 Quality Education Data, Inc., Technology Purchasing Forecast 2000- 2001,
6th Edition

(Denver, Colo.: 2000).

010 20 30 40 50 60 70 80 90100 Problem Solving or Data Analysis

Research Learning Subject Matter

Computer Applications Practice Drill Work 40

38 52 32 17 Percentage of Schools Classroom Activities

Page 23 GAO- 01- 934 BIA and DOD Schools

Most of the BIA school administrators responding to our survey reported
significant problems with the condition of their facilities. They noted
problems with specific building features, such as roofs and plumbing, and
with certain environmental conditions, such as indoor air quality. In
addition, they indicated that work spaces for teachers and students, such as
classrooms and libraries, were insufficient. The extent of the concerns they
reported in the survey are reflected in BIA?s backlog of maintenance and
repair needs. Among the backlog of such needs, which was recently verified
by an independent engineering firm, is more than $960 million in needed
repairs for school facilities and dormitories. A contributing factor to the
facility needs of BIA schools is their isolation. Many schools are located
in remote areas where, in addition to maintaining school buildings and
grounds, facilities managers must also maintain other components of the
infrastructure such as water and sewer systems, electric utilities, and fire
stations- systems normally provided by local municipalities for most public
schools. The Congress has recently increased funding used to address the
backlog. However, budget allocations for the maintenance and repair of
facilities have generally been less than amounts recommended by national
guidelines.

Compared to an earlier survey of administrators in public schools,
significantly more BIA school administrators responding to our survey
reported that the overall condition of their buildings and specific building
features was inadequate. As shown in table 4, administrators at 65 percent
of BIA schools reported that one or more of their school buildings was in
less- than- adequate condition and 76 percent of boarding schools reported
the same for their dormitory buildings. 25 In comparison, in a 1999
Education survey on the condition of public school buildings, 26
administrators at about one- quarter of public schools reported that one or
more of their buildings was in less- than- adequate condition. Looking back

25 Our survey asked school administrators to rate the condition of original
buildings, attached or detached permanent additions to original buildings,
and temporary buildings such as portables or trailers.

26 NCES, Condition of America?s Public School Facilities: 1999, NCES 2000-
32 (Washington, D. C.: U. S. Department of Education, June 2000). Extensive
Problems With

BIA School Facilities Reported

BIA School Buildings, Features, and Work Spaces Reported To Be Less Than
Adequate

Page 24 GAO- 01- 934 BIA and DOD Schools

to reported conditions in 1994, the conditions of both BIA and public
schools appear to have generally improved. 27

Administrators also reported that specific features of school buildings and
dormitories, such as the plumbing system or roof, needed substantial repair
or replacement. Overall, more than 70 percent of responding BIA schools
reported that one or more of the features of school buildings was in less-
than- adequate condition and 87 percent of boarding schools reported the
same for dormitory buildings. In contrast, administrators at half of public
schools reported that one or more of the features in their school buildings
was inadequate. As shown in table 4, for many of the features listed, BIA
administrators were at least twice as likely as public school administrators
to report that their school buildings and dormitories had inadequate
building features.

27 GAO conducted a survey of public school and BIA school building
conditions in 1994. Our current survey (2000) asked questions similar to
those used in our 1994 survey and Education?s 1999 survey of public schools.
For results from our previous study, see School Facilities: Condition of
America?s Schools (GAO/ HEHS- 95- 61, Feb. 1995) and School Facilities:
Reported Condition and Costs to Repair Schools Funded by the Bureau of
Indian Affairs (GAO/ HEHS- 98- 47, Dec. 1997). It should be noted that the
results of our 1994 study of BIA schools reflect the reported conditions of
responding schools; survey response rates were too low (41 percent) to
permit generalization to all BIA schools.

Page 25 GAO- 01- 934 BIA and DOD Schools

Table 4: Reported Facility Deficiencies for BIA Schools and Dormitories and
Public Schools Percentage of schools reporting inadequacy Building
deficiencies BIA schools a BIA dormitories a Public schools b Onsite school
buildings

One or more buildings in less- than- adequate condition (original buildings,
additions to original buildings, and temporary buildings) 65 76 24

Features of onsite buildings

One or more inadequate building features 72 87 50 Roofs 48 50 22 Framing,
floors, foundation 41 61 14 Exterior walls, finishes, windows, doors 49 57
24 Interior finishes, trim 39 58 17 Plumbing 47 66 25 Heating, ventilation,
air conditioning 54 76 29 Electric power 43 66 22 Electrical lighting 34 65
17 Safety features such as sprinklers, fire alarms, smoke detectors 37 53 20

a Source: GAO survey. b Source: NCES, Condition of America?s Public School
Facilities: 1999, NCES 2000- 32 (Washington, D. C.: U. S. Department of
Education, June 2000).

Administrators responding to our survey also indicated that some
environmental conditions were a problem in their schools. For example, over
40 percent reported unsatisfactory indoor air quality, acoustics or noise
control, or physical security in their buildings. Administrators at over 70
percent of responding boarding schools reported the same for dormitory
buildings. In comparison, these environmental conditions were a problem for
20 percent or less of public schools in 1999.

During our site visits, school officials indicated various problems with
their school buildings. For example, the principal of a school in Washington
said that in one portable classroom building the roof and windows leaked,
and in two buildings there was a continuous problem with mold in the walls,
which was difficult to control due to the wet climate. She said that the
mold was a health problem for children with certain allergies. In a South
Dakota school, a teacher expressed concerns about aging asbestos floor tiles
in her classroom that had to be partially removed due to cracking, leaving
other tiles exposed. Asbestos floor tiles, which are present in more than 90
of BIA?s schools, can be a threat to the

Page 26 GAO- 01- 934 BIA and DOD Schools

health of students and staff if the asbestos fibers are disturbed, released
into the air, and inhaled. 28

In addition to questions about the condition of buildings and aspects of
their environment, our survey asked administrators about the sufficiency of
work spaces for students and teachers, such as classrooms, libraries, labs,
and eating areas. We also asked whether some of these work spaces were
adequately equipped- for example, whether libraries had sufficient numbers
of books or whether buildings were adequately wired to support technology
needs. As shown in figure 7, many BIA school administrators responding to
our survey reported that various work spaces and equipment were insufficient
at their school. For example, administrators at 55 percent of responding
schools reported that the number of classrooms was insufficient and over 60
percent reported the space in their libraries was a problem. Insufficient
classroom space was noted by officials at some schools we visited. For
example, at a large K- 12 school in South Dakota, officials said that the
number of students in the high school building exceeded capacity. According
to school officials, the building was intended to house 250 students, but
they currently had about 400 students enrolled. In an Arizona school,
because there was no other space, the school held special education classes
in an area previously used for storage and teachers used partitions to
separate the different groups of students.

28 Asbestos is a fibrous mineral that was used widely from the 1950s through
the early 1970s in construction materials, such as shingles for roofing and
siding, pipe and boiler insulation, floor tiles, ceiling panels, coatings,
and gaskets. It is present in most BIA schools in the form of floor tiles
and/ or insulation. A danger occurs when tiny asbestos fibers are released
into the air and can be inhaled or swallowed, increasing the potential for
conditions such as lung cancer.

Page 27 GAO- 01- 934 BIA and DOD Schools

Figure 7: Percentage of BIA Schools Reporting That Various Work Spaces and
Equipment Are Insufficient

Source: GAO survey.

A large proportion of BIA schools (60 percent) also reported that the
capacity of the building?s telephone and electrical wiring systems to
support computers, the Internet, and telephones was inadequate for their
buildings. For BIA schools with insufficient telephone and electrical
wiring, the age of the buildings may be a factor. Almost half of the
buildings are more than 30 years old and almost 15 percent are between 50
and 100 years old. The principal of an Arizona school built in the 1930s
said that most of the electrical wiring in the school is original and as new
technology has been added, they have had frequent problems with system
overloads. He said that this is likely to get worse as more new technology
is added.

010 20 30 40 50 60 70 80 90100 Telephone and Electrical Wiring

Size of Eating Area Number of Science Labs

Number of Computer Labs Number of Library Books

Library Space Size of Classrooms

Number of Classrooms 55 29

62 64 38 31

44 60

Percentage of Schools Work Space and Equipment

Page 28 GAO- 01- 934 BIA and DOD Schools

Administrators? concerns about the condition of their facilities are
reflected in the BIA backlog of deferred maintenance and repairs. 29 This
backlog is a catalog of deficiencies- related to buildings, surrounding
grounds, and other infrastructure- identified by architects, engineers,
facilities managers, and safety officers over the years. The inventory
describes in detail individual work items required in schools to meet
national standards and codes such as the Uniform Building Code and National
Fire Codes. The backlog contains a description of the work that needs to be
done and the estimated cost for each item. The deficiencies may involve
safety and health, access for persons with disabilities, or noncompliance
with building codes. Recently, BIA?s backlog was validated by an independent
engineering firm. 30 According to BIA officials and representatives from the
engineering firm, this process included on- site inspections of each BIA
school building. Based on their inspections, the engineers validated current
projects and their cost estimates, removed items that had already been
addressed, and added new deficiencies they identified.

As of February 2001, BIA?s backlog for educational facilities totaled $962
million 31 -an increase of almost $270 million (in current dollars), or 39
percent, since October 1997. 32 Table 5 shows estimated costs from BIA?s
backlog to address deficiencies related to certain building features at BIA
schools. BIA school administrators often indicated problems with these
features in our survey. For example, the cost to address all deficiencies

29 Maintenance is generally defined as the upkeep of property and equipment
and includes the adjustment, cleaning, or lubrication of equipment;
replacement of parts; painting; and other actions to assure continuing
service and prevent breakdown. Deferred maintenance is maintenance that was
not performed when it should have been or when it was scheduled to be and
which, therefore, is put off or delayed for a future period. Repair is work
to restore damaged or worn- out property to a normal operating condition.
Repairs are curative, while maintenance is preventive.

30 The review of BIA?s backlog was conducted by Applied Management
Engineering, Inc. (AME). In addition to the validation of the backlog, AME
developed a process for BIA to ensure the consistency and quality of
information going into the backlog. AME is also conducting a study on the
adequacy of operations and maintenance funding for BIA schools and an
assessment of their facilities with respect to educational program needs,
according to AME?s general manager.

31 According to a BIA official, in some cases, the backlog may include both
the deficiencies for a certain building and the estimated cost for planning,
design, and/ or construction to replace that same building.

32 See School Facilities: Reported Condition and Costs to Repair Schools
Funded by Bureau of Indian Affairs (GAO/ HEHS- 98- 47, Dec. 31, 1997). BIA
Reports Needing More

Than $960 Million To Address Educational Facilities Deficiencies but Funding
Has Been Increasing

Page 29 GAO- 01- 934 BIA and DOD Schools

related to heating, ventilation, and air conditioning systems in schools
totaled more than $63 million at the time of this review. Administrators at
more than half of the schools responding to our survey indicated problems
with these systems. One school we visited in Arizona did not have a central
cooling system, so each classroom had an air conditioner or evaporative
cooling unit. The principal said the units were quite old and so noisy that
they disrupted the students? learning environment. 33

Table 5: Selected Reported Deficiencies and Estimated Repair Costs for BIA
Schools and Dormitories

Building features often indicated as inadequate a Estimated cost to repair
related

deficiencies b (dollars in millions)

Heating, ventilation, air conditioning $63.2 Plumbing 31.6 Roofs 24.5 a
Source: GAO survey.

b Source: BIA backlog data as of February 2001.

More than $127 million of the backlog total represents deficiencies related
to the health and safety of students, and a significant portion of this ($
44 million) relates to fire safety alone. For example, the backlog lists
buildings at more than 100 schools that need to have their fire alarm
systems replaced or upgraded because they are old, not working, or missing.
At one school we visited, the facilities manager said that the fire alarm
system on the school campus was no longer reliable, and that it was so old
that buying replacement parts was becoming difficult. With respect to
sprinkler systems, the BIA Chief of Safety and Risk Management said that it
is BIA?s goal to have sprinkler systems and smoke detectors in all
dormitories to ensure the safety of the residents. However, at the time of
our review, the backlog showed that more than half of the 54 boarding
schools? dormitories needed sprinkler systems either to be installed or
upgraded. None of the boarding schools we visited had sprinkler systems in
student dormitories. In addition, a dormitory at one school was closed due,
in part, to fire- related safety deficiencies, according to the school
superintendent.

33 Some studies have concluded that school facility conditions, such as
inadequate ventilation, faulty heating systems, and excessive noise, can
affect student learning as well as the health and morale of staff and
students. However, overall, the research is inconclusive and many scholars
say that more research is needed to establish a link between school facility
conditions and student performance.

Page 30 GAO- 01- 934 BIA and DOD Schools

A contributing factor to BIA school facility needs and the backlog is that
BIA schools have to support a more extensive infrastructure than most public
schools. Because many BIA schools are located in isolated areas, they
maintain their own water and sewer systems, electric utilities, fire
stations, and other important services that are generally provided to public
schools by municipalities. Among the 171 BIA schools and 14 peripheral
dormitories, 146 sites (79 percent) had responsibility for maintaining some
kind of water infrastructure such as wells, water distribution lines, and/
or water treatment facilities; 138 (75 percent) had responsibility for sewer
infrastructure, such as sewage lagoons and sewage lines; and 52 schools (30
percent) operated and maintained fire trucks.

BIA officials indicated that providing and maintaining this additional
infrastructure was a considerable drain on resources. For example, at a K- 8
school we visited in Arizona, the facilities manager said that his staff
must maintain a water system that includes five wells with distribution
lines, pumps, pump houses, and water storage tanks; a sewer system that
includes sewer lines, a lagoon, and a lift station; a fire station, fire
truck, and hydrants; and a landfill. The facilities manager said the school
has chronic problems with the sewer system, and an independent engineering
firm has recommended that the sewer mains and service lines be replaced. For
water and sewer systems alone, deficiencies in BIA?s backlog for educational
facilities totaled almost $56 million.

BIA?s infrastructure also includes dormitory facilities for boarding
students and housing for staff and their families. In total, BIA schools
support 157 dormitory buildings for students and 1,879 single- family
quarters and 194 apartment buildings for employees. However, we did not
include employee housing in the scope of our study because the rent paid by
staff is meant to cover the upkeep of these facilities.

Funding from BIA?s construction budget that is used largely to address the
backlog was increased substantially by the Congress last year. This funding
more than doubled from $40 million in fiscal year 2000 to $110 million in
fiscal year 2001. BIA uses this funding to support a wide range of facility-
related activities, including minor projects, such as the replacement of
boiler systems, and major projects, such as the replacement of a natural gas
line or the renovation of a school office building. In addition, the
Congress provided even greater increases for replacement school construction
the last 2 years. Funding rose from $17 million in fiscal year 1999 to $141
million in fiscal year 2001. This funding can reduce the backlog when new
facilities are constructed to replace

Page 31 GAO- 01- 934 BIA and DOD Schools

those that cannot accommodate current education programs or are beyond
repair.

While funding has increased during the last few years for certain
facilityrelated line items in BIA?s budget, we attempted to look more
closely at funding available specifically for maintenance and repair needs.
Adequately funding maintenance and repair is important because deferral of
these tasks can increase the backlog. One guideline set forth by the
National Research Council (NRC) 34 considers maintenance and repair budgets
in terms of the current replacement value of buildings. 35 The NRC has
recommended that budget allocations for routine maintenance and repair be in
the range of 2 to 4 percent of the current replacement value of buildings.
36

Using NRC?s definitions, we extracted and combined pieces of BIA?s budget to
identify funding available for the maintenance and repair of buildings and
other infrastructure and determined that funding levels were below those
recommended by the NRC. From 1997- 2000, funding requested for and allocated
to the maintenance and repair of BIA schools was between 1.5 percent and 1.8
percent- below the lower end of the recommended range of 2 to 4 percent.
Funding increased in fiscal year 2001, resulting in maintenance and repair
budget allocations providing about 2.5 percent of current replacement value
that year. However, based on discussions with recognized experts in
budgeting and facilities operations and maintenance, even with this
increase, budget allocation levels over the past 5 years have been too low.
37 These experts suggested that because of the extensive infrastructure
beyond buildings that BIA must support, the isolation of the schools which
dramatically increases maintenance and repair costs, and the age of BIA
facilities, BIA schools

34 NRC is the principal operating agency of both the National Academy of
Sciences and the National Academy of Engineering in providing advisory
services in science and technology to the federal government and other
entities.

35 Current replacement value is the amount in current dollars it would cost
to duplicate the facility. 36 NRC, Committing to the Cost of Ownership:
Maintenance and Repair of Public Buildings (Washington, D. C.: National
Academy Press, 1990). 37 The experts we spoke with included a member of the
National Academy of Engineering, two engineering consultants, and a retired
director of facilities for a historical society with past experience
developing capital and facilities operations budgets for the Congress. BIA
Funding for Maintenance

and Repair Is Less Than National Guidelines

Page 32 GAO- 01- 934 BIA and DOD Schools

likely require maintenance and repair funding in the range of 4 to 6 percent
of current replacement value.

Estimated per- pupil expenditures for BIA day schools were $9,647 in 1997-
98. 38 In comparison, the national average per- pupil expenditure for public
schools was $6,189 in 1997- 98, the latest year for which national data were
available at the time we did our work. It is important to note that per-
pupil expenditures can range widely. Within the 50 states, per- pupil
expenditures for public schools in 1997- 98 ranged from $4,288 in
Mississippi to $9,643 in New Jersey.

Estimated per- pupil expenditures at BIA schools also varied widely
depending upon the type of school. As shown in table 6, BIA boarding school
expenditure estimates were almost $2,000 more per pupil than BIA day schools
overall; the additional estimated expenditures for the boarding schools were
likely due to their residential components. Among BIA day schools, estimated
expenditures per pupil were higher for tribally operated schools than BIA-
operated schools.

Table 6: Comparison of Per- Pupil Expenditures at BIA and Public Schools,
1997- 98 BIA- funded schools

Estimated per pupil expenditure Public schools Per- pupil

expenditure b

Day schools $9,647 All schools $6,189 BIA- operated 7,677 Rural schools
5,817 Tribally operated 10,423 Large city schools 6,293 Boarding schools a
$11,643 a Includes BIA- operated and tribally operated boarding schools.

b Data are from the National Public Education Financial Survey, Fiscal Year
1998, as cited in NCES?

Statistics in Brief: Revenues and Expenditures for Public Elementary and
Secondary Education: School Year 1997- 98, NCES 2000- 348 (Washington, D.
C.: U. S. Department of Education, May 2000).

One reason for the difference between estimated per- pupil expenditures for
day schools operated by BIA and those by tribes is that tribally operated
schools bear higher administrative costs- each school must

38 Our analysis of expenditures at BIA schools considered expenditures from
federal funding sources alone. According to BIA officials, federal funds
constitute by far the largest part of funding for BIA schools. The national
per- pupil expenditure figures include funding from federal, state, local,
and private sources. Estimated Expenditures

Per Pupil for BIA Schools Are Higher Than Public Schools?, and Vary
Substantially Between Categories of Schools

Page 33 GAO- 01- 934 BIA and DOD Schools

provide its own accounting and other support services that otherwise would
be provided by BIA for a comparable BIA- operated school. To support these
additional operating expenses, BIA provides administrative cost grants to
tribally operated schools, which averaged about $1,400 per student in 1997-
98.

In comparing BIA and public school per- pupil expenditures, it is difficult
to draw meaningful conclusions because the many factors that distinguish BIA
schools from public schools may add to their educational costs. Special-
needs students generally require additional educational resources and, as
mentioned previously, a high proportion of BIA students have special needs.
For example, according to BIA records for the 2000- 2001 school year, about
21 percent of BIA students are enrolled in special education compared with
13 percent of public school students, and 58 percent of BIA students have
limited English proficiency, compared with 8 percent of public school
students. 39 Available data also indicate that a high proportion of BIA
students are poor. 40 The student populations of BIA schools in fact may be
more similar to those of large city or rural public schools, where higher
proportions of students tend to have special needs.

Other characteristics of BIA schools may lead to increased expenditures. The
isolation that results in additional infrastructure needs for BIA schools
also results in higher costs in other areas, such as transportation. For
example, the number of miles transported per student for BIA schools in
school year 1999- 2000 was about 296, while that of public schools was about
165 for school year 1998- 1999. Many of the miles traveled by BIA students
are on unimproved roads or roads in poor condition, which increases wear and
tear on vehicles. BIA also provides long- distance transportation four times
a year for many of the 8,000 students living in boarding schools. Because
BIA schools are generally small and often far

39 Data on the proportion of public school students in special education are
for 1998- 99 and are from Digest of Education Statistics 2000 (Washington,
D. C.: U. S. Department of Education, Jan. 2001). Data on the proportion of
public school students who are considered to have limited English
proficiency are for 1996- 97 and are from the National Clearinghouse for
Bilingual Education Summary Report of the Survey of the States? Limited
English Proficient Students and Available Educational Programs and Services
1996- 97 (Washington, D. C.: George Washington University, Graduate School
of Education and Human Development, Sept. 1998).

40 As noted earlier, available data for 123 of the 171 BIA schools show that
over 80 percent of students were eligible for free or reduced- price lunch
for school year 1999- 2000. According to data from USDA, about 40 percent of
students enrolled in schools nationwide are approved for free or reduced-
price lunch.

Page 34 GAO- 01- 934 BIA and DOD Schools

from each other, it can be difficult for them to achieve economies of scale.
Many, for example, must have their own facility maintenance shops and
garages; in contrast, public school districts generally have such facilities
that serve several schools.

DOD students? academic achievement generally exceeds that of elementary and
secondary students as measured by national standardized tests. DOD school
administrators indicated that nearly all teachers in DOD schools are
certified in the subjects or grades they teach, and the majority of teachers
hold advanced degrees- this proportion is greater than the national average
for public school teachers. Students? access to computers reported by DOD
school administrators is greater than that reported for public schools
nationwide. In addition, the vast majority of DOD schools provide their
teachers with technical and instructional assistance for using computers in
the classroom. Many DOD school administrators reported some problems with
their school facilities and some reported that work spaces for teachers and
students were not adequate. However, overall the conditions of facilities
reported by DOD schools and public schools in 1999 were not substantially
different. The estimated per- pupil expenditure for DOD schools is higher
than the national average, and the estimated per- pupil expenditure is
higher for overseas schools than for domestic schools.

On standardized achievement tests, DOD students on average score at or above
national norms. DOD schools use the TerraNova Multiple Assessments to assess
students in grades 3 through 11. In math and language arts, DOD students?
average national percentile rankings ranged from the 61st percentile to the
72nd percentile (see table 7). Overseas and domestic schools scored in a
comparable range. By definition, the national average is the 50th
percentile. At most grade levels, the rankings for overseas schools were a
few points higher than those for domestic schools. For both boys and girls,
and for various ethnic groups, DOD students were at or above national
averages. On the National Assessment The DOD School

System Academic Performance of DOD Students Exceeds That of Students
Nationally

Page 35 GAO- 01- 934 BIA and DOD Schools

of Educational Progress (NAEP), another standardized test, DOD students
scored at or above national benchmarks. 41

Table 7: National Percentile Rankings for Students in DOD Schools on the
TerraNova Multiple Assessments Standardized Tests for School Year 1999- 2000

National percentile rankings Grade Language arts Math

3 6361 4 6464 5 6463 6 6865 7 6665 8 6862 9 7066 10 72 72 11 68 67

Source: DOD data.

On other measures that are often used as indicators of students? academic
achievement, such as college admissions tests and plans to attend college,
DOD students generally perform as well as or better than public school
students. A high proportion of DOD students take college admissions tests,
scoring at or near national averages. (See table 8.) In addition, DOD school
administrators reported that about three- quarters of DOD graduates plan to
pursue higher education. Nationally, just under two- thirds of high school
graduates are enrolled in college the year after they complete high school.
DOD administrators responding to our survey also estimated that about 60
percent of their graduates planned to enter 4- year colleges and another 15
percent planned to enter 2- year colleges. Nationally, about 40 percent of
high school graduates enroll in 4- year colleges the year after they
graduate and about 24 percent enroll in 2- year colleges.

41 The NAEP is a nationally administered test that assesses students?
proficiency in different subject areas, such as reading, math, and writing,
in different years. We examined DOD data on students? performance on the
NAEP in math (1996), reading (1998), and writing (1998).

Page 36 GAO- 01- 934 BIA and DOD Schools

Table 8: Average SAT and ACT Scores for DOD and Public School Students
Graduating in 2000

SAT scores ACT scores Verbal Math English Math

DOD students 504 500 21.3 20.8 National average 505 514 20.5 20.7

Source: DOD data.

The academic achievement of DOD students may be influenced by socioeconomic
factors and a ?common culture? engendered by military service. Education
research has strongly associated student academic achievement with parental
education, income levels, and family structure. In each of these areas,
military families fare relatively well. For example, nearly all military
enlistees are high school graduates and relatively high proportions of the
parents of DOD students have attended or graduated from college. Moreover,
while a considerable portion of children in DOD schools are eligible for
free or reduced- price lunch programs (32 percent), this is less than the
proportion in public schools (40 percent); in addition, military families
receive benefits, such as tax- free housing allowances or free housing on
base, that are not considered in eligibility determinations for the lunch
programs. Finally, nearly 90 percent of military school- age children reside
in two- parent families compared with 70 percent of children in public
schools, although DOD families are affected by the absence of parents
deployed away from home.

In addition to these socioeconomic factors, DOD schools receive unique
support from the military and its culture. Military commands support the
schools in their jurisdictions and encourage their staffs to do so as well-
for instance, by making parent- teacher conferences a duty or fostering
volunteer work in the schools. Staff at DOD schools are sensitive to the
unique circumstances of military families, such as the deployment of
military parents away from home and frequent family moves to new duty
stations. According to a recent study, military children move through an
average of six school districts between kindergarten and the completion of
high school. Therefore, DOD is developing standardized curricula in its
schools that are intended to help overcome some problems associated with
high student mobility.

Page 37 GAO- 01- 934 BIA and DOD Schools

Administrators responding to our survey indicated that virtually all
teachers in DOD schools are certified in the subjects or grades they teach,
both in domestic and overseas schools. About two- thirds of teachers in DOD
schools have advanced degrees compared to about 46 percent of public school
teachers generally. Again, this proportion was about the same for both
domestic and overseas schools.

DOD teachers have access to various kinds of professional development, and
DOD schools support professional development in several ways. For example,
95 percent of the DOD schools that responded to our survey reported that the
majority of their teachers received in- service training provided by the
school during the 1999- 2000 school year. Other types of training are shown
in figure 8.

Figure 8: Percentage of DOD Schools Reporting That a Majority of Teachers
Received Various Types of Training

Source: GAO survey.

DOD schools facilitate their teachers? professional development in several
ways. Nearly all DOD schools reported allowing time off to participate in
training and setting days aside for training. In addition, nearly nine out
of ten schools reported paying travel or per diem for training; about half
reported paying tuition or fees. These types of training and supports for
DOD Teachers Are

Certified and Most Have Advanced Degrees

010 20 30 40 50 60 70 80 90100 College Courses in Teaching Methods

or Assessment College Courses in Subject Matter

Professional Association Workshop In- Service Provided by District

In- Service Provided by School 95 71 6 7 5 Percentage of Schools Type of
Training

Page 38 GAO- 01- 934 BIA and DOD Schools

professional development are comparable to those for public school teachers
generally, nearly all of whom reported having some professional development
with similar kinds of support from their schools. 42

DOD officials related few problems attracting teachers to domestic schools
but noted that recruiting was somewhat more difficult in certain overseas
locations, particularly in parts of Asia. To ease recruiting, DOD often pays
moving expenses for teachers it hires in the U. S. to work in overseas
schools. DOD provides transportation for the employees and their family
members, and ships their household goods to and from the overseas area.
Applicants recruited in the U. S. for overseas posts also receive housing
allowances to cover rent and utility costs. 43

Class sizes reported by DOD administrators are close to national averages
with no consistent pattern between domestic and overseas schools.
Nationally, in 1998, the average class size was 23 students for general
elementary classrooms and 24 students for teachers in departmentalized
settings. The median number of students per classroom reported by DOD
schools was in the low 20s in grades K through 8 and in the high teens in
grades 9 through 12, as shown in table 9.

Table 9: Median Class Sizes at DOD Schools Median number of students per
class Grades

K to 3 Grades

4 to 6 Grades

7 to 8 Grades

9 to 12

Overseas schools 22 24 20 19 Domestic schools 20 23 23 19

Source: GAO survey.

Beginning in school year 1999- 2000, DOD obtained additional funding to
reduce class sizes in grades 1 through 3 in order to enhance learning
opportunities and improve the quality of instruction. The target is a
pupilteacher ratio of 18 to 1. Reduced class sizes were to be implemented in
18 overseas schools and 25 domestic schools, with all DOD schools phased in
over the next 6 years.

42 NCES, Schools and Staffing Survey, 1993- 94 (Washington, D. C.: U. S.
Department of Education). 43 According to a DOD official, a small number of
staff in DOD overseas schools live in government- provided quarters, but the
majority receive allowances intended to cover the cost of housing in the
areas where they choose to live.

Page 39 GAO- 01- 934 BIA and DOD Schools

Based on the responses of DOD school officials to our survey, DOD students
appear to have greater access to computers, including computers connected to
the Internet, than do public school students nationwide. These
administrators reported having about 1 instructional computer for every 3.7
students. This ratio is slightly better than the national average reported
by NCES for 2000. (See table 10.) We do not know about the quality or
capacity of computers in DOD schools, but a significant portion of these
computers (59 percent) are 3 years old or older.

Administrators at DOD schools reported that students had 1 computer
connected to the Internet available for every 4.8 students. This ratio also
is better than the ratio for public schools nationwide (1 computer for every
7 students). In addition, DOD administrators reported that almost 90 percent
of instructional rooms in their schools have at least one computer with a
connection to the Internet. This figure exceeds the national average in 2000
by more than 10 percentage points.

Table 10: Technology Statistics for DOD and Public Schools DOD schools a
Public

schools b

Number of students per instructional computer 3.7 5 Number of students per
instructional computer with Internet access 4.8 7 Percentage of schools
connected to the Internet 100 98 Percentage of instructional rooms with
Internet access 89 77 Percentage of schools using dial- up connections to
the Internet c 5 11 a Data for DOD schools come from GAO?s survey, with the
exception of the percentage of schools

connected to the Internet, which was provided by a DOD official. Data
represent technology access in DOD schools at the time of our survey, which
was administered in Nov. 2000. b Public school data are for school year
1999- 2000 and come from NCES? Internet Access in U. S. Public Schools and
Classrooms: 1994- 2000, NCES 2001- 071 (Washington, D. C.: U. S. Department
of Education, May 2001). c Other types of connections schools may be using
include dedicated line, cable modem, and wireless

connection.

DOD schools also appear to have greater staff support for technology than
public schools nationwide. Almost all DOD school administrators reported
that in 2000 they had paid (versus volunteer) staff that maintained
technology and provided technical assistance. In addition, 98 percent of DOD
schools have paid staff to support teachers in using technology for
instruction. In a survey conducted by Education in 1999, 95 percent of
public schools reported having some paid staff to support advanced
telecommunications in their school. A school technology administrator at
Technology Access and

Support Reported by DOD Schools is on Average Greater Than Public Schools

Page 40 GAO- 01- 934 BIA and DOD Schools

one Army base told us that each school on the base has a technologist
available to instruct teachers on how to use technology in the classroom,
and that, as a result, teachers were integrating the technology into the
curriculum on a daily basis. At one school in this district, first graders
were learning how to create and deliver PowerPoint presentations.

As previously noted, though research suggests that effective use of
technology in the classroom can improve student learning, research is
inconclusive concerning what types of classroom activities are best served
by technology. In responding to our questions concerning teachers? use of
computers in the classroom, about half of DOD school administrators reported
that the majority of their teachers regularly engage students in the use of
computers to learn subject matter, conduct research, and learn computer
applications. Fewer DOD administrators reported that a majority of their
teachers are routinely using computers to develop student problem- solving
or data- analysis skills or for drill and practice purposes. (See fig. 9.)

Figure 9: Percentage of DOD Schools Reporting That a Majority of Teachers
Assign Students Computer or Internet Work At Least Once Per Week for Various
Classroom Activities

Source: GAO survey. 010 20 30 40 50 60 70 80 90100

Problem Solving or Data Analysis Research

Learning Subject Matter Computer Applications

Practice Drill Work 40 47

54 48 25 Percentage of Schools Classroom Activities

Page 41 GAO- 01- 934 BIA and DOD Schools

DOD schools appear to be making significant investments in educational
technology. According to agency records and estimates, the DOD school system
obligated about $356 per student for educational technology in fiscal year
2000 for domestic and overseas schools. By comparison, according to a
national survey, public school districts planned to spend $113 per student
on educational technology in the 2000- 2001 school year. 44

Many DOD school administrators responding to our survey reported problems
with their school facilities but, for the most part, their responses were
not substantially different from those reported nationally by public school
administrators in 1999. 45 As shown in table 11, 32 percent of the school
administrators reported that they had one or more buildings in less than
adequate condition; in 1999, 24 percent of public school administrators
responding to a national survey reported that they had one or more buildings
in less than adequate condition. Looking at specific building features, such
as roofs and plumbing, administrators at about half of both DOD and public
schools reported having at least one inadequate feature. For many features,
about the same proportion of DOD and public schools reported problems. For
some features, including electric power, electrical lighting, and safety
features, fewer DOD schools reported problems than public schools.

44 Quality Education Data, Inc., Technology Purchasing Forecast 2000- 2001,
6th Edition

(Denver, Colo.: 2000). 45 NCES, Condition of America?s Public School
Facilities: 1999, NCES 2000- 32 (Washington, D. C.: U. S. Department of
Education, June 2000). Administrators Report

Some Problems With DOD School Facilities

Page 42 GAO- 01- 934 BIA and DOD Schools

Table 11: Reported Facilities Deficiencies for DOD and Public Schools
Percentage of schools reporting

inadequacy DOD schools a Public

schools b Building deficiencies Onsite school buildings

One or more buildings in less than adequate condition (original buildings,
additions to original buildings, and temporary buildings)

32 24

Features of onsite buildings

One or more inadequate building features 52 50 Roofs 21 22 Framing, floors,
foundation 15 14 Exterior walls, finishes, windows, doors 20 24 Interior
finishes, trim 14 17 Plumbing 22 25 Heating, ventilation, air conditioning
29 29 Electric power 14 22 Electrical lighting 11 17 Safety features such as
sprinklers, fire alarms, or smoke detectors 14 20 a Source: GAO survey.

b Source: NCES, Condition of America?s Public School Facilities: 1999, NCES
2000- 32 (Washington, D. C.: U. S. Department of Education, June 2000).

When asked about the condition of certain environmental factors in DOD
schools, such as indoor air quality, again about the same proportion of
administrators at DOD schools as public schools reported unsatisfactory
conditions. About 20 percent or less of administrators at DOD and public
schools reported that indoor air quality, acoustics or noise control, or
physical security were inadequate in their buildings.

Some DOD schools reported that various work spaces and equipment for
students and teachers were insufficient at their schools: 29 percent of
schools reported that the number of classrooms was insufficient; about the
same proportion reported inadequate library space (see fig. 10). The
principal of a 40- year- old school we visited that had not been renovated
said that the lack of space for storage was a major problem. In addition, 40
percent of DOD schools reported that the telephone and electrical wiring in
their facilities was inadequate. The problems with inadequate telephone and
electrical wiring may be due to the age of DOD education facilities. About
60 percent of the DOD school buildings are 30 to 70 years of age,

Page 43 GAO- 01- 934 BIA and DOD Schools

and as the level of technology introduced into the schools has increased, so
have the demands on the telephone and electrical systems.

Figure 10: Percentage of DOD Schools Reporting That Various Work Spaces and
Equipment Are Insufficient

Source: GAO survey.

DOD estimates that about $529 million is needed to improve its educational
facilities, as of May 2001. This estimate comes from the 5- Year Facilities
Plan for DOD schools. This plan lists projects for repair and improvement
identified by school engineers and principals. 46

Each year, each DOD school principal and a DOD school district facilities
engineer inspect the school facilities and grounds to validate existing
projects on the plan, identify new projects, coordinate routine maintenance,
identify construction projects, and discuss prioritization. Headquarters-
level facilities officials prioritize the projects for DOD schools
systemwide and determine which projects will be funded based on

46 The 5- Year Facilities Plan for DOD schools also includes the cost to
replace some schools. At the time of this review, the plan included the
costs to replace six schools. DOD Reports Needing More

Than $500 Million to Improve Educational Facilities

010 20 30 40 50 60 70 80 90100 Telephone and Electrical Wiring

Size of Eating Area Number of Science Labs

Number of Computer Labs Number of Library Books

Library Space Size of Classrooms

Number of Classrooms 29 24

28 16

14 18

28 40

Percentage of Schools Work Space and Equipment

Page 44 GAO- 01- 934 BIA and DOD Schools

the availability of funds. Included on this list are repair and maintenance
requirements that cost more than $25,000 for domestic schools and $10,000
for overseas schools. Projects below these levels are funded centrally out
of school or district operations and maintenance budgets, according to the
DOD?s chief of school facilities. DOD school superintendents/ principals and
facilities managers at the schools we visited indicated that they were
generally able to meet the repair and maintenance needs of their facilities
with these funds.

Table 12 shows DOD?s estimated costs to address deficiencies related to
certain building features at DOD schools. DOD school administrators often
indicated problems with these features in our survey. For example, the cost
to address all deficiencies related to heating, ventilation, and air
conditioning systems in schools totaled more than $44 million at the time of
this review. Administrators at almost 30 percent of schools responding to
our survey indicated problems with these systems. During our site visits to
seven DOD schools in Alabama and Georgia, the condition of the heating,
ventilation, and cooling systems was the most common complaint about school
facilities.

Table 12: Selected Reported Deficiencies and Estimated Repair Costs for DOD
Schools

Building features often indicated as inadequate a Estimated cost to repair
related

deficiencies b (dollars in millions)

Heating, ventilation, and air conditioning $44.5 Roofs 29 Playground
surfaces and equipment 21 a Source: GAO Survey.

b Source: 5- Year Facilities Plan for DOD schools, May 2001.

Funding to address projects in the 5- year plan comes primarily from repair
and maintenance and minor construction allocations from the operation and
maintenance budget for DOD schools, according to a DOD official. These
allocations have decreased from about $71 million in fiscal year 1998 to $40
million in fiscal year 2000. According to agency officials, these decreases
were implemented in part to help pay for escalating payroll costs at DOD
domestic schools. In addition to allocations from the operation and
maintenance budget, funding from the construction budget for DOD schools is
used for projects in the 5- year plan. The construction budget has
fluctuated over the last several years. In fiscal year 1999, the
construction budget for DOD schools totaled $44 million; in fiscal year

Page 45 GAO- 01- 934 BIA and DOD Schools

2000, this amount rose to $82 million; and in fiscal year 2001, it decreased
to $36 million.

Estimated per- pupil expenditures for DOD domestic and overseas schools were
above the national average for 1997- 98 (see table 13), and DOD schools?
estimated expenditures per pupil vary depending on their locations. 47
Schools located overseas tend to have substantially higher estimated per-
pupil expenditures ($ 10, 097) than DOD domestic schools ($ 7,725). Much of
this difference, according to DOD data, reflects the added costs of having
teachers and other school- related staff overseas. 48 These costs include
such things as benefits paid to employees in connection with moves to
overseas locations and allowances to compensate for substantially higher
living costs outside the United States.

Table 13: Comparison of Per- Pupil Expenditures at DOD and Public Schools,
1997- 98

DOD schools Estimated

per pupil expenditure Public schools Per- pupil expenditure

Domestic $7,725 All schools $6,189 a Overseas 10,097 a Data are from the
National Public Education Financial Survey, Fiscal Year 1998, as cited in
NCES?

Statistics in Brief: Revenues and Expenditures for Public Elementary and
Secondary Education: School Year 1997- 98, NCES 2000- 348 (Washington, D.
C.: U. S. Department of Education, May 2000).

We provided officials at DOD, Education, and the Department of the Interior
(Interior), which oversees BIA, an opportunity to comment on a draft of this
report. In its written comments, DOD concurred with the content of the
report and had no technical comments. DOD?s letter is printed in appendix
II. Education found the report to be helpful and informative and provided
technical comments which we incorporated into the report when data were
available. Interior provided written comments and agreed with many aspects
of our report. However, the agency pointed

47 DOD does not have readily available expenditure data to determine per-
pupil expenditures. Therefore, we used obligation data as a proxy for
expenditures. 48 Although there are some such costs associated with teachers
and school- related staff in Guam and Puerto Rico, they are much less than
those for overseas schools because there are fewer schools in these
locations than overseas. Estimated Expenditures

Per Pupil for DOD Schools Are Higher Than Public Schools? and Vary
Substantially Between Domestic and Overseas Schools

Agency Comments

Page 46 GAO- 01- 934 BIA and DOD Schools

out that BIA?s students and schools are very different from those in the
public school system and stated that in some cases GAO had made
inappropriate comparisons between BIA schools and the public school system.
Interior stressed this most with respect to measures we used to consider the
academic performance of BIA students and our calculation of expenditures per
pupil. However, we believe that we have provided a framework for our
analysis that shows a sensitivity to these issues and that our analysis is a
fair and balanced representation of the achievement of and expenditures for
BIA students. We discuss Interior?s comments in appendix III following the
agency?s printed letter. Interior also provided technical corrections which
we incorporated where appropriate.

As agreed with your offices, unless you publicly release its contents
earlier, we will make no further distribution of this report until 30 days
after its issue date. At that time, we will send copies of this report to
the Honorable Gale A. Norton, Secretary of the Interior; Neal McCaleb,
Assistant Secretary of Indian Affairs, Department of the Interior; the
Honorable Donald H. Rumsfeld, Secretary of Defense; John M. Molino, Deputy
Assistant Secretary of Defense for Military Community and Family Policy; the
Honorable Roderick R. Paige, Secretary of Education; appropriate
congressional committees; and other interested parties. Please call me at
(202) 512- 7215 if you or your staff have any questions about this report.
Key contacts and staff acknowledgments for this report are listed in
appendix IV.

Marnie S. Shaul Director, Education, Workforce,

and Income Security

Appendix I: Scope and Methodology Page 47 GAO- 01- 934 BIA and DOD Schools

This appendix discusses in more detail the scope and methodology for
examining the following aspects of Bureau of Indian Affairs (BIA) and
Department of Defense (DOD) school systems: (1) student achievement, (2)
teacher staffing, (3) access to educational technology, (4) condition of
school facilities, and (5) expenditure levels.

The scope of our review included BIA day schools and boarding schools, and
DOD schools located both in the United States and overseas. To the extent we
could, we excluded from our analyses BIA peripheral dormitories, which house
Indian students on reservations who attend nearby public schools and
generally do not have academic programs. Our focus was the 1999- 2000 school
year, and most of the data we collected relate to that year.

A major source of information for this review was a mail survey of
administrators at all BIA and DOD schools. We pretested a draft
questionnaire at eight schools and revised it based on their comments. In
November 2000, we mailed the final questionnaire to all 171 BIA schools and
224 DOD schools. We did follow- up mailings in January and March 2001 and
accepted returns through early May 2001, yielding response rates of 81
percent for BIA schools and 92 percent for DOD schools. We did not
independently verify the accuracy of the information provided in the
questionnaire responses. In addition to information collected through the
survey, we obtained budget, facilities, and other data from BIA and DOD, and
testing data from state departments of education and college testing
organizations. We obtained and reviewed supporting documentation for some
budget and expenditure data but otherwise did not verify this information.

In the course of this review, we interviewed various staff at BIA?s Office
of Indian Education Programs and DOD?s Department of Defense Education
Activity who are responsible for education programs, budgets, technology,
and facilities. As shown below, we also conducted site visits to nine BIA
schools located in South Dakota, Arizona, Minnesota, and Washington and
seven DOD schools located in the United States at three military
installations- Maxwell Air Force Base in Alabama and Fort Benning and Robins
Air Force Base in Georgia. At these schools, we toured the school facilities
and met with a variety of officials, including principals, superintendents,
education specialists, and facilities management staff. The BIA schools we
visited were

 Greasewood Springs Community School, Ariz.,

 Hopi Day School, Ariz., Appendix I: Scope and Methodology

Appendix I: Scope and Methodology Page 48 GAO- 01- 934 BIA and DOD Schools

 Muckleshoot Tribal School, Wash.,

 Chief Leschi School, Wash.,

 Flandreau Indian Boarding School, S. Dak.,

 Pine Ridge School, S. Dak.,

 Loneman Day School, S. Dak.,

 Fond du Lac Ojibwe School, Minn., and

 Nay- Ah- Shing School, Minn. The DOD schools we visited were

 Maxwell AFB Elementary School, Maxwell AFB, Ala.,

 Dexter Elementary School, Fort Benning, Ga.,

 Faith Middle School, Fort Benning, Ga.,

 Loyd Elementary School, Fort Benning, Ga.,

 White Elementary School, Fort Benning, Ga.,

 Linwood Elementary School, Robins AFB, Ga., and

 Robins Elementary School, Robins AFB, Ga. We obtained data on public
schools from readily available sources, typically from Department of
Education (Education) studies.

We conducted our work between June 2000 and July 2001 in accordance with
generally accepted government auditing standards.

To provide an indication of BIA and DOD students? academic achievement, we
reviewed their performance on standardized tests and state assessments. We
also obtained data on their performance on college entrance examinations,
namely the Scholastic Assessment Test (SAT) and the American College Test
(ACT). In addition, our survey asked administrators to report on the post-
high- school plans of graduating seniors.

For BIA schools, we attempted to collect data on BIA students? performance
on state assessments from four states that account for more than 70 percent
of all BIA schools- Arizona, New Mexico, North Dakota, and South Dakota. We
were able to obtain data for all BIA schools in North Dakota and South
Dakota and for 19 of 50 BIA schools in Arizona that participated in
Arizona?s state assessment program. No data were available for BIA schools
in New Mexico, which has about one- quarter of all BIA students. In total,
about 36 percent of BIA students attend the Student Achievement

Appendix I: Scope and Methodology Page 49 GAO- 01- 934 BIA and DOD Schools

schools to which our data relate. All data obtained were for school year
1999- 2000.

From the North Dakota Department of Public Instruction, we obtained data on
students? performance on the statewide assessment. North Dakota uses the
TerraNova Basic Multiple Assessments to assess students in grade 4 and the
TerraNova Complete Battery Plus to assess students in grades 6, 8, and 10.
The state provided data on statewide averages for students? performance on
the test, a breakdown of these scores by student ethnicity, and the scores
for all BIA schools. We focused on students? total scores on the test and
reported the range of scores across all grades. In South Dakota, schools use
the Stanford Achievement Test to assess students in grades 2, 4, 8, and 11.
From South Dakota?s Department of Education, we obtained the average scores
for students statewide, and the scores for all BIA schools on the complete
battery and for specific skill areas such as language arts, reading, and
math. 1 We reported specifically on the range of scores on the complete
battery across all grades. Arizona uses an assessment test customized for
its standards, called the Arizona Instrument to Measure Standards (AIMS), in
grades 3, 5, 8, 10, and 11. AIMS classifies students into four categories in
relation to the standards: Falls Far Below, Approaches, Meets, or Exceeds.
The Arizona Department of Education provided us with statewide averages for
students? performance on the AIMS, including a breakdown of statewide scores
by ethnicity, and the scores for participating BIA schools. We reported the
range for the proportion of children who met or exceeded the state
standards, across all grades and across all subjects (reading, writing, and
math).

We requested data on the performance of BIA students on the SAT and ACT from
the respective college testing organizations. With BIA?s approval, the
testing organizations provided us data on the scores for BIA students who
graduated in 1999 or 2000. It should be noted that a small number of BIA
graduates (fewer than 70) take the SAT. Far more BIA students (about 800)
take the ACT because BIA schools are located in areas of the country where
the ACT is the predominant college admissions test.

BIA also provided the annual reports for virtually all BIA- funded schools,
which include data on the proficiency of students as assessed by schools.

1 The complete battery score is a composite score that includes all subject
areas tested.

Appendix I: Scope and Methodology Page 50 GAO- 01- 934 BIA and DOD Schools

These assessments are based on multiple measures, which in addition to
students? performance on standardized tests can include measures such as
portfolios illustrating students? work, grades, teacher observations, and
work samples.

DOD students in both overseas and domestic schools are assessed using the
TerraNova Multiple Assessments in grades 3 through 11. We obtained data from
DOD on students? performance for school year 1999- 2000 and focused on
students? scores in language arts and math. We reported the range of scores
for students in all grades, across both language arts and math, and across
students in both overseas and domestic schools. We also obtained data from
DOD on students? performance on the National Assessment of Educational
Progress (NAEP). The NAEP is a nationally administered test offered in
different subject areas, such as reading, math, science, and writing, in
different years. We examined DOD data on students? performance on the NAEP
in math (1996), reading (1998), and writing (1998). We reviewed overall
averages for DOD students in both overseas and domestic schools and compared
these scores with national averages. Finally, we obtained SAT and ACT
results from DOD for students graduating during the 1999- 2000 school year.

Nearly all information about BIA and DOD teacher qualifications and training
was obtained through GAO?s survey. Some information on teacher turnover
rates and staff tenure was obtained from agency records. To determine class
sizes, we asked administrators on the survey to report the average number of
students in a classroom for grades K- 3, 4- 6, 7- 8, and 9- 12.

Nearly all information on BIA and DOD students? access to technology in
schools, technology staffing levels, and how technology is used in the
classroom was obtained through the survey. We calculated student- tocomputer
ratios systemwide for BIA and DOD schools by dividing the total number of
computers reported by responding schools by the total number of students at
the schools. For individual schools in the BIA and DOD systems, student- to-
computer ratios varied somewhat from the ratio for the systems overall. With
respect to technology spending, DODEA provided us with estimates of how much
the agency spent to support Teacher Staffing and

Educational Technology

Appendix I: Scope and Methodology Page 51 GAO- 01- 934 BIA and DOD Schools

educational technology programs in its schools for the 1999- 2000 school
year. BIA schools provided this information to us through the survey.

We gathered information on the condition of BIA and DOD school buildings
primarily through the survey. We asked school administrators to report on
the physical condition of buildings and on specific building features, such
as roofs and plumbing. We also asked about whether certain environmental
conditions were satisfactory, such as indoor air quality and acoustics or
noise control, and whether working facilities for teachers and students,
such as classrooms, libraries, and computer labs, were sufficient. Many of
the questionnaire items were based on those used in two previous surveys: a
1999 Education survey on the condition of school buildings, which was
conducted on a nationally representative sample of public schools; 2 and a
1994 GAO survey on the condition of school buildings, conducted on a
nationally representative sample of public schools and all BIA schools. 3 We
asked administrators to report separately on the condition of school
buildings and on the condition of dormitory buildings.

As a further indication of the condition of schools, we reviewed BIA and DOD
agency data on needed maintenance and repairs at school facilities. In
reviewing BIA?s backlog of deferred maintenance and repairs, we first
excluded data for buildings not related to schools, such as administrative
buildings, employee quarters, and buildings supporting law enforcement
services. We then looked at estimated costs to address certain types of
deficiencies according to categories used by BIA- for example, those related
to student and staff safety; environmental needs, such as asbestos and radon
testing; and energy- related needs. Finally, we asked BIA officials to
conduct searches on the backlog database for deficiencies related to certain
items asked about on our survey- for example, roofs; plumbing; fire safety;
handicapped access requirements; sewer and water infrastructure; asbestos;
heating, ventilation, and air conditioning; and playground surfaces and
equipment- and then totaled the estimated costs

2 National Center for Education Statistics (NCES), Condition of America?s
Public School Facilities: 1999, NCES 2000- 32 (Washington, D. C.: U. S.
Department of Education, June 2000).

3 School Facilities: Condition of America?s Schools (GAO/ HEHS- 95- 61, Feb.
1995). Facilities

Condition Backlogs of Maintenance and Repair Needs

Appendix I: Scope and Methodology Page 52 GAO- 01- 934 BIA and DOD Schools

for these deficiencies. We also obtained the 5- Year Facilities Plan for DOD
schools, which includes a total for the estimated amount needed to address
all identified items. It should be noted that the 5- year plan does not
include repair projects costing under $25,000 for domestic schools and
$10,000 for overseas schools. We reviewed the 5- year plan data for specific
building deficiencies that were noted more often as problems by DOD schools
in our survey: roofs; heating, ventilation, and air conditioning; and
playground surfaces and equipment.

To assess maintenance and repair funding levels for BIA schools, we measured
funding against guidelines set by the National Research Council (NRC). The
NRC has recommended that budget allocations for maintenance and repair be
between 2 and 4 percent of current replacement value. 4 We first determined
the current replacement value of BIA schools by multiplying the square
footage of the schools by the cost to construct a new school per square
foot. We obtained BIA?s total education facility square footage for
maintained buildings for 1997- 2001. (These data included 14 peripheral
dormitories and two colleges, which account for about 11 percent of BIA?s
total education facility square footage.) We then obtained data on the
average cost per square foot of a new school for each year from 1997 through
2000, from annual education construction reports published by American
School and Universities. 5 (The average cost per square foot of a new school
was not yet available for 2001, so we used the average amount per square
foot reported for 2000 for this year.) Then, for each year from 1997 to
2001, we multiplied BIA?s educational facility square footage by the average
cost per square foot to construct a school that year.

To identify BIA budget allocation categories that met the NRC?s definition
of maintenance and repair, we reviewed BIA documents describing the
categories and discussed these with experts. Once we identified the
appropriate categories, we obtained the amounts BIA requested and allocated
for these categories for fiscal years 1997- 2001. Then for each year, we
determined the percentage that the requested and allocated

4 NRC, Committing to the Cost of Ownership: Maintenance and Repair of Public
Buildings (1990). 5 J. Agron, 24th, 25th, 26th and 27th Annual Official
Education Construction Reports,

(Overland Park, Kan.: American School and University, May 1998, May 1999,
May 2000, and May 2001). BIA?s Allocations for

Maintenance and Repairs of Schools as a Percentage of Replacement Value

Appendix I: Scope and Methodology Page 53 GAO- 01- 934 BIA and DOD Schools

amounts represented of the current replacement value and compared these with
NRC guidelines.

To calculate estimated per pupil expenditures for DOD and BIA schools that
would be comparable to public school figures, we followed guidelines for
identifying expenditures set forth by the National Center for Education
Statistics (NCES) in its National Public Financial Education Survey. 6 The
results of this survey are used to develop the national per- pupil figure.
National per- pupil expenditure figures reflect current expenditures from
federal, state, local and private sources, at the district level for
prekindergarten through grade 12. Current expenditures include salaries,
benefits, purchased services, and supplies for the following functions:
instruction; support services; noninstructional services, such as food
services; and direct program support, such as state contributions to
employee retirement funds. Long- term expenditures such as capital outlay,
debt service, facilities acquisition and construction services, and property
expenditures (for example, for equipment and vehicles) are excluded from
current expenditures by NCES and, in most cases, by our evaluation. At the
time we did our work, the latest school year for which national public
school per- pupil expenditures were available was 1997- 98.

Data needed to precisely compute per- pupil expenditures for both BIA and
DOD schools were not available. Therefore, our calculations are estimates
based on BIA and DOD agency data and federal expenditure information from
tribal audit reports submitted under the Single Audit Act. Because BIA and
DOD data were generally not available by categories specified by NCES, we
used agency or tribal expenditure or obligation data for programs or budget
line items which included these categories. Using this approach we were able
to include expenditures for salaries, benefits, purchased services, and
supplies for the categories included in the national per- pupil expenditure:
instruction, support services, and noninstructional services. Direct program
support generally did not apply. Similarly, we were generally able to
exclude expenditures for capital outlay, debt service, and facilities
acquisition and construction services by excluding budget line items that
contain these expenditures. In some

6 NCES, The National Public Education Financial Survey Instruction Booklet,
January 1999 (Washington, D. C.: U. S. Department of Education). Because
1997- 98 was the latest year for which national public school per- pupil
expenditures were available at the time we did our work, we used the
guidelines dated January 1999, since these guidelines applied to the 12-
month period beginning July 1, 1997, and ending June 30, 1998. Estimated Per
Pupil

Expenditures

Appendix I: Scope and Methodology Page 54 GAO- 01- 934 BIA and DOD Schools

cases, we were not able to extract specific expenditures or obligations
relating to property expenditures within selected programs; however, these
amounts were likely to be relatively small.

For BIA schools, we developed estimated per- pupil calculations for (1) BIA-
operated day schools, (2) tribally operated day schools, (3) day schools
overall, and (4) boarding schools. For BIA- operated schools, we obtained
expenditure information from the Department of the Interior?s (Interior)
financial system based on codes indicating programs that support
kindergarten through grade 12. We were not able to identify expenditures for
administrative services provided by Interior for BIAoperated schools, such
as payroll and procurement. We used fiscal year 1998 data. A fiscal year
reasonably approximates the period July 1, 1997, and ending June 30, 1998,
used by NCES. For tribally operated schools, we relied primarily on federal
expenditure information from audit reports submitted under the Single Audit
Act. These reports were generally for the period July 1, 1997, through June
30, 1998. We were able to obtain expenditure information for 54 of the 98
schools that were tribally operated that year. We could not obtain
expenditure information for the remaining 44 schools because, in some cases,
the school expenditure information in the audit report was included with
other expenditure information for tribal or other organizations and could
not be readily extracted, or because related audit reports were not
available.

For both BIA- operated and tribally operated schools, we considered
expenditures from federal funding sources only, which, according to BIA
officials, represent by far the greatest source of funding for BIA schools.
This included expenditure data for Education programs, such as Title I; and
the Department of Agriculture?s (USDA) child nutrition programs, such as the
National School Lunch and Breakfast programs. The per- pupil expenditure
figures for public schools nationally consider funding from all sources.
Most funding for public schools comes from state, intermediate, and local
sources.

In order to determine overall estimated per- pupil expenditures for day and
boarding schools, we weighted the calculated estimated per- pupil
expenditures for BIA- operated and tribally operated schools of both types,
according to the number of students attending each type of school. We used
student enrollment figures from BIA internal reports for each school type.

In order to calculate DOD overseas and domestic estimated per- pupil
expenditures, we used obligations from DOD internal financial reports for

Appendix I: Scope and Methodology Page 55 GAO- 01- 934 BIA and DOD Schools

fiscal year 1998 as a proxy for expenditures, since DOD does not maintain
expenditure information in the detail necessary for this type of analysis.
We also used fiscal year 1998 data since, similar to other federal agencies,
DOD tracks obligations by fiscal year and not for other periods. For student
enrollment figures, we relied on data from DOD internal reports for domestic
and overseas schools. We also obtained information for the value of USDA?s
child nutrition programs, such as the National School Lunch and Breakfast
programs, for fiscal year 1998, which we included in the estimated
expenditure per pupil calculation.

Appendix II: Comments from the Department of Defense

Page 56 GAO- 01- 934 BIA and DOD Schools

Appendix II: Comments from the Department of Defense

Appendix III: Comments from the Department of the Interior

Page 57 GAO- 01- 934 BIA and DOD Schools

Appendix III: Comments from the Department of the Interior

Appendix III: Comments from the Department of the Interior

Page 58 GAO- 01- 934 BIA and DOD Schools

See comment 3. See comment 2.

See comment 1.

Appendix III: Comments from the Department of the Interior

Page 59 GAO- 01- 934 BIA and DOD Schools

See comment 6. See comment 5.

Now on p. 12. See comment 4.

Now on p. 10.

Appendix III: Comments from the Department of the Interior

Page 60 GAO- 01- 934 BIA and DOD Schools

See comments 8 & 9. Now on p. 14. See comment 7.

Appendix III: Comments from the Department of the Interior

Page 61 GAO- 01- 934 BIA and DOD Schools

Now on p. 19 See comment 10.

Now on pp. 17- 18.

Appendix III: Comments from the Department of the Interior

Page 62 GAO- 01- 934 BIA and DOD Schools

See comment 12. See comment 11.

Appendix III: Comments from the Department of the Interior

Page 63 GAO- 01- 934 BIA and DOD Schools

Now on p. 28. Now on p. 31. See comment 13.

Appendix III: Comments from the Department of the Interior

Page 64 GAO- 01- 934 BIA and DOD Schools

Now on p. 32. See comment 16.

See comment 15. See comment 14.

Appendix III: Comments from the Department of the Interior

Page 65 GAO- 01- 934 BIA and DOD Schools

See comment 17. See comment 18. See comment 17.

Appendix III: Comments from the Department of the Interior

Page 66 GAO- 01- 934 BIA and DOD Schools

See comment 19. Now on p. 33.

Appendix III: Comments from the Department of the Interior

Page 67 GAO- 01- 934 BIA and DOD Schools

Appendix III: Comments from the Department of the Interior

Page 68 GAO- 01- 934 BIA and DOD Schools

1. We stated that BIA?s funding for school operations and facilities needs
increased 30 percent from fiscal years 2000 to 2001. Interior suggested that
GAO differentiate the actual amount of the increase between the Operation of
Indian Programs and Construction appropriation accounts, initially stating
that each account had increased by 25 percent. Based on our calculations, we
added information to the report specifying that funding for Operation of
Indian Programs increased by about 5 percent and funding for Construction
increased 121 percent between fiscal years 2000 and 2001. This supports our
statement that nearly all the increase was for the repair and replacement of
school facilities while funding for school operations received only a modest
increase.

2. Interior commented that our report should acknowledge that the Department
of Education (Education) funding that BIA schools receive is available to
all public schools across the country. Interior specifically noted that
public schools that educate Indian children on or near Indian reservations
receive Impact Aid Funds from Education. We added text to the report
indicating that like BIA schools, many public schools also receive funding
from Education.

3. Interior noted that in an agreement with Education, BIA serves as a state
education agency and is authorized to receive flow through funds from
Education. We have added this information in the report.

4. Interior suggested that we provide some examples to further highlight
differences between BIA and DOD schools. Interior stated that the DOD school
system has no schools with small enrollments or combinations of grades like
those in the BIA school system. The DOD school system does have several
schools with enrollments under 100 students and schools that span both
elementary and secondary grades, but we agree that these types of schools
are less prevalent in DOD?s school system. Interior also said that it would
have been helpful for GAO to provide an example of how BIA schools are less
able to take advantage of economies of scale and that this would help
explain the school system?s per- pupil expenditure. In our report section on
per pupil expenditures, we describe a situation that illustrates how
difficult it can be for BIA schools to achieve economies of scale. Finally,
Interior stated that the report could comment on how the physical conditions
found in BIA schools could affect student outcomes. Thus, we added a note to
the report stating that though some studies conclude that school facility
conditions- such as GAO Comments

Appendix III: Comments from the Department of the Interior

Page 69 GAO- 01- 934 BIA and DOD Schools

inadequate ventilation or faulty heating systems- can affect student
learning, the research is inconclusive overall.

5. Regarding the performance of BIA students on state assessments, Interior
argued that state standardized tests may be culturally biased and may be a
less reliable measure for Indian students than other minorities. In the
report we acknowledged that standardized tests have been criticized as being
culturally biased, and we therefore provided information on authentic
assessments of BIA students. However, we are not aware of any research that
shows standardized tests are less reliable for Indian students than other
minority students.

6. Interior said GAO should have compared BIA student achievement with
public schools serving large numbers of Indian students in rural areas. We
agree that these would be worthwhile analyses. However, the additional audit
work and data required to perform these analyses did not allow us to include
them in this report.

7. Interior noted BIA?s efforts to address differences in student
achievement through programs designed to moderate the effects of poor
economic and family conditions on student achievement. The agency also noted
that BIA had established long term goals that address student performance.
We have added information to the report to acknowledge BIA?s efforts in
these areas.

8. Interior questioned our comparison of BIA students? plans to attend
college and performance on college entrance examinations with those of
students nationally. The agency argued that the economies of rural areas,
where most BIA schools are located, do not offer jobs that require post-
secondary education, and for this and other reasons, less cultural emphasis
is placed on BIA students to perform well on tests like the SAT and to
attend college. The agency stated that BIA students generally expect to go
to tribal colleges that do not require college entrance examinations, such
as the SAT. We believe that these measures are appropriate given that one of
the Office of Indian Education Program?s goals is to increase post-
secondary enrollment rates. Also, a significant number of BIA students take
the SAT and ACT, perhaps to enter colleges that require these examinations.

9. Interior also said that we should have compared BIA students? plans to
attend college and their performance on college entrance examinations with
students from other rural regions of the U. S. or students in public schools
on Indian reservations. Some data are

Appendix III: Comments from the Department of the Interior

Page 70 GAO- 01- 934 BIA and DOD Schools

available for students living in rural areas but these are not specifically
focused on Indian students. For example, data from 1993- 94 show that on
average 54 percent of 12th graders living in small towns or rural areas
applied to college, compared with BIA school administrators? estimates that
28 percent of graduating seniors planned to enroll in college. With regard
to the SAT, for students graduating in 2000, rural students received an
average total score of 992 on the verbal and math sections of the SAT,
compared with 1,019 for students nationally, and 765 for BIA students.

10. Interior stated that teacher turnover correlates with reduced student
achievement, but did not provide support for this statement. We were not
able to identify any studies establishing a link between teacher turnover
and student achievement.

11. Interior noted that in fiscal year 2000, BIA conducted a census of
personal computers in BIA schools and concluded that the school system had a
ratio of 1 computer for every 3.3 students. BIA added, however, that 30
percent of these computers are for administrative usage and that almost half
are reaching or have surpassed their life cycle usefulness. Our survey asked
school administrators about computers used specifically for instructional
purposes. Using these data, we determined that the school system had a ratio
of 1 computer for instructional purposes for every 3.5 students.

12. Interior said that we had failed to address the fact that state
legislatures have appropriated funds for school technology and that these
funds are not included in district funding appropriations and therefore are
not included in calculations of expenditures per pupil by NCES. We contacted
NCES and were told that state funding for educational technology should be
included in the data that states report to NCES for expenditure per pupil
calculations. Thus, based on this information, we did not modify our report.
Interior also noted that the Office of Indian Education Programs had to rely
on private partnerships and grants to provide enough resources to place
technical capabilities in all its schools. (It should be noted that these
additional resources were not identified during our review and therefore
were not included in our expenditure per pupil calculation for BIA schools.)
In addition, Interior stated that our report was unclear about the resources
DOD uses to support technology in its schools. In our report we provided an
estimate for the amount of money the DOD school systems obligated per
student for educational technology in fiscal year 2000, which was less than
the amount BIA schools spent per student

Appendix III: Comments from the Department of the Interior

Page 71 GAO- 01- 934 BIA and DOD Schools

based on our survey results. This funding comes from DOD appropriations.

13. With respect to housing for teachers and staff, Interior said that we
had assumed that rents collected from teachers living in employee housing
covered the cost of maintaining this housing and had not realized that
schools end up covering these costs out of their facilities maintenance
funds. Interior argued that school maintenance funding diverted for teacher
housing should not have been included in our calculation of expenditures per
pupil. Interior also stated that staff in DOD schools live in government
provided quarters expense free. Early in our review, BIA officials informed
us that rents paid by teachers and staff did not cover the full cost of
maintaining employee housing. However, when requested, BIA was unable to
identify expenditures made by schools for such purposes. More importantly,
Interior noted in its comments that providing housing is necessary for
maintaining a qualified teaching staff. Because of this, we believe that
expenditures to maintain employee housing are part of the cost of providing
instruction for these schools. An NCES official agreed and stated that
expenditures from school facilities maintenance funds used to maintain
employee housing should be included in the calculation of expenditures per
pupil. For the same reason, we included DOD schools? expenditures for
housing allowances provided to overseas staff in our calculation of
expenditures per pupil. We have also added a footnote stating that a small
number of staff in DOD schools overseas live in government- provided
quarters; most receive housing allowances intended to cover the cost of
housing in the areas where they choose to live.

14. Interior commented that our report failed to address the adequacy of
funding for facilities operations. We were not able to identify a widely
used benchmark to assess the sufficiency of BIA?s facilities operations
funding. We were only able to compare BIA?s obligations for both facilities
operations and maintenance with similar expenditures for public schools. We
found that BIA?s obligations per square foot of building space were higher
than the national median for public schools. However, it was difficult to
draw conclusions from these calculations as they do not consider the
additional infrastructure BIA schools operate and maintain. Also, this
analysis does not examine the adequacy of BIA?s funding for operations and
maintenance, it merely compares the amounts used by BIA and public schools
for these purposes. Further, we were aware at the time we conducted our work
that BIA had contracted with a private engineering firm to assess the

Appendix III: Comments from the Department of the Interior

Page 72 GAO- 01- 934 BIA and DOD Schools

adequacy of BIA?s facilities operations and maintenance funding. This work
may provide a clearer picture of BIA?s needs. This report is expected to be
completed in Fall 2001.

15. Interior commented that the cost of operating facilities could be
reduced if school replacement construction funding covered the cost of
demolishing and removing old structures. Currently, BIA provides for the
security of these idle structures out of operations funding until a decision
is made on their disposal. This issue may be worthy of attention, but BIA?s
own budget formula for determining operations and maintenance needs suggests
that this is not a major expense.

16. Interior noted that BIA, DOD, and public schools define and account for
maintenance funds and expenditures differently, and these differences make
?a straight across the board comparison? difficult. We determined that it
would be challenging to complete such an analysis within the timeframe of
this review. Thus, we did not attempt to compare maintenance budgets or
expenditures among these school systems. Rather, we analyzed BIA?s and DOD?s
maintenance and repair backlogs and provided information on the amount of
funding allocated to address them and school construction.

17. Interior commented that our estimates of expenditures per pupil were not
adjusted to reflect differences between students in day schools and in
residential schools. We did calculate separate per- pupil expenditures for
day schools and residential schools and did not combine them. Interior also
stated that we did not adjust our per- pupil expenditure estimates for
students with special education needs or for the costs of small schools in
sparsely populated areas. The agency recommended that we use an alternative
approach for calculating expenditures per pupil that takes into account
differences in student and school needs. Interior pointed to funding
allocation systems used by some states where students are weighted
differently based on the additional education resources they may require.
For example, a student requiring special education services may be counted
as 1.5 students in determining state aid. Interior argued that we should
have used a weighted count of its students to calculate expenditures per
pupil, since a high proportion have special needs. Such a calculation may be
useful, but comparisons with public schools would not be possible since NCES
does not calculate the national expenditure per pupil using a weighted
student count.

Appendix III: Comments from the Department of the Interior

Page 73 GAO- 01- 934 BIA and DOD Schools

18. Interior stated that we should have compared BIA expenditures per pupil
with that of school districts with similar student populations. Interior
suggested several school districts with similar student populations in South
Dakota and New Mexico that could be used for comparison with BIA?s per-
pupil expenditure. We obtained the perpupil expenditures for these districts
from NCES and found they ranged from $4,979 to $8, 706 in 1997- 98, all
below BIA?s estimated expenditure per pupil of $9,647.

19. Interior recommended that we should not have included administrative
cost grants for tribally operated schools in our expenditure per pupil
calculation. The agency argued that these grants are for the purpose of
administering the schools and carrying out support functions that are not
part of the basic academic instruction, such as fiscal and personnel
activities. According to an NCES official, these expenditures should be
included as they fall into the category of general administration
expenditures, which is included in the national per pupil expenditure
calculation.

Appendix IV: GAO Contacts and Staff Acknowledgments

Page 74 GAO- 01- 934 BIA and DOD Schools

Karen A. Whiten, (202) 512- 7291 Linda Y. A. McIver, (206) 287- 4821

In addition to the individuals named above, Mary E. Abdella, Beverly G.
Burke, Julianne Hartman Cutts, Eric M. Eliasen, Kopp F. Michelotti, Nancy R.
Purvine, Joel I. Grossman, Elsie M. Picyk, Stanley G. Stenersen, and James
P. Wright made key contributions to this report. Appendix IV: GAO Contacts
and Staff

Acknowledgments GAO Contacts Staff Acknowledgments

(102002)

The first copy of each GAO report is free. Additional copies of reports are
$2 each. A check or money order should be made out to the Superintendent of
Documents. VISA and MasterCard credit cards are also accepted.

Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.

Orders by mail:

U. S. General Accounting Office P. O. Box 37050 Washington, DC 20013

Orders by visiting:

Room 1100 700 4 th St., NW (corner of 4 th and G Sts. NW) Washington, DC
20013

Orders by phone:

(202) 512- 6000 fax: (202) 512- 6061 TDD (202) 512- 2537

Each day, GAO issues a list of newly available reports and testimony. To
receive facsimile copies of the daily list or any list from the past 30
days, please call (202) 512- 6000 using a touchtone phone. A recorded menu
will provide information on how to obtain these lists.

Orders by Internet

For information on how to access GAO reports on the Internet, send an email
message with ?info? in the body to:

Info@ www. gao. gov or visit GAO?s World Wide Web home page at: http:// www.
gao. gov

Contact one:

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

 E- mail: fraudnet@ gao. gov

 1- 800- 424- 5454 (automated answering system) Ordering Information

To Report Fraud, Waste, and Abuse in Federal Programs
*** End of document. ***