p23Y2

unbd
stutee
General
Accounthg
ofrice
GAO
Wulthgton,

Beall,

1

D.C. 20648

mhlrtlon

c

and Ellmur services Dmdorl

3 -257466

August 5, 1994
The Honorable
United States
Dear Senator

Harris Wofford
Senate

k

Wofford:

1

The Congress has been considering
legislation,
as part of a
crime reduction
package, that would prohibit
-prison inmates
from receiving
federal funding for education,
such as Pell
grants.
You requested that we (1) identify
the number of
inmates receiving
Pell grants,
(2) describe the effect
on
grants for other needy students,
and (3) measure the impact
of education on recidivism,
or relapse into criminal
.
behavior.
As agreed with your staff,
this letter
conveys
the information
you requested.

I
[
i
I

1

1

Backoround
At $6 billion
in 1993-94, the Pell grant program is the
largest
federal program providing
grants to help students
from low-income families
finance their undergraduate
postsecondary
education.
First authorized
in 1972, the
grants are fully
funded by the federal government.
They
are awarded on the basis of need, as determined by the
difference
between the student's
financial
resources and
the cost to him or her to attend school, including
tuition
transportation,
and fees, room and board, booka, supplies,
miscellaneous
expenses, and, In some cases, child or
dependent care and disability-related
expenses.
The
maximum award appropriated
for award year 1993-94 was
$2,300.'

e

In addition
to demonstrating
need, students must be
enrolled
in an undergraduate
course of study, and must meet
numerous other eligibility
requirements,
including
(1) a
hfgh school degree, (2) a recognized equivalent,
or (3)
have an ability
to benefit
from the education.
The Higher

I

'An award year is a 12 month period of time from July 1 of
one year to June 30 of the next.
Award year 1993-94 is the
period July 1, 1993, through June 30, 1994.

I
6
c

GAO/HEHS-94-224R Pell

Grants

for

Prison

Inmates

1
I

r

I
/

B-257466
Education Amendments of 1992 removed any requirement
for a
minimum number of courses or any limit
on the number of
years a student may take to complete an undergraduate
degree.
These amendments also prohibited
'incarcerated
students,
who are serving a death sentence or life sentence
without possibility
of parole,
from receiving
grants and
limited
the grants for other incarcerated
students to
tuition,
fees, books, and supplies.
On the government's behalf,
schools approved by the
Department of Education determine students'
eligibility
and
Approved schools fall into
process and award the grants.
(1) public or private
non-profit
one of three categories:
(2) public or private
nonschools of higher education,
profit
postsecondary vocational
schools, and (3) private
for-profit
proprietary
scho01s.~ All eligible
schools must
have legal authorization
to operate within a state,
accreditation
by a nationally
recognized accrediting
agency
or approval by state agencies endorsed by the Secretary of
Education,
and admit as regular students only those with
or
(1) a high school diploma, (2) a recognized equivalent,
(3) be beyond the state age of compulsory school
According to the 1992 amendments, a school
attendance.
with more than 25 percent of its students incarcerated
is
not eligible
to participate
in federal student financial
Pell grants.3
The amendments also
aid programs, including
limit
eligibility
of correspondence
programs because a
school is generally
eligible
only if (1) less than 50
percent of its students are enrolled
in correspondence
courses, and (2) less than 50 percent of its courses are
taught through correspondence.
Scoue and Methodolocv
To obtain the information
you requested,
we interviewed
program officials
at the Departments of Education and
In addition,
we
Justice,
as well as academic experts.
examined Pell grant eligibility
criteria
and procedures,
analyzed preliminary
1993-94 data on Pell grants awarded to

2We use "schools"
instead of t'institutions,*t
in higher education legislation,
to refer
providing
the education in order to avoid
prison institutions.

as referred
to
to the entities
confusion with

'The Secretary may waive this prohibition
for a nonprofit
institution
that offers a two-year associate degree and/or
a four-year
baccalaureate
degree.
2

GAO/HEHS-94-224R

Pell

Grants

for

Prison

Inmates

B-257466

incarcerated
recidivism.

students,

and reviewed

the literature

on

The Department of Education first
began collecting
data
during the 1993-94 award year on whether or not Pell
The Department, although
recipients
were incarcerated.
still
correcting
the data , provided us with preliminary
data on incarcerated
students as of April 4, 1994.
According to Department officials,
these data are very
close to the final totals
because most of the awards had
However, the accuracy of the
been made as of this date.
data is limited
by its preliminary
nature and Education's
Department officials
limited
verification
procedures.
estimate that the data will not be complete until
approximately
one year after the end of the award year.
The verification
procedures are limited
because schools
The Department only attempts to verify
the
self-report.
schools'
reports
if there is an obvious anomaly, such as an
Ohio school that reported more incarcerated
students than
the total
for any other state.
Particiuation
of Inmates
Small Part Of Pell Proaram
Pell
Inmate participation
is a small part of the total
Only
grant program, according to our data analysis.
23,000, of the approximately
4 million
Pell recipients
for
This represents
the 1993-94 award year,
were incarcerated.
less than 1 percent,
that is, 1 out of every 500 Pell
Inmates received $35 million
of the $6 billion
recipients.
This represents
less
awarded in Pell grants in 1993-94.
than 1 percent,
that is, 6 cents out of every 10 Pell
In addition,
the average award for
program dollars.
incarcerated
students was the same as for non-incarcerated
students,
$1,500 out of the $2,300 dollar
maximum award.'
Finally,
only a small percentage of all inmates receive
This is because the 23,000 inmates represent
Pell grants.
just 2 out of every 100 federal or state inmates.
Incarcerated
students receiving
Pell grants are
concentrated
in a few states and schools, as shown in
figure
1. Almost three-fourths
of inmate Pell recipients
Inmates in non-federal
prisons in
are in nine states.
'The data for non-incarcerated
students is for the 1992-93
However, we have no
award year, the most recent available.
reason to believe that the 1993-94 average award would be
significantly
different,
particularly
as the maximum
appropriated
award was the same.
3

GAO/HEHS-94-224R

Pell

Grants

for

Prison

Inmates

B-257466
seven states may not receive Pell grants because of the
states'
failure
to comply with a 1992 amendment, which
requires
that states demonstrate that Pell-awards
will be
state funding.
used to supplement, rather than supplant,
The distinction
between supplementing and supplanting
is
determined by requiring
the state to maintain the fiscal
year 1988 funding level of postsecondary education
See enclosure I for
assistance
to incarcerated
students.5
the number of incarcerated
Pell recipients
in each state.
Fiuure 1: Geoaraohic Distribution
Incarcerated
Students Receivina

of
Pell Grants

'The Department of Education denied California,
Michigan,
Nebraska,and New Jersey certification
that they met the
funding requirements,
Arizona and Florida did not apply,
and Missouri's
initial
certification
was withdrawn on
February 1, 1994 after they submitted corrected
funding
data.
4

GAO/HEHS-94-224R Pell

Grants

for

Prison

Inmates

B-257466

Almost one-third
of inmates receiving
Pell grants are at 11
Less than 3 percent (269 of 9,468) of schools
schools,
eligible
to award Pell grants report inmates as recipients.
See enclosure II for a list of the schools with the highest
As shown in table
1,
number of incarcerated
students.
incarcerated
Pell recipients
are most likely
to enroll
in
programs at public post-secondary
schools of at least 2
years, but less than 4, or g ro g rams at private non-profit
schools of 4 or more years.
Table 1: Percentaue of Inmate Pell
Tvpe of School and Proaram Lenath

Less than
2 years
At least 2
years, but :
less than
4
4 years or
more
Total

Grants

II

bv

ReciDients

2

0

8

10

39

3

1

43

12

35

0

47

53

I

38

to Incarcerated
Students
Grants to Other Needv

Do Not Affect

9

I

I

100

Students

According to Department officials,
grants to inmates do not
affect
the eligibility
or size of grants to other students.
If incarcerated
students received no Pell grants, no
student currently
denied a Pell award would have received
The
one and no award amount would been increased.
Department operates the Pell program as an entitlement
6The type of school categories
in the database
differ
from
those in the legislation
in that they are defined solely by
financial
control rather than the distinction
between
schools of higher education and vocational
schools.
5

GAO/HEHS-94-224R

Pell

Grants

for

Prison

Inmates

B-257466
program in that every eligible
student demonstrating
a
need, as defined by the program, is awarded a grant, up to
a maximum award. To meet the needs of current students,
This makes
the program borrows from future appropriations.
it unlikely
that the $35 million
that was awarded to
incarcerated
students could have been used to increase the
According to Department
award for other Pell recipients.
officials,
the money gained by reducing current
expenditures
by $35 million,
less than 1 percent, would
have been used to reduce the amount borrowed from future
In addition,
Department officials
years' appropriations.
if the money had been distributed
among other
estimated,
individual
awards would have increased by about
recipients,
$3, at most.
Measurino the Imnact
of Education on Recidivism
Has Resulted in Conflictina

Findinas

Many studies have attempted to isolate
the impact of
education from the many other factors affecting
recidivism.
These studies have resulted
in conflicting
findings.
Differences
in findings
may be partly explained by the
significant
methodological
challenges such studies face,
First,
both in terms of design and availability
of data,
design is complicated by the varied,
complex and interThe sociology
of
related
factors
influencing
recidivism.
criminology
suggests that manyffactors
are difficult
to
For example, a factor such as
either define or measure.
the prisoner's
level of community support, which could
include marital
relationship,
may be significant
in
marital
However, a positive
reducing recidivism,
relationship
may be a measure of community support for
some,
but an abusive marital
relationship
may increase the
probability
of recidivism
for others,
Second, design is complicated
by the process of selfBecause prisoners
selection
for education programs.
generally
volunteer
to participate
in education programs,
different
from those who do not
they are, by definition,
volunteer.
These differences
may affect recidivism
as
The
much, or more, than the education programs themselves.
best way to control
for these differences,
both known and
unknown, would be to conduct an experiment in which
prisoners
are randomly provided or not provided education
programs. The experience of individual
inmates from both
groups could then be followed,
over a period of years,
With or without a
after the inmates leave prison.
the impact would require
randomized study, determining
6

GAO/HEHS-94-224R

Pell

Grants

for

Prison

Inmates

B-257466
tracking
individual
prisoners
over time because of the
inter-related
factors
that influence
recidivism.
We have provided.your
staff with an annotated literature
search and copies of some recent studies on the impact of
Included was a study the Bureau
education on recidivism.
of Prisons released in 1992 on its Federal Post Release
we
Employment Project
(PREP). As we reported previously,'
found the study to be generally well-designed,
it
demonstrates
some of the difficulties
associated with
PREP found that inmates
studies of recidivism.
participating
in work or vocational
education were less
likely
to have their parole revoked (as a result
of
of their
committing
a crime or a technical
violation
parole)
than other inmates who had similar
background
characteristics
but did not participate
in work or
As we reported previously,
a
vocational
training
programs.
potentially
serious threat to the validity
of the study
includes
(1) the absence of random assignment of prisoners
and (2) differences
between those who did and those who did
not participate
in work and vocational
programs that might
influence
the success of the inmates after release from
prison.
If you have any questions or would like to discuss this
letter,
please call me at (202) 512-7014 or Charles Jeszeck
at (202) 512-7036.

Linda G. Morra, Director
Education and
Employment Issues

'GAO/GGD-93-33, Federal Prisons: Inmate and Staff Views on
Education and Work Trainina
Proarams, January 19, 1993
7

GAO/HEHS-94-224R

Pell

Grants

for

Prison

Inmates

I

ENCLOSUREI

ENCLOSUREI

State Totals of Incarcerated
Students Awarded Pell
for Award Year 1993/94, as of April 4, 1994
Ohio
New York
M issouri
Texas
Illinois
Massachusetts
Georgia
Alabama
Virginia
Pennsylvania
Indiana
Colorado
Tennessee
South Carolina
Maryland
Louisiana
M ississippi
Kentucky
Oregon
Oklahoma
Nevada
Kansas
W isconsin
Arkansas
M innesota
New Jersey
California
M ichigan
South Dakota
West Virginia
Wyoming
New Hampshire
Delaware
North Dakota
Utah
Montana
Alaska
North Carolina
Washington
Arizona
Florida
Iowa
Total

8

GAO/HEHS-94-224R Pell

Grants

3,793
3,034
1,557
1,479
1,467
1,306
1,186
1,094
1,054
708
655
567
564
517
455
363
341
319
313
309
264
249
245
214
148
143
134
112
2
58
51
44
41
39
15
4
4
2
1
1
1
22,993

Grants

for

Prison

Inmates

ENCLOSUREII

ENCLOSUREII

Schools With High Numbers of Incarcerated
Grants for Award Year 93-94, as of April

Students
4, 1994

Receiving

Pell

Number of
Incarcerated
Students
1,094
818
729
716
708
585
580
572
538
444
414

Ohio University,Ohio
Park College, Missouri
J F Ingram State Vocational
School, Alabama
Ashland University,
Ohio
Atlantic
Union College, Massachusetts
Mount Wachusett Community College, Massachusetts
Microcomputer Technology Institute,
Texas
Wilmington College, Ohio
Brewter Parker College, Georgia
Genesee Community College, New York
Tennessee
Branell College - Nashville,

104774
9

GAO/HEHS-94-224R Pell

Grants

for

Prison

Inmates

!