Head Start Programs: Participant Characteristics, Services, and Funding
(Letter Report, 03/31/98, GAO/HEHS-98-65).

Pursuant to a congressional request, GAO provided information on the
Head Start Program in preparation for its reauthorization, focusing on:
(1) the number and characteristics of those served; (2) the services
provided and the way they are provided; (3) federal and nonfederal
program dollars received and spent by programs delivering Head Start
services; and (4) other programs providing similar early childhood
services.

GAO noted that: (1) Head Start served about 782,000 disadvantaged
children and 711,000 families in program year 1996-97, according to
GAO's review; (2) the demographics of these children and families were
similar in many respects; (3) most children were 4 years old and spoke
English as their main language; (4) families typically had more than one
child and were very poor; (5) through Head Start, children received
access to a large array of services, as did their families in some
cases; (6) most child and family services, however, were neither paid
for nor provided directly by Head Start programs; (7) instead, Head
Start programs often functioned as a coordinator or facilitator,
referring and linking children and families to needed services; (8)
although many families required full-day, full-year child care, Head
Start services were typically provided in centers that operated part day
on schedules that paralleled the school year; (9) only a small
percentage of children attended programs in centers that operated year
round; (10) virtually no programs operated on weekends, and only a few
operated before 7 a.m. or after 5 p.m.; (11) almost half of the families
identified as needing full-day services left their children at a
relative's or unrelated adult's home when the children were not in Head
Start; (12) most programs responding to GAO's survey secured funding for
their operations from multiple sources; (13) among all programs in the
states and territories, the average amount of Head Start grant funds per
child was $4,637, ranging from a low of $792 to high of $16,206; (14)
the additional income programs received from other sources increased the
amount of funds available per child to an average of $5,186, 12 percent
more income per child; (15) total funds per child varied widely by
program, ranging from $1,081 to $17,029 per child; (16) programs spent
their income on a variety of services and activities; however, the
largest promotion of programs' overall income was spent on education
services; (17) most Head Start programs reported that state-funded
preschools, other preschools, child development centers and child care
centers, and family day care homes operated in the same communities as
Head Start programs; and (18) although GAO's review did not determine
the extent to which these programs resemble Head Start, some that serve
disadvantaged children sometimes help children and families obtain
additional services, such as medical services, as Head Start does.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-98-65
     TITLE:  Head Start Programs: Participant Characteristics, Services, 
             and Funding
      DATE:  03/31/98
   SUBJECT:  Locally administered programs
             Preschool education
             Food programs for children
             Disadvantaged persons
             Preschoolers
             Child care programs
             Labor costs
             Block grants
             Authorization
             Surveys
IDENTIFIER:  Head Start Program
             Child Care and Development Block Grant
             Social Services Block Grant
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Requesters

March 1998

HEAD START PROGRAMS - PARTICIPANT
CHARACTERISTICS, SERVICES, AND
FUNDING

GAO/HEHS-98-65

Head Start Program Participants

(104869)


Abbreviations
=============================================================== ABBREV

  ACF - Administration for Children and Families
  AFDC - Aid to Families With Dependent Children
  HHS - Department of Health and Human Services
  PIR - Program Information Report
  TANF - Temporary Assistance for Needy Families
  USDA - U.S.  Department of Agriculture

Letter
=============================================================== LETTER


B-276121

March 31, 1998

Congressional Requesters

Head Start--now over 30 years old--has long enjoyed both
congressional and public support.  Since its inception, Head Start
has served over 16 million children at a total cost of $35 billion. 
Funding for Head Start has tripled in the past 10 years.  In
addition, the President recently announced several proposals to help
working parents secure affordable, quality child care, including
significantly expanding Head Start so it could serve more eligible
children. 

Growing out of the War on Poverty in the mid-1960s, Head Start was
created to provide comprehensive educational, health, social, and
mental health services to disadvantaged preschool children.  The
program was built on the philosophy that effective intervention in
children's lives can be best accomplished through family and
community involvement. 

Much has changed since the mid-1960s, and many questions exist about
how the program operates in today's environment.  For example,
although it is widely known that Head Start provides a large array of
services to children, less is known about the services it provides
families and how services are delivered to participants.  Head Start
regulations emphasize that programs secure and use community
resources to provide services before using Head Start funds; however,
the extent to which programs secure other funding is not known. 
Furthermore, upon Head Start's creation, although it was many
communities' main early childhood program for serving disadvantaged
children and their families, other programs now also serve
disadvantaged children.  Finally, questions exist about how
predominantly part-day Head Start programs will adapt to meet the
changing needs of the families they serve, particularly in view of
the increased need for full-day child care resulting from changes in
welfare policy.\1

Although Head Start does collect information on all its programs
annually, that information does not completely portray Head Start
program operations.  Head Start collects information on the type and
extent of services provided and the types of children and families
being served through an annual survey.  The program collects no
information, however, on the number of hours and months of the year,
for example, that children attend center programs.  Nor does it
collect information on Head Start program income and expenditures or
other early childhood programs operating in communities in which Head
Start programs operate.  Moreover, although Head Start does collect
information on the number of families the program serves, it collects
no information on the number of individual family members served. 
Consequently, as the Congress prepares to reexamine the Head Start
program in 1998, information needed for answering many questions is
unavailable, incomplete, or lacking enough detail to facilitate
important decisions about the program. 

To address some of these questions in preparation for Head Start's
reauthorization, we were asked to describe the (1) number and
characteristics of those served, (2) services provided and the way
they are provided, (3) federal and nonfederal program dollars
received and spent by programs delivering Head Start services, and
(4) other programs providing similar--in part or in whole--early
childhood services.  Although many other early childhood programs
exist, our review focused solely on programs operating in Head Start
service delivery areas and those operated by Head Start agencies. 

The methodology for our study had several components:  we (1)
surveyed all Head Start programs, (2) analyzed data from Head Start's
annual survey of programs, (3) conducted telephone interviews of a
number of programs to gather illustrative information about Head
Start programs, and (4) visited several Head Start programs to
observe the programs and discuss survey responses.  (See app.  I for
a discussion of our objectives, scope, and methodology.) This report
presents information on what we call Head Start's regular\2 program,
that is, those programs that operate in the 50 states and the
District of Columbia and serve 85 percent of the children.  Thus,
programs serving special populations--Migrant, Native American, and
pregnant women and infants--are excluded.  Such programs represent a
small percentage of Head Start children served, and each program is
unique.  Because of this focus, and other reasons discussed in the
report, certain figures, such as enrollment data, may differ from
Head Start's official 1996-97 figures. 

The scope of our work for this report excluded several issues.  For
example, we did not address issues of Head Start program quality,
although some early childhood experts and the Advisory Committee on
Head Start Quality and Expansion have voiced concern about the uneven
quality of some Head Start programs.\3 Nor did we address Head
Start's impact on the lives of those it serves.  In a previous
report, we examined the research conducted on the program in the past
20 years and found it inadequate to draw conclusions about the impact
of the national program.\4 As agreed with your offices, we will be
issuing a separate report on how Head Start uses its processes,
systems, and performance measures to ensure program and fiscal
accountability. 


--------------------
\1 Under welfare reform legislation passed in 1996, many more welfare
families will be expected to seek and keep jobs.  The Personal
Responsibility and Work Opportunity Reconciliation Act of 1996 made
sweeping changes to national welfare policy by ending the Aid to
Families With Dependent Children (AFDC) program.  Replacing AFDC are
the Temporary Assistance for Needy Families (TANF) block grants,
which provide federal funds to help states help needy families. 

\2 Head Start programs in Guam, American Samoa, the Virgin Islands,
the Federated States of Micronesia, the Republic of the Marshall
Islands, the Commonwealth of the Northern Marianna Islands, Puerto
Rico, and Palau are also funded under the regular program. 

\3 The Secretary of the Department of Health and Human Services (HHS)
initiated a bipartisan task force, the Advisory Committee on Head
Start Quality and Expansion, to review the Head Start program and
make recommendations.  The Committee issued a report in Dec.  1993
titled Creating a 21st Century Head Start.  In its technical comments
on the report, the agency noted that it has taken steps since that
time to improve the quality of all Head Start programs nationwide,
including terminating more than 80 programs that were not meeting
required levels of quality.  The agency provided no additional
information, however, about the overall quality of current Head Start
programs or their impact on the children and families served. 

\4 Head Start:  Research Provides Little Information on Impact of
Current Programs (GAO/HEHS-97-59, Apr.  15, 1997). 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Head Start served about 782,000 disadvantaged children and 711,000
families in program year 1996-97, according to our review.  The
demographics of these children and families were similar in many
respects.  Most children were 4 years old and spoke English as their
main language.  Moreover, families typically had more than one child
and were very poor. 

Through Head Start, children received access to a large array of
services as did their families in some cases.  For example, in
addition to education services, children received medical and dental
care, immunizations, social services, child care, meals, and other
nutrition services.  Families received access to social services, and
parents received access to job and literacy training.  Most child and
family services, however, were neither paid for nor provided directly
by Head Start programs.  Instead, Head Start programs often
functioned as a coordinator or facilitator, referring and linking
children and families to needed services. 

Although many families required full-day, full-year child care, Head
Start services were typically provided in centers that operated part
day--usually 3 to 4 hours a day--on schedules that paralleled the
school year--about 9 months a year.  Only a small percentage of
children attended programs in centers that operated year round. 
Virtually no programs operated on weekends, and only a few operated
before 7 a.m.  or after 5 p.m.  Almost half of the families
identified as needing full-day services left their children at a
relative's or unrelated adult's home when the children were not in
Head Start. 

Most programs responding to our survey secured funding for their
operations from multiple sources, such as states and other federal
programs.  Among all programs in the states and territories, the
average amount of Head Start grant funds per child was $4,637; it
ranged from a low of $792\5 to a high of $16,206.  The additional
income programs received from other sources increased the amount of
funds available per child to an average of $5,186--a difference of
about $549 or 12 percent more income per child.  Total funds per
child varied widely by program, ranging from $1,081 to $17,029 per
child. 

Programs spent their income on a variety of services and activities. 
However, the largest portion of programs' overall income was spent on
education services.  Personnel-related expenses were the largest
expense, and personnel delivering education services--services that
the program typically provides directly--accounted for the largest
portion of expenses. 

Most Head Start programs reported that state-funded preschools, other
preschools, child development centers and child care centers, and
family day care homes operated in the same communities as Head Start
programs.  Although our review did not determine the extent to which
these programs resemble Head Start, some that serve disadvantaged
children sometimes help children and families obtain additional
services, such as medical services, as Head Start does. 


--------------------
\5 This program operated in American Samoa.  Within the United
States, the lowest reported Head Start funding per child was $1,081. 


   BACKGROUND
------------------------------------------------------------ Letter :2

Head Start, the centerpiece of federal early childhood programs, was
created in 1965 as part of President Johnson's War on Poverty.  Head
Start's primary goal is to improve the social competence of children
in low-income families.  Social competence is the child's everyday
effectiveness in dealing with both the present environment and later
responsibilities in school and life.  Social competence involves the
interrelatedness of cognitive and intellectual development, physical
and mental health, nutritional needs, and other factors.  To support
the social competence goal, Head Start programs deliver a broad range
of services to children.  These services include educational,
medical, nutritional, mental health, dental, and social services. 
Another essential part of every program is parental involvement in
parent education, program planning, and operating activities. 

Head Start programs are governed by performance standards, which set
forth the expectations and minimum requirements that all Head Start
programs are expected to meet.  Program officials expect these
standards, however, to be largely self-enforcing, with the exception
that Head Start's 12 regional offices conduct on-site monitoring of
Head Start programs every 3 years.\6 The program also has a separate
set of performance standards for services for children with
disabilities.  Both sets of performance standards, which have
governed the program since 1975, were revised in the 1990s.  Head
Start issued performance standards for children with disabilities in
1993.  The performance standards for the rest of the programs became
effective in January 1998 and attempt to reflect the changing Head
Start population, the evolution of best practices, and program
experience with the earlier standards.\7

Head Start targets children from poor families, and regulations
require that at least 90 percent of the children enrolled in each
program be low income.  By law, certain amounts are set aside for
special populations of children, including those with disabilities
and Native American and migrant children.  The program is authorized
to serve children at any age before the age of compulsory school
attendance; however, most children enter the program at age 4. 


--------------------
\6 An annual fiscal audit is also required. 

\7 Head Start programs may voluntarily implement these standards
before the effective date.  At the time of our study, however, the
1975 performance standards were in effect. 


      HEAD START HAS THREE
      APPROVED PROGRAM OPTIONS
---------------------------------------------------------- Letter :2.1

Head Start programs may be delivered in any of three Head
Start-approved program options.  One option involves the enrolled
child receiving the bulk of Head Start services at a center; however,
some home visits are required.  Centers operate varying numbers of
hours per day for either 4 or 5 days per week.  Providing services at
children's homes is a second option.  The children receive the bulk
of services at home, with some opportunities for them to interact in
a group setting.  The combination option--the third--
entails both center attendance and home visits.  In addition,
programs may implement a locally designed option, which, as the name
implies, is developed at the local program level.  Locally designed
options may take many forms, such as family day care homes. 

How are services delivered in a center setting, the most common
option?  The center may be housed in a church basement, at a parent's
work site, in a public school building, at a college or university,
or some other location.  A Head Start teacher as well as a second
adult instruct the children using a curriculum relevant to and
reflective of the needs of the population served.  Head Start
regulations emphasize that large and small group activities take
place throughout the day.  Children should be encouraged to solve
problems, initiate activities, explore, experiment, question, and
gain mastery through learning by doing.  In addition to educational
services, children receive other services.  Meals and snacks are
provided as appropriate.  Within a certain number of days of entering
the program, children receive a thorough health screening and medical
and dental examination.  This screening may take place on or off
site.  Program staff ensure that treatment and follow-up services are
arranged for all health problems detected.  In addition, Head Start
staff are expected to visit the children's homes to assess their and
their families' need for services.  For example, these visits may
identify the families' need for services such as emergency assistance
or crisis intervention.  Staff may also provide families with
information about community services and how to use them.  During
these visits, staff are expected to develop activities for family
members to use at home that will reinforce and support the child's
total Head Start experience. 


      HEAD START ADMINISTRATION
      INVOLVES GRANTEE AGENCIES
---------------------------------------------------------- Letter :2.2

Head Start is administered by HHS' Administration for Children and
Families (ACF), which includes the Head Start Bureau--one of several
under ACF.  Grantees, which deliver Head Start services at the local
level, numbered about 1,440 in fiscal year 1996.  Grantees may
contract with organizations--called delegate agencies--in the
community to run all or part of their local Head Start programs. 
Grantees and delegate agencies include public and private school
systems, community action agencies and other private nonprofit
organizations, local government agencies (primarily cities and
counties), and Indian tribes.  Unlike some other federal social
service programs funded through the states, HHS awards Head Start
grants directly to local grantees.  HHS distributes Head Start funds
using a complex formula, based upon, among other things, previous
allotments and the number of children, aged 5 and under, below the
poverty line in each state compared with the number in other states. 
Head Start, a federal matching grant program, requires grantees to
typically obtain 20 percent of program costs from nonfederal funds. 
These funds can be in the form of cash, such as state, county, and
private money, or in-kind contributions such as building space and
equipment.  Head Start regulations require that programs identify,
secure, and use community resources in providing services to Head
Start children and their families before using Head Start funds for
these services.  As a result, Head Start programs have established
many agreements for services. 

Head Start has served over 16 million children since its inception. 
The passage of the 1990 Head Start Expansion and Quality Improvement
Act resulted in increased funding for Head Start to allow more
children the opportunity to participate in Head Start as well as
improve the quality of Head Start services.  In fiscal year 1996,
Head Start received $3.6 billion\8 in funding and served about
752,000 children.  This figure reflects children served through all
of Head Start's programs.  The regular Head Start program serves
children and families residing in the 50 states and the District of
Columbia.\9 About 85 percent of Head Start children are served
through the regular Head Start program.  Head Start also operates
programs for migrant and Native American populations. 

Recognizing that the years from conception to age 3 are critical to
human development, the Congress established Early Head Start in 1994. 
This program targets children under age 3 from low-income families as
well as expectant mothers.  Since 1967, however, Head Start has
served children and families now targeted by the Early Head Start
program through Parent Child Centers. 


--------------------
\8 Head Start's fiscal year 1997 appropriation is about $4 billion. 

\9 Head Start programs in Guam, American Samoa, the Virgin Islands,
the Federated States of Micronesia, the Republic of the Marshall
Islands, the Commonwealth of the Northern Marianna Islands, Puerto
Rico, and Palau are also funded under the regular program. 


      RECENT GAO REPORTS ON HEAD
      START
---------------------------------------------------------- Letter :2.3

In the past 3 years, we have issued several reports on the Head Start
program.  One report discussed local perspectives on barriers to
providing Head Start services.\10 That report, among other things,
concluded that Head Start lacked enough qualified staff to meet the
complex needs of children and families.  Other barriers included a
limited availability of health professionals in the community willing
to help Head Start staff in providing services and programs having
difficulties getting suitable facilities at reasonable costs.  In our
most recent report,\11 we concluded that the body of research
conducted on the Head Start program does not provide information on
whether today's Head Start is making a positive difference in
participants' lives.  Specifically, we found that the body of
research conducted on the program was inadequate for use in drawing
conclusions about the impact of the national program in any area in
which Head Start provides services such as school readiness or
health-related services.  We also stated that no single study of the
program used a nationally representative sample so that findings
could be generalized to the national program.  We recommended that
the Secretary of HHS include in HHS' research plan an assessment of
the impact of regular Head Start programs.  In commenting on this
report, HHS mentioned, among other things, that estimating program
impact at the national level is not appropriate because of the
extreme variability of local programs.  That is, local Head Start
sites have great flexibility, and, even though all programs share
common goals, they may operate very differently.  Thus, HHS considers
a single, large-scale, national study of impact to be
methodologically inappropriate. 


--------------------
\10 Early Childhood Programs:  Local Perspectives on Barriers to
Providing Head Start Services (GAO/HEHS-95-8, Dec.  21, 1994). 

\11 GAO/HEHS-97-59, Apr.  15, 1997. 


   HEAD START SERVES BOTH CHILDREN
   AND FAMILIES
------------------------------------------------------------ Letter :3

Head Start programs were funded to serve about 701,000 children at
any one time in program year 1996-97; however, the number of
different children enrolled in the program throughout the 1996-97
program year was about 782,000,\12 which averaged about 454 children
per program, ranging from a low of 17\13 to a high of 6,045.  The
number of different children enrolled in the program includes
children who are funded with all sources of funds, such as those
received from state agencies, and who have been enrolled in Head
Start for any length of time, even if they dropped out or enrolled
late, provided they have attended at least one class or, in
home-based programs, received at least one home visit. 

Head Start estimates capacity or the number of children that can be
served at any one time in two ways.  Total funded enrollment
(701,000) is the number of children that can be served at any one
time with Head Start grant funds, as well as funds from other
sources, such as state agencies.  This estimate includes children,
regardless of funding source, who are an integral part of the Head
Start program and who receive the full array of Head Start services. 
Head Start-funded enrollment (667,000) is an estimate of the number
of children that can be served at any one time with Head Start grant
funds only (see table II.1 in app.  II for enrollments by state). 
Although programs are authorized and expected to serve a certain
number of children, according to Head Start Bureau officials, local
programs may negotiate with their regional offices to adjust their
enrollment.  Thus, programs may choose to fill fewer slots or
establish more slots.  To illustrate, a program authorized to serve
50 children may choose to actually serve only 40 children or to serve
60.  By serving fewer children, the program can support other
enhancements, such as providing employees with full benefits.  Head
Start Bureau officials also stated that some states have regulations
and laws that also affect the number of slots that can be filled.  A
state that requires training and licensing of its early childhood
staff, for example, might be limited in the number of children it
could serve if licensed staff cost more.  Differences in the cost of
living can also affect the number of slots that can be filled. 

In addition, Head Start programs served about 711,000\14 families of
Head Start children, which Head Start regulations define as all
people living in the same household who are

  -- supported by the income of the parent or guardian and

  -- related by blood, marriage, or adoption. 

Head Start does not require that programs count the number of
individual family members served, however, so the number of services
provided them is unknown. 

The children and families Head Start served had some similar
demographic characteristics (see fig.  1).  Most were either 3 (31
percent) or 4 (63 percent) years old.  Most of the children--79
percent--spoke English as their main language.  Spanish-speaking
children constituted the next largest language group--18 percent. 
About 38 percent of the children were black, 33 percent were white,
and 25 percent were Hispanic.  About 13 percent of Head Start
children had some sort of disability. 

   Figure 1:  Age, Ethnicity, and
   Dominant Languages of Head
   Start Children

   (See figure in printed
   edition.)

\a Regular Head Start, which excludes Early Head Start and Migrant
programs (which serve a number of children in this age group), also
serves children who are under 3 years old--as well as children who
are 6.  However, both groups represent less than 1 percent of the
total. 

\b Other includes children who are Asian or Pacific Islanders and
American Indian or Alaska Native. 

\c Other includes children whose dominant language is an Asian,
Native American, or other language. 

Source:  Head Start's 1996-97 survey. 

Most Head Start families have more than one child; most have two or
three children (see fig.  2).  In addition, most (61 percent) have
only one parent or are headed by other relatives, or they are foster
families or have other living arrangements.  Head Start families are
generally very poor as indicated by several measures (see fig.  3). 
More than one-half are either unemployed or work part time or
seasonally, and about 60 percent have family incomes under $9,000 per
year.  Furthermore, only 5 percent have incomes that exceed official
poverty guidelines, and 46 percent receive TANF\15 benefits. 

   Figure 2:  Number of Children
   in Head Start Families and
   Family Type

   (See figure in printed
   edition.)

Note:  Other includes families headed by other relatives, foster
families, or those with some other living arrangement. 

Source:  Head Start's 1996-97 survey. 

   Figure 3:  Employment and
   Income Status of Head Start
   Families

   (See figure in printed
   edition.)

Source:  Head Start's 1996-97 survey. 


--------------------
\12 These figures may differ from Head Start's total official 1996-97
enrollment figure because this report focuses on the regular Head
Start program.  Thus, enrollment for Migrant and Native American
programs, as well as that for programs that serve pregnant women and
infants are not included.  In cases where enrollment data were not
available or not provided by our survey respondents, we imputed it
from Head Start's 1996-97 survey, which also collected this
information.  In cases where the data were not available from Head
Start's 1996-97 survey, we obtained it from the 1995-96 Head Start
survey. 

\13 We did find one program serving seven children, but this is an
anomaly, according to ACF, done only to ensure that children living
in a geographically isolated county have the opportunity to
participate in the Head Start program. 

\14 This figure, which was taken from Head Start's 1996-97 survey,
reflects the number of Head Start families with children enrolled in
regular Head Start programs.  For 96 percent of these families, Head
Start programs completed a family needs assessment.  Of all families,
some number received access to certain services such as education and
employment training. 

\15 TANF, established in 1996, replaced AFDC. 


   CHILDREN AND FAMILIES RECEIVED
   ACCESS TO AN ARRAY OF SERVICES
------------------------------------------------------------ Letter :4

Through Head Start, children received access to a large array of
services.  Children received medical and dental services,
immunizations, mental health services, social services, child care,
and meals.  According to Head Start's annual survey, nearly all
children enrolled in Head Start received medical screening/physical
exams, dental exams, and immunizations in the 1996-97 program year. 
Most children received medical screening, including all appropriate
tests and physical examinations as well as dental examinations by a
dentist.  Most had also received all immunizations required by the
Head Start immunization schedule for the child's age.  Children also
received education services in various settings. 

In addition, Head Start programs provided children's families access
to services (see table II.2 in app.  II).  Of the services we asked
about, parent literacy, social services, job training, and mental
health were the most frequently provided (see table II.4 in app. 
II).  Programs were least likely to provide dental and medical
services to siblings and other family members, with 64 percent
reporting they never provided dental services and 56 percent
reporting they never provided medical services. 


      SERVICES FOR CHILDREN
      PROVIDED IN PRIMARILY
      PART-DAY, PART-YEAR PROGRAMS
---------------------------------------------------------- Letter :4.1

Most children attended centers that operated part day and part year. 
About 90 percent of the children received services through center
programs.  Fifty-one percent of children attending centers went to
centers that operated 3 to 4 hours per day (see fig.  4).  Another 42
percent went to centers that operated between 5 and 7 hours per day. 
Only 7 percent of the children went to centers that operated 8 or
more hours per day.  In addition, 63 percent of the children attended
centers that operated 9 months of the year.  However, only 27 percent
of the children attended centers that operated 10 to 11 months, and
even less--7 percent--
attended centers that operated year round. 

   Figure 4:  Most Children Attend
   Part-Day, Part-Year Programs

   (See figure in printed
   edition.)

Note:  For this figure, programs operating 8 hours per day or more
are considered full day. 

Source:  GAO survey. 

According to Head Start's survey, about 38 percent of the families
needed full-day, full-year child care services.  However, this
proportion may increase dramatically as welfare reform is
implemented.\16 About 44 percent of the families needing full-day,
full-year child care services left their children at a relative's or
unrelated adult's home when the children were not in Head Start,
according to Head Start's survey. 

In 1997, the Congress appropriated additional funds to, among other
things, increase local Head Start enrollment by about 50,000
children.  Recognizing that an increasing proportion of Head Start
families work and many who may receive public assistance are
participating in welfare reform initiatives in response to TANF, the
Head Start Bureau announced that programs that provide more full-day,
full-year Head Start services will receive special priority for
funding.  Head Start urged programs to consider combining Head Start
expansion funds with other child care and early childhood funding
sources and to deliver services through partnerships such as
community-based child care centers.  This focus on providing
full-day, full-year services departs from previous expansion
priorities, which emphasized part-day, part-year, or home-based
services. 

For our review we talked with Head Start program officials who had
applied for expansion funds to meet the needs of working parents. 
Officials operating a program in Florida, for example, stated that
they plan to expand the number of days and hours the program
currently operates:  hours of operation will be expanded from 7:30
a.m.  to 4:00 p.m.  to 6:30 a.m.  to 7:00 p.m.  In addition,
officials operating a program in Vermont stated that it plans to
provide full-day, full-year services as well.  Their strategy
involves collaborating with an existing private center that will
offer children extended-day services. 


--------------------
\16 Under TANF, to avoid financial penalties, states must place 25
percent of adults receiving TANF benefits in work and work-related
activities in fiscal year 1997.  This required participation rate
rises to 50 percent in fiscal year 2002. 


   HEAD START'S MAJOR ROLE IN
   PROVIDING SERVICES IS
   FACILITATING ACCESS
------------------------------------------------------------ Letter :5

Head Start provides services in a number of ways.  In some instances,
Head Start programs both delivered and paid for services.  In most
cases, however, Head Start arranged for or referred participants to
services, and some other agency delivered and paid for the services. 
In these cases, Head Start provided information to help participants
get services from some other source.  For example, when asked the
main methods the programs used to provide medical services for
enrolled children, 73 percent of survey respondents said that they
referred participants to services, and some other entity or program,
such as Medicaid, primarily paid for the service (see fig.  5 and
table II.3 in app.  II).  Because most Head Start children are
eligible for Medicaid's Early and Periodic Screening, Diagnosis, and
Treatment Program, Head Start programs may refer children to Medicaid
providers; thus, Head Start provides access to these services with
little or no impact on the Head Start programs' budgets.  The same
was true of dental services and immunizations. 

About 40 percent of the programs reported Head Start funds, however,
as the primary source for meals and food, even though Head Start
expects programs to seek reimbursement for these expenses from the
U.S.  Department of Agriculture's (USDA) Adult and Child Care Food
Program. 

Education was the service most directly provided by Head Start for
enrolled children.  Nearly 90 percent of programs reported that they
both delivered and funded education services for enrolled children. 
Some Head Start program officials we interviewed, however, told us
that they contracted with a private preschool or child care centers
to provide education services.  These cases are rare, however; only 3
percent of respondents to our survey reported that Head Start funded,
but someone else delivered, education services.  These programs
purchased "slots" in centers operated by other organizations for
about 2,000 children. 

   Figure 5:  Head Start's Role in
   Providing Services for Enrolled
   Children

   (See figure in printed
   edition.)

Note:  Head Start programs deliver services in a variety of ways. 
This figure highlights the most direct and indirect ways Head Start
programs deliver services. 

Source:  GAO survey. 

In addition, Head Start typically provides services for children's
siblings and other family members indirectly (see table II.4 in app. 
II).  Of those respondents to our survey who indicated that they
provided services to siblings and other family members, at least half
reported that Head Start programs neither delivered nor paid for the
services.  As shown in figure 6, programs were more likely to report
full Head Start involvement (that is, the program paid for and
delivered the service) in the areas of education; social services;
child care; and meals, food, and nutrition.  For our review, we asked
several Head Start directors about some of the services they provided
directly to family members.  Program officials stated that they
typically provided services to the siblings, while providing services
to the enrolled child.  For example, education services provided to
enrolled children in a home-based program may be provided to siblings
as well, benefiting all enrolled children and their siblings.  The
director of a program in Montana, for example, stated that staff
bring along snacks for the siblings during home visits.  The director
of a program in Ohio stated that if the enrolled child, as well as
the child's siblings, needs a physical exam, they will ensure that
the siblings are also referred for physical exams. 

   Figure 6:  Head Start's Role in
   Providing Services for Family
   Members

   (See figure in printed
   edition.)

Note:  Head Start programs deliver services in a variety of ways. 
This figure highlights the most direct and indirect ways Head Start
programs deliver services. 

\a Percentages are based on those respondents who indicated that they
provided services. 

Source:  GAO survey. 


   MULTIPLE FUNDING SOURCES
   SUPPORTED PROGRAMS
------------------------------------------------------------ Letter :6

When asked to report the funds received from all sources to operate
their Head Start programs, survey respondents reported that different
funding sources supported Head Start programs (see fig.  7).  Most
programs--
about 90 percent--had multiple sources.  The number of different
funding sources that respondents reported varied (see fig.  8).  The
largest portion of programs, 40 percent, reported one other non-Head
Start funding source followed by 27 percent of the respondents who
reported two other non-Head Start funding sources.  At the other
extreme, however, the number of programs reporting six to seven
funding sources was small--
about 1 percent. 

   Figure 7:  Head Start Programs
   Supported by Multiple Funding
   Sources

   (See figure in printed
   edition.)

Source:  GAO survey. 

   Figure 8:  Number of Different
   Funding Sources Varied by
   Program

   (See figure in printed
   edition.)

Source:  GAO survey. 

The multiple funding sources included other federal programs, such as
the Child Care and Development Block Grant Program and the Social
Services Block Grant Program, both of which provide funding for child
care.  USDA was also a source of federal funding for programs, which,
among other things, supplemented Head Start program food and
nutrition resources by reimbursing food costs for eligible children. 
States, charitable organizations, and businesses also provided
program funds.  Some of this non-Head Start funding may have been
part of the 20 percent of nonfederal matching funds that programs
typically have to provide.  In addition, programs received in-kind
support for their operations such as building space, transportation,
training, supplies and materials, and health services.  In fact, many
Head Start agencies also operated other programs from which Head
Start participants sometimes received services but whose budgets were
separate from Head Start.  For example, we spoke to one Head Start
director whose program was operated by a public school.  According to
this official, the school district bears a number of the Head Start
program expenses.  For example, the school district bears a portion
of the cost of facilities, Head Start children receive their meals in
the cafeteria using school staff, and some staff funded with title I
and special education money provide services for Head Start children. 

As shown in table 1, respondents reported receiving a total of $3.1
billion to operate their Head Start programs in their most recently
completed budget year, of which $2.7 billion,\17 or 85 percent, was
income from the Head Start grant. 



                          Table 1
          
            For Responding Programs, Head Start
          Grants Were Programs' Largest Source of
                           Funds

                                                Percentage
                                                  of total
Source                                Amount         funds
------------------------------  ------------  ------------
Head Start                      $2,648,213,3            85
                                          51
Child Care and Development         9,338,689            \a
 Block Grant
USDA                             168,109,049             5
Social Services Block Grant        8,532,352            \a
 (title XX)
Other federal                     23,370,625             1
State                            168,885,256             5
Foundations, charities, and        9,408,674            \a
 businesses
Other nonfederal                  68,263,099             2
==========================================================
Total                           $3,104,119,0
                                          95
----------------------------------------------------------
\a Less than 1 percent of the total funding. 

Head Start grant funds were the largest single source of funding for
most programs.  For example, for about 77 percent of the respondents,
Head Start funding represented between 80 and 100 percent of the
programs' total funds. 

Other non-Head Start funding totaled about $456 million and
represented about 15 percent of the total funds received.  The states
provided the largest source of other funding, which totaled about
$169 million and represented about 5 percent of the total funds in
programs' last budget year.  The next largest source of funds came
from a federal source--USDA.  USDA funding of $168 million also
represented about 5 percent of the total program funds. 


--------------------
\17 This figure is significantly lower than the 1996 Head Start
program appropriation of $3.6 billion and the 1997 appropriation of
about $4 billion for several reasons.  First, only $3.2 billion of
the 1996 appropriation and $3.6 billion of the 1997 appropriation
were allocated to directly support local Head Start programs in the
states and territories.  Second, this amount includes the amount
spent by Early Head Start Programs and Parent Child Centers, which
were not included in our analysis.  Third, the amounts of funding
received by programs serving no children are excluded from this
analysis.  Programs that serve no children may maintain a central
office staff responsible for, among other things, monitoring and
overseeing programs.  Finally, a number of programs did not provide
income information. 


      NON-HEAD START FUNDING
      INCREASED AMOUNT AVAILABLE
      PER CHILD; FUNDING ACROSS
      PROGRAMS AND STATES VARIED
      WIDELY
---------------------------------------------------------- Letter :6.1

The non-Head Start funding increased the amount of funds available
per child.\18

Average Head Start grant funds per child were $4,637\19 for the
responding programs.  The total amount of funds per child, including
Head Start grant funds, was $5,186 per child,\20 a difference of
about $549 or 12 percent Head Start-wide.  Across most states and
territories, the non-Head Start funding increased the amount
available per child (see table II.5 in app.  II).  As shown in figure
9, for the majority of states, the additional funds increased the
amount available per child by over 10 percent; in four states and the
District of Columbia, additional funds increased the amount available
per child by at least 21 percent. 

   Figure 9:  Additional Funding
   From Other Sources Increased
   Amount Available in Almost All
   States

   (See figure in printed
   edition.)

Source:  GAO survey. 

Head Start and total funding per child varied considerably (see table
II.6 in app.  II).  Across all programs, the median amount of Head
Start grant funds per child was $4,450 for the responding programs
but ranged from a low of $792 to a high of $16,206.\21 Median total
funds per child of $4,932 across all programs ranged from $1,081 to
$17,029 per child. 

Several reasons may explain the funding variation by state and
program such as the hours and days of program operation and the
characteristics of the children served.  We spoke with a Head Start
director in the District of Columbia, whose program had high per
child Head Start and total funding.  The director told us that the
program provided service for children in centers that operated year
round and for 10 hours or more per day.  We also spoke with a
director of a program in New York City that had high funding per
child.  That program provided part-day center services.  The children
it served, however, had multiple disabilities or special needs.  We
also spoke with directors whose funding per child was low.  One
director stated that because the Head Start program is operated by
the public school, the school bears a number of the expenses--such as
facilities and food cost as well as some staff costs--of the Head
Start program. 


--------------------
\18 We instructed respondents not to include nonmonetary
contributions even though they may have received such contributions
as part of the required 20-percent match. 

\19 Average Head Start grant funding per child was calculated by
dividing Head Start grant funds by Head Start-funded enrollment. 

\20 Total funding per child was calculated by dividing the funding
from all sources, including Head Start grant funds, by the total
funded enrollment. 

\21 The amount of Head Start funds per child for 5 percent of the
programs was $3,000 or less; for 27 percent, $4,000 or less; for 71
percent, $5,000 or less; and for 89 percent, $6,000 or less.  For
about 11 percent of the programs, Head Start grant funds per child
were over $6,000. 


      PERSONNEL COSTS ACCOUNTED
      FOR MOST PROGRAM
      EXPENDITURES
---------------------------------------------------------- Letter :6.2

Head Start programs spent 68 percent of their overall funds on
personnel.  Personnel included teachers, teacher aides, home
visitors, social service workers, and administrators.  Personnel
costs for educational services were the single largest personnel
expense (53 percent).  According to Head Start's annual survey, Head
Start programs employed many staff.  About 129,000 staff worked
either full or part time in regular Head Start programs nationwide
(see fig.  10).  These staff, in addition to providing direct
services, such as education, facilitated children's and families'
access to services.  One way Head Start tries to encourage parental
involvement is by providing parents preference for employment in Head
Start programs as nonprofessionals.  Thus, about one-third of the
staff were parents of current and former Head Start children. 

The remaining funds--32 percent--were spent on nonpersonnel-related
expenses.  Interestingly, direct payment for medical services
accounted for only 3 percent of nonpersonnel-related expenses.  In
this area, programs are encouraged to seek non-Head Start sources of
funds, and many programs link families and children to the Medicaid
Early and Periodic Screening, Diagnosis, and Treatment Program. 

   Figure 10:  Head Start Programs
   Employed Many Types of Staff

   (See figure in printed
   edition.)

Source:  GAO and Head Start surveys. 

In addition, programs spent their funds on a range of services.  As
shown in figure 11, education services were the largest expense (39
percent).  The smallest expenses were for health (4 percent),
disabilities services (3 percent), and parent involvement services (3
percent). 

   Figure 11:  Programs Spent
   Funds on a Range of Services

   (See figure in printed
   edition.)

Source:  GAO survey. 


   OTHER PROGRAMS SERVING HEAD
   START-ELIGIBLE CHILDREN
   OPERATED IN SAME COMMUNITIES AS
   HEAD START PROGRAMS
------------------------------------------------------------ Letter :7

Many Head Start programs reported that state-funded preschools (70
percent), other preschools, child development and child care centers
(90 percent), and family day care homes (71 percent) operated in
their communities serving Head Start-eligible children.  The extent
to which these programs resemble Head Start is not known. 

However, programs that serve disadvantaged children may--like Head
Start--help children and families obtain additional services such as
medical and social services.  To test this assumption, we gathered
information on Head Start agencies that also operated other early
childhood programs.  About 11 percent of the Head Start respondents
(in 39 states) reported that they operated other early childhood
programs and that these programs served Head Start-eligible children. 
These children received some or most--but not all--of the services
typically provided by Head Start programs.  Respondents reported
serving about 14,000 Head Start-eligible children through these other
programs.  California served the greatest number of such children
(3,216) followed by Kentucky (2,652) (see table II.7 in app.  II). 
These programs provided many of the same services as Head Start
programs, but not all services were provided to all children. 
Education services, meals, social services, and immunizations were
the most often provided; dental, medical, and other nutrition
services were the least often provided.  Thirty percent of the
programs responded that they provided no services to families. 
Families or siblings were more likely to receive social services and
parent literacy training through Head Start and less likely to
receive medical services, such as dental, mental health, and
immunizations. 


   CONCLUSIONS
------------------------------------------------------------ Letter :8

In many respects, the Head Start program is at a crossroads because
the context in which it operates today differs greatly from that of
30 years ago when the program was established.  The services
available to poor children have changed and communities have enhanced
resources for serving poor children and their families. 
Consequently, Head Start facilitates or brokers many services
provided by others, referring and linking families to these services,
rather than providing them directly.  The one service that almost all
Head Start programs provide directly is education, although the
number of early childhood education programs other than Head Start
has grown in the past 30 years. 

Furthermore, changes in welfare policy have important implications
for Head Start.  Most Head Start programs operate for only part of
the day and part of the year.  As changes in welfare policy require
increasing numbers of poor people--including Head Start parents--to
seek and maintain employment, however, the need for full-day,
full-year services will intensify.  The administration's proposals to
help working parents secure affordable, quality child care include
substantially increasing Head Start enrollment.  Head Start's
predominantly part-day, part-year programs present obstacles for
meeting the needs of working families.  Head Start will need to
balance the administration's wish to serve more eligible children,
which has typically been done by creating more part-day, part-year
slots, with the need for more full-day, full-year services more
compatible with working families' needs. 

Finally, information about Head Start's effectiveness and the
efficiency of various Head Start models is lacking.  As we reported
earlier, although Head Start research has been conducted, it does not
provide information on whether today's Head Start is positively
affecting the lives of today's participants whose world differs
vastly from that of the 1960s and early 1970s.\22 In addition,
funding for Head Start programs varies widely.  We do not know to
what extent, however, this variation may be attributable to
efficiencies in providing services or to other factors such as
programs' ability to leverage other community resources,
characteristics of the population served, or program structure. 


--------------------
\22 GAO/HEHS-97-59, Apr.  15, 1997. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :9

ACF provided general comments about the Head Start program and
specific technical comments, which we incorporated in the report as
appropriate.  Four of ACF's comments that were not incorporated in
the report addressed services provided to children's siblings, data
on hours and months of attendance, use of funds for food costs, and
hiring of parents. 

ACF commented that our discussion of services provided to enrolled
children's siblings is misleading because it implies that Head Start
programs are actively providing services to such children.  ACF
contends that Head Start programs do not use grant funds to provide
services to siblings and that such services are provided only to the
extent that they are part of the enrolled child's services. 
Nevertheless, a small percentage of Head Start survey respondents
reported that they did use Head Start funds to deliver services to
families and siblings.  Our report emphasizes, however, that when
provided, many of these services are neither paid for nor delivered
by Head Start.  Head Start facilitates siblings' and families' access
to services in much the same way as it does to enrolled children.  We
also report that our interviews with Head Start officials showed that
siblings sometimes receive services as part of the program's services
to the enrolled child.  For example, Head Start staff may bring along
snacks for siblings during home visits and provide education services
for the siblings during such visits.  It is likely that in such a
situation, the Head Start program would consider this to be providing
services directly because Head Start funds might have been used to
pay the staff's salary and the cost of siblings' snacks. 

In addition, ACF commented that Head Start does collect data on the
number of hours per day or months per year that enrolled children
attend center programs and that such information is available through
its Head Start Cost data system.  During this study, we reviewed the
Head Start Cost data system and found--and Head Start officials had
previously confirmed--that reporting of Head Start Cost data is
optional and not all programs provide such data.  Furthermore, the
data collected by the system on the number of hours per day or months
per year that children attend center programs really reflect
programs' projected center operating schedules, not their actual
schedules. 

ACF also stated that our discussion of USDA reimbursement is somewhat
inaccurate and that USDA covers the vast majority of all food costs
incurred by Head Start programs, with Head Start grant funds paying
only a small portion of these costs.  AFC stated that it is not
conceivable that 40 percent of Head Start programs are using Head
Start funds as their primary source of meals and food because
programs are required to seek such reimbursement from USDA.  We did
not change our figures in the report, however, because they directly
reflect the reports of our survey respondents. 

In addition, ACF stated that the discussion of hiring parents should
clarify that Head Start hires parents only for jobs for which they
are qualified and that many parents have advanced through the Head
Start ranks and now hold professional-level positions in the program. 
We assessed, however, neither the qualifications of the parents Head
Start employs nor the number who hold professional-level positions in
the programs and therefore the report does not address these issues. 

We are sending copies of this report to the Secretary of Health and
Human Services, the Head Start Bureau, appropriate congressional
committees, and other interested parties.  Please call me at (202)
512-7014 if you or your staff have any questions about this report. 
Major contributors to this report are listed in appendix V. 

Carlotta C.  Joyner
Director, Education and
 Employment Issues

List of Requesters

The Honorable William F.  Goodling
Chairman
The Honorable William L.  Clay
Ranking Minority Member
Committee on Education and the Workforce
House of Representatives

The Honorable Frank D.  Riggs
Chairman
The Honorable Matthew G.  Martinez
Ranking Minority Member
Subcommittee on Early Childhood, Youth and Families
Committee on Education and the Workforce
House of Representatives

The Honorable Daniel R.  Coats
Chairman
The Honorable Christopher J.  Dodd
Ranking Minority Member
Subcommittee on Children and Families
Committee on Labor and Human Resources
United States Senate

The Honorable Randy Cunningham
House of Representatives

The Honorable Dale E.  Kildee
House of Representatives


OBJECTIVES, SCOPE, AND METHODOLOGY
=========================================================== Appendix I


   OBJECTIVES
--------------------------------------------------------- Appendix I:1

In preparation for Head Start's reauthorization, the Chairman and
Ranking Minority Member, House Committee on Education and the
Workforce; the Chairman and Ranking Minority Member, Subcommittee on
Early Childhood, Youth and Families, House Committee on Education and
the Workforce; Chairman and Ranking Minority Member, Subcommittee on
Children and Families, Senate Committee on Labor and Human Resources;
and Representatives Cunningham and Kildee asked us to describe the
(1) number and characteristics of Head Start participants, (2)
services provided and the way they are provided, (3) federal and
nonfederal program dollars received and spent by programs delivering
Head Start services, and (4) other programs providing similar--in
part or in whole--early childhood services.  As agreed with the
requesters' offices, however, we did not comprehensively review other
early childhood programs. 


   SCOPE
--------------------------------------------------------- Appendix I:2

We focused on collecting information on Head Start's regular program;
thus, programs serving special populations, such as migrant and
Native American and pregnant women and infants, were excluded.  About
85 percent of Head Start children are served through regular Head
Start programs.  Programs for special populations represent only a
small portion of Head Start children served and each program is
unique. 


   METHODOLOGY
--------------------------------------------------------- Appendix I:3

We administered our survey about the same time Head Start conducted
its annual survey (May 1997), which we also analyzed.  Both surveys
collected information on the 1996-97 program year, which spanned
September 1996 to May 1997.  Head Start refers to its annual survey
as the Program Information Report (PIR). 

Our survey was mailed to 1,783 regular Head Start programs; of these,
1,722 were determined to be active Head Start programs that served
children.\23 The PIR was a second source of information on programs. 
(Both instruments are described in more detail in the following
section.) Because the mailing list HHS provided us was the same one
used for the PIR, all regular Head Start programs should have
received both our survey and the PIR. 


--------------------
\23 We omitted from our analyses those grantees who indicated on the
survey that they did not directly operate a program that served
children.  We omitted 55 programs on the list we were given that we
later discovered were inactive or were being deactivated as well as 6
programs that appeared inactive because they did not respond to our
survey, the 1995-96 PIR, or the 1996-97 PIR. 


      DESCRIPTION OF OUR SURVEY
------------------------------------------------------- Appendix I:3.1

To obtain a broader understanding of Head Start, our questionnaire
mostly avoided questions appearing on the PIR.  For example, we asked
respondents to report the number of months and hours of the day
children attended centers, the number of classes operated on
weekends, and whether Head Start programs paid for children to attend
centers operated by someone else.  We also asked them the number of
months they provided services in their home-based programs.  In
addition, we asked how services are provided to enrolled children and
their family members and the extent to which family members are
served.  We also asked them about the funds they received to operate
their Head Start programs as well as their Head Start program
expenditures.  We asked Head Start programs if they served Head
Start-eligible children through other early childhood programs they
operated and about the services provided them and their families. 
Our complete survey appears in appendix III. 


      DESCRIPTION OF THE PIR
------------------------------------------------------- Appendix I:3.2

HHS requires that all grantees and delegate agencies complete annual
PIRs.  Although the questions asked in the report change somewhat
from year to year, in general, the report asks about program
management issues.  Among other things, the 1996-97 report asked
about the numbers of children served by the Head Start program in
that program year, the number receiving particular kinds of services,
and details about the Head Start staff, for example, the number of
staff in various kinds of positions, their educational level, and so
forth.  All Head Start programs are required to complete a PIR;
however, not all had done so at the time of our analyses. 


      RESPONSE RATES
------------------------------------------------------- Appendix I:3.3

Because we collected data from two major sources, response rates are
shown in table I.1 in several ways.  The overall response rate (98
percent) is based on the number of eligible respondents divided by
the number from which information was obtained from at least one
source.  Our survey response rate is based on the number of eligible
respondents divided by the number completing and returning our survey
(86 percent).  Finally, the PIR response rate (94 percent) is based
on the number of eligible respondents for whom HHS provided us with
completed 1996-97 PIR information. 



                         Table I.1
          
            Number Responding and Response Rates

                                                  Response
                                      Number          rate
Survey                            responding     (percent)
------------------------------  ------------  ------------
GAO survey only                           72            86
PIR only                                 206            94
Both GAO survey and PIR                1,412            98
----------------------------------------------------------

      NONSAMPLING ERRORS AND DATA
      IMPUTATIONS
------------------------------------------------------- Appendix I:3.4

All surveys are vulnerable to some nonsampling errors, including
errors due to imperfect population lists, measurement errors due to
ambiguous questions or inaccurate responding, or errors due to lack
of response.  These errors may affect both our survey and the PIR to
some unknown degree. 

We took several steps to minimize the impact of these errors.  First,
we examined responses for extreme values.  In many cases, we reviewed
questionnaires for explanations of questionable responses.  When we
could not resolve questions, we called survey respondents for
clarification.  In a few cases, respondents had reported numbers
incorrectly; and, in these cases, we corrected the data, or, if
correction was not possible, we rejected the erroneous data.  Second,
we looked for a systematic pattern in the distribution of
nonrespondents.  Because we thought that program size (defined by
total funded enrollment) might be related to response patterns, we
examined whether programs of various sizes were more or less likely
to respond.  Although smaller programs tended to be somewhat less
likely to respond, the difference in the response rate, coupled with
the small number of the nonrespondents, yielded an inconsequential
overall impact. 

In most cases we based our analyses simply on the answers of survey
respondents.  No weighting for nonresponse was done because our
response rate was so high that adjustments for nonresponse would have
hardly affected our findings.  In reporting total enrollment
information, however, we adjusted the data so that more complete
total enrollment could be reported.  For those programs lacking
enrollment data, we imputed enrollment from the 1996-97 PIR (or in
cases where the 1996-97 PIR was not available, we used the 1995-96
PIR). 


      TELEPHONE INTERVIEWS AND
      SITE VISITS
------------------------------------------------------- Appendix I:3.5

To gather illustrative information, we conducted telephone interviews
of nine Head Start programs in Florida, Iowa, Montana, New York,
Ohio, Pennsylvania, Vermont, Arkansas, and Oregon, which were
judgmentally selected.  We selected large and small programs in
different parts of the country and programs representing a mixture of
the types of program options Head Start offers such as centers and
homes.  We selected programs operated by different types of
agencies--including community action agencies, universities, and
nonprofit organizations.  In addition, we selected grantees that
operated the program directly as well as those that did not and
programs that received funds from various sources to operate their
program as well as those operating with only Head Start grant funds. 
Finally, we selected programs in which a portion of the total
enrollment was funded with non-Head Start income.  We asked Head
Start program officials a number of questions, including whom they
served, their funding sources, availability of other early childhood
programs in their communities, and general questions about program
operations.  We also asked programs about further program expansion. 
Finally, we validated selected responses to our survey by visiting
several Head Start programs, which we also wanted to observe.  We
visited programs in Philadelphia, Pennsylvania; Boston,
Massachusetts; Kansas City, Missouri; Chicago, Illinois; Atlanta,
Georgia; and Seattle, Washington. 

We conducted our work between March 1997 and November 1997 in
accordance with generally accepted government auditing standards. 


DATA ON HEAD START PROGRAMS
========================================================== Appendix II

The tables in this appendix provide selected information on Head
Start programs.  Table II.1 presents data on Head Start enrollments
by state.  Tables II.2 provides data on the extent to which families
received services, and tables II.3 and II.4 present information on
how services are provided to enrolled children and their families. 
Table II.5 presents by state information on the average Head Start
grant funding per child and the average funding per child from all
sources, including Head Start grants.  Table II.6 presents data on
the variation in funds per child by and within state.  Table II.7
presents information on the number of Head Start-eligible children
receiving services through other early childhood programs that Head
Start agencies operate. 



                         Table II.1
          
               Head Start Enrollment by State

                      Funded enrollments
                  --------------------------
                    Head Start         Total        Actual
State                 funded\a      funded\b    enrollment
----------------  ------------  ------------  ------------
Head Start-wide        666,695       701,029       781,889
Alabama                 14,184        14,184        15,266
Alaska                   1,173         1,509         1,759
Arizona                  9,290         9,467        11,672
Arkansas                 8,622         9,037        10,324
California              70,337        74,512        87,459
Colorado                 5,580         5,969         6,671
Connecticut              5,556         5,892         6,555
Delaware                 1,492         1,757         1,955
District of              2,867         2,869         3,267
 Columbia
Florida                 26,545        26,807        28,330
Georgia                 19,159        19,159        21,048
Hawaii                   2,126         2,126         2,487
Idaho                    1,872         2,027         2,106
Illinois                32,260        33,050        36,464
Indiana                  9,993        10,059        11,224
Iowa                     5,926         5,935         6,787
Kansas                   5,574         5,723         6,344
Kentucky                13,737        14,205        15,738
Louisiana               18,388        18,388        19,938
Maine                    2,781         3,144         3,511
Maryland                 8,262         9,342        10,215
Massachusetts           10,497        11,086        12,170
Michigan                30,417        31,409        34,810
Minnesota                8,491        10,892        11,982
Mississippi             23,743        23,743        24,972
Missouri                13,774        13,953        16,514
Montana                  2,257         2,257         2,538
Nebraska                 3,477         3,517         3,991
Nevada                   1,749         1,749         2,039
New Hampshire            1,122         1,122         1,206
New Jersey              12,349        12,676        13,646
New Mexico               5,997         6,002         6,262
New York                37,170        38,641        45,289
North Carolina          15,318        15,437        16,682
North Dakota             1,678         1,678         1,848
Ohio                    34,218        47,550        51,286
Oklahoma                11,085        11,182        13,329
Oregon                   4,700         5,701         6,486
Outer Pacific            5,860         6,040         6,216
Pennsylvania            24,617        24,623        27,242
Puerto Rico             31,012        31,012        33,393
Rhode Island             1,853         2,175         2,435
South Carolina          10,070        10,070        10,634
South Dakota             1,975         1,975         2,298
Tennessee               13,350        13,462        14,893
Texas                   49,395        50,506        57,495
Utah                     4,051         4,051         4,541
Vermont                  1,078         1,078         1,182
Virginia                10,518        11,903        13,004
Virgin Islands           1,430         1,430         1,246
Washington               8,107         8,300         9,907
West Virginia            5,993         6,013         6,926
Wisconsin               12,341        13,346        14,820
Wyoming                  1,279         1,289         1,487
----------------------------------------------------------
\a Head Start-funded enrollment is an estimate of the number of
children who can be served at any one time with Head Start grant
funds only. 

\b Total funded enrollment is the number of children who can be
served at any one time with Head Start grant funds as well as other
sources of funds such as those received from state agencies.  It
includes children, regardless of funding source, who are an integral
part of the Head Start program and who receive the full array of Head
Start services. 



                         Table II.2
          
           Percentages of Enrolled Children Whose
                 Families Received Services

                                             50 to
                                     25 to    less
                            Less      less    than   75 or
Services          None   than 25   than 50      75    more
--------------  ------  --------  --------  ------  ------
Education for       50        33        10       4       3
 siblings
Medical             56        33         7       2       2
Dental              64        28         5       2       2
Mental health       24        52        16       4       3
Immunizations       54        29         9       5       4
Social              12        26        21      18      23
 services
Meals/food          34        40        15       6       5
Other               31        40        14       8       7
 nutrition
 services
Child care          50        34        11       4       2
Parent              11        44        25      12       9
 literacy
Job training        18        42        22      12       6
 for parents
----------------------------------------------------------
Note:  Percentages may not add up to 100 due to rounding. 



                         Table II.3
          
          Who Delivered and Who Paid for Services
                     for Enrolled Child

                                    Head
                Head      Head     Start  Others
               Start     Start    set up  delive
              delive  delivere        or    red;
                 red        d;  referred    Head
                 and    others  ; others   Start       Not
Services      funded    funded  funded\a  funded  provided
------------  ------  --------  --------  ------  --------
Education         88         3         6       2         1
Medical            6         7        73      12         2
Dental             7         7        69      17         1
Mental            20         5        40      33         1
 health
Immunization       2         6        82       6         4
 s
Social            46         6        44       3         0
 services
Meals/food        31        43        17       9         1
Other             44        11        30      12         4
 nutrition
 services
Child care        21         8        30       3        39
----------------------------------------------------------
Note:  Percentages may not add up to 100 due to rounding. 

\a Under this scenario, Head Start acts as a facilitator, neither
delivering nor funding the service. 



                          Table II.4
           
           Who Delivered and Paid for Services for
                     Other Family Members

                                    Head
                Head      Head     Start
               Start     Start    set up    Others
              delive  delivere        or  delivere
                 red        d;  referred   d; Head
                 and    others  ; others     Start       Not
Services      funded    funded    funded    funded  provided
------------  ------  --------  --------  --------  --------
Education         16         3        31         1        48
Medical            1         3        48         1        46
Dental             1         3        42         2        53
Mental            10         4        56        12        17
 health
Immunization       1         4        53         1        41
 s
Social            19         7        62         2        10
 services
Meals/food        18        12        35         3        33
Other             17         6        47         3        27
 nutrition
 services
Child care        12         6        33         3        47
Job training       9         8        66         3        15
Parent            17         8        62         5         8
 literacy
------------------------------------------------------------
Note:  Percentages may not add up to 100 due to rounding. 



                         Table II.5
          
          Average Amount of Funding From Non-Head
          Start Sources Increased Amount Available
                         per Child

                                          Difference\a
                                      --------------------
              Head Start
                   grant       Total
               funds per   funds per
               child (in   child (in
State           dollars)    dollars)      Amount   Percent
------------  ----------  ----------  ----------  --------
Head Start-       $4,637      $5,186        $549        12
 wide
Alabama            4,065       4,558         493        12
Alaska             6,296       7,862       1,566        25
Arizona            4,753       4,874         121         3
Arkansas           3,682       4,291         610        17
California         5,507       5,829         321         6
Colorado           4,214       4,458         244         6
Connecticut        5,335       6,551       1,216        23
Delaware           4,187       4,691         504        12
District of        4,402       6,583       2,181        50
 Columbia
Florida            4,671       5,743       1,072        23
Georgia            4,649       5,313         664        14
Hawaii             4,592       5,274         682        15
Idaho              5,255       5,640         386         7
Illinois           4,175       4,565         391         9
Indiana            4,266       4,579         313         7
Iowa               4,286       5,123         837        20
Kansas             4,072       4,416         343         8
Kentucky           4,142       4,464         322         8
Louisiana          4,475       5,021         546        12
Maine              4,763       5,593         829        17
Maryland           3,968       4,971       1,004        25
Massachusett       5,762       6,756         994        17
 s
Michigan           4,271       4,926         655        15
Minnesota          4,597       4,796         199         4
Mississippi        4,069       4,596         527        13
Missouri           4,496       4,853         357         8
Montana            4,271       4,563         292         7
Nebraska           3,948       4,167         219         6
Nevada             4,859       5,156         296         6
New                5,402       5,909         508         9
 Hampshire
New Jersey         6,128       6,928         800        13
New Mexico         4,653       5,293         640        14
New York           5,519       6,259         741        13
North              4,625       5,226         601        13
 Carolina
North Dakota       4,069       4,312         243         6
Ohio               4,102       4,177          76         2
Oklahoma           3,736       4,281         545        15
Oregon             5,997       6,515         518         9
Outer              1,692       2,290         598        35
 Pacific
Pennsylvania       4,853       5,247         394         8
Puerto Rico        4,138       5,045         907        22
Rhode Island       4,922       5,864         942        19
South              4,766       5,708         941        20
 Carolina
South Dakota       4,125       4,375         249         6
Tennessee          4,657       5,317         660        14
Texas              4,758       5,362         603        13
Utah               3,985       4,148         163         4
Vermont            4,881       5,263         382         8
Virginia           4,572       5,103         531        12
Washington         5,727       6,564         836        15
West               4,619       5,260         641        14
 Virginia
Wisconsin          4,447       4,719         273         6
Wyoming            4,108       4,458         350         9
----------------------------------------------------------
\a Because we rounded Head Start grant funds per child to the nearest
dollar, our calculations of the difference between the two in some
cases differ slightly from the relative difference calculated by
others. 



                         Table II.6
          
           Funding per Child Varied by and Within
                           State

                Head Start funds per    Total funds per
                       child                 child
                --------------------  --------------------
                Averag                Averag
                     e                     e
                (media                (media
State               n)   Low    High      n)   Low    High
--------------  ------  ----  ------  ------  ----  ------
Head Start-     $4,450  $792  $16,20  $4,932  $1,0  $17,02
 wide                              6            81       9
Alabama          4,048  3,21   6,064   4,511  3,21   6,692
                           6                     6
Alaska           6,689  2,61   8,618   7,693  3,13  10,455
                           5                     1
Arizona          2,898  2,26   7,948   3,072  2,26   8,159
                           7                     7
Arkansas         3,589  1,46   4,523   4,075  2,48   7,435
                           0                     2
California       4,912  2,33  14,984   5,330  2,27  15,386
                           9                     7
Colorado         4,420  3,16   5,591   4,774  3,41   9,542
                           6                     4
Connecticut      5,111  3,84   7,839   6,253  4,69   8,445
                           6                     3
Delaware         4,157  3,32   5,445   4,421  3,43   7,091
                           7                     3
District of      4,493  3,14   9,077   7,914  4,44  15,203
 Columbia                  4                     7
Florida          4,547  2,12   6,147   5,302  3,90   7,898
                           2                     9
Georgia          4,287  2,42   6,048   4,744  2,42   7,880
                           2                     2
Hawaii           4,400  4,35   4,703   4,797  4,65   5,560
                           6                     2
Idaho            5,231  4,66   5,734   5,615  4,71   6,192
                           4                     5
Illinois         4,157  2,19   9,195   4,593  2,22   9,471
                           2                     4
Indiana          4,238  3,03   6,583   4,443  3,03  12,724
                           3                     3
Iowa             4,252  1,66   8,331   4,837  2,31   9,705
                           9                     6
Kansas           4,110  2,72   5,453   4,508  2,72   5,768
                           5                     5
Kentucky         4,076  2,40   7,339   4,388  2,61   8,356
                           3                     0
Louisiana        4,179  3,20   7,347   4,740  3,32   8,082
                           7                     6
Maine            4,749  3,90   6,489   5,659  4,90   6,668
                           3                     7
Maryland         4,612  2,83   5,542   4,959  3,01  10,216
                           0                     6
Massachusetts    5,707  3,60  11,697   6,739  4,46  11,752
                           6                     1
Michigan         4,178  1,79   6,724   4,396  1,79  10,611
                           4                     4
Minnesota        4,365  3,76   5,757   4,583  3,95   5,780
                           5                     0
Mississippi      3,996  3,85   5,218   4,544  4,24   5,963
                           3                     0
Missouri         4,319  1,72   9,518   4,641  2,07   7,905
                           7                     4
Montana          4,431  3,80   5,356   4,684  4,14   5,570
                           2                     6
Nebraska         4,025  3,26   5,368   4,148  3,27   6,710
                           8                     4
Nevada           6,126  4,10  12,882   6,375  4,42  13,167
                           8                     4
New Hampshire    5,535  4,71   6,246   6,021  5,06   7,131
                           0                     7
New Jersey       5,890  4,10   9,760   6,426  4,10  10,409
                           8                     8
New Mexico       4,240  3,09   7,963   4,626  3,09   9,029
                           9                     9
New York         5,587  1,82  16,206   6,153  1,82  17,029
                           5                     5
North Carolina   4,604  3,70   7,083   5,132  3,70   7,433
                           1                     1
North Dakota     4,123  3,81   4,462   4,483  3,88   4,603
                           7                     0
Ohio             4,001  2,83   8,936   4,170  2,41   7,615
                           5                     3
Oklahoma         3,763  3,26   4,298   4,337  3,62   4,807
                           4                     0
Oregon           6,162  3,89   8,625   6,326  4,92   8,041
                           4                     0
Outer Pacific    2,217   792   4,071   2,217  1,30   5,633
                                                 5
Pennsylvania     4,815  3,52   6,640   5,206  3,70   9,684
                           8                     8
Puerto Rico      4,085  3,55   5,585   4,790  4,07   7,350
                           6                     1
Rhode Island     4,965  4,24   5,301   6,120  5,63   6,278
                           8                     1
South Carolina   4,424  3,78   9,080   4,873  4,28  11,492
                           0                     7
South Dakota     4,176  3,74   4,757   4,319  4,06   5,307
                           0                     8
Tennessee        4,536  3,54   7,306   5,001  3,87   7,999
                           7                     2
Texas            4,447  1,08   8,103   4,995  1,08   8,938
                           1                     1
Utah             3,957  2,98   5,021   4,048  3,00   5,229
                           2                     7
Vermont          4,785  4,69   5,557   5,075  4,85   6,914
                           1                     4
Virginia         4,409  3,24   6,763   4,935  3,03   7,156
                           9                     6
Washington       5,782  3,33   7,453   5,986  4,49  12,175
                           1                     7
West Virginia    4,176  3,01   7,020   4,519  3,40   9,085
                           3                     5
Wisconsin        4,513  2,61   6,175   4,625  2,66   7,151
                           4                     2
Wyoming          4,328  3,69   4,708   4,385  3,69   5,522
                           3                     3
----------------------------------------------------------


                         Table II.7
          
           Head Start-Eligible Children Served by
             Head Start Agencies in Other Early
          Childhood Programs Received Some or Most
            of the Services Head Start Children
                          Received

                                           Receive some or
State                                        most services
--------------------------------------  ------------------
Alabama                                                 \a
Alaska                                                  48
Arizona                                                 40
Arkansas                                               443
California                                           3,216
Colorado                                                24
Connecticut                                            128
Delaware                                                \a
District of Columbia                                    20
Florida                                                538
Georgia                                                467
Hawaii                                                  51
Idaho                                                   63
Illinois                                               341
Indiana                                                 66
Iowa                                                    41
Kansas                                                  81
Kentucky                                             2,652
Louisiana                                               \a
Maine                                                  129
Maryland                                                40
Massachusetts                                          111
Michigan                                               314
Minnesota                                              109
Mississippi                                             \a
Missouri                                                \a
Montana                                                 31
Nebraska                                                58
Nevada                                                  \a
New Hampshire                                          113
New Jersey                                              98
New Mexico                                              \a
New York                                               462
North Carolina                                         157
North Dakota                                            \a
Ohio                                                   171
Oklahoma                                                \a
Oregon                                                 137
Outer Pacific                                           \a
Pennsylvania                                           451
Puerto Rico                                          1,667
Rhode Island                                            34
South Carolina                                          \a
South Dakota                                            30
Tennessee                                               40
Texas                                                1,249
Utah                                                    \a
Vermont                                                 18
Virginia                                                37
Virgin Islands                                          \a
Washington                                             467
West Virginia                                           60
Wisconsin                                              175
Wyoming                                                 14
==========================================================
Total                                               14,391
----------------------------------------------------------
\a Respondents in these states and territories did not report serving
children who received some or most Head Start-like services. 




(See figure in printed edition.)Appendix III
GAO'S NATIONAL SURVEY OF HEAD
START PROGRAMS
========================================================== Appendix II



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)




(See figure in printed edition.)Appendix IV
HEAD START'S SURVEY
========================================================== Appendix II



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
=========================================================== Appendix V

GAO CONTACTS

D.  Catherine Baltzell, Assistant Director, (202) 512-8001
Sherri Doughty, Project Manager, (202) 512-7273

STAFF ACKNOWLEDGMENTS

In addition to those named above, the following individuals made
important contributions to this report:  Deborah Edwards developed
the survey, performed the statistical analyses, and co-wrote the
report; Donnesha Correll co-wrote the report and managed survey
operations; Wayne Dow performed the statistical analyses; Liz
Williams edited the report; and Ann McDermott created the report
graphics. 


*** End of document. ***