This is the accessible text file for GAO report number GAO-13-567 
entitled 'VA Education Benefits: Student Characteristics and Outcomes 
Vary across Schools' which was released on July 25, 2013. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

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

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

July 2013: 

VA Education Benefits: 

Student Characteristics and Outcomes Vary across Schools: 

GAO-13-567: 

GAO Highlights: 

Highlights of GAO-13-567, a report to congressional requesters. 

Why GAO Did This Study: 

In fiscal year 2012, various VA education programs provided nearly $11 
billion in education payments to almost 1 million veterans. The 2008 
Post-9/11 GI Bill significantly increased education benefits for 
veterans and provided for separate payments for tuition and fees, for 
housing expenses, and for books. In fiscal year 2012, VA made about 
$8.5 billion in tuition, housing, and other payments under the Post-
9/11 GI Bill. GAO was asked to review data on VA’s education programs. 
Specifically, GAO examined: (1) the distribution of VA education 
payments among schools; (2) how student characteristics and outcomes 
at highly VA-funded schools compare to those of other VA-funded 
schools; and (3) how student characteristics and outcomes compare at 
highly VA-funded public, nonprofit, and for-profit schools. 

To address these topics, GAO collected fiscal year 2010 and 2011 data 
from VA (the most recent data available at the time of our study) and 
school year 2010-2011 and 2011-2012 data from the Department of 
Education. GAO used these data to compare VA payment amounts, as well 
as student characteristics and outcomes, at highly VA-funded schools 
and other schools. Veterans typically comprise a small proportion of a 
school’s total enrollment. GAO also conducted regression analysis to 
compare student outcomes at public, nonprofit and for-profit schools 
after controlling for other factors, such as school enrollment and 
tuition. 

What GAO Found: 

The majority of the Department of Veterans Affairs (VA) education 
payments were made to a small percentage of schools receiving VA 
funding in fiscal year 2011, primarily through the Post-9/11 GI Bill. 
About 5 percent of schools (654 schools) received more than $3.8 
billion in aggregate VA education payments used for tuition and fees 
in fiscal year 2011, over 60 percent of such funding. These 654 “
highly VA-funded schools” each received at least $2 million (and as 
much as $113 million) in Post-9/11 GI Bill tuition and fee payments 
from fiscal year 2010 through fiscal year 2011 and enrolled more total 
students on average than other schools. Almost half of tuition and fee 
payments for all VA education programs were used at public schools. 
However, the breakdown of Post-9/11 GI Bill payments differed 
somewhat, with for-profit and public schools receiving about the same 
proportion of Post-9/11 GI Bill tuition and fee payments. 

Figure: Fiscal Year 2011 Tuition and Fee Payments by Sector for All VA 
Education Programs and for the Post-9/11 GI Bill: 

[Refer to PDF for image: 2 pie-charts] 

All VA Education Programs including Post-9/11 GI Bill: 
Public schools: 45%; 
Nonprofit schools: 21%; 
For profit schools: 34%. 

Post-9/11 GI Bill: 
Public schools: 38%; 
Nonprofit schools: 25%; 
For profit schools: 37%. 

Source: GAO analysis of VA data. 

Note: For VA education programs that pay veterans a fixed monthly 
amount, payments include the total payments made to veterans because 
it was not possible to determine what portion was used for tuition 
payments to schools. 

[End of figure] 

Highly VA-funded schools generally had more positive outcomes than 
other VA-funded schools. Compared to other schools, highly VA-funded 
schools generally had higher retention rates (percentage of students 
returning to the same school from 1 year to the next) and graduation 
rates. Because data on student veterans are limited, the student 
characteristics and outcomes in GAO’s analyses are for school 
populations as a whole. Student outcomes were generally similar when 
GAO tested more narrow definitions of highly VA-funded schools (at 
least $5 million and $10 million in Post-9/11 GI Bill funding). 

Among highly VA-funded schools, student characteristics and outcomes 
differed at public, nonprofit, and for-profit schools. For instance, 
highly VA-funded for-profit schools enrolled a higher percentage of 
low-income and minority students than public or nonprofit schools. 
After controlling for differences in school and student 
characteristics, for-profit schools had lower retention rates compared 
to public and nonprofit schools. However, for-profit schools had 
graduation rates that were higher than public schools and similar to 
nonprofit schools. 

What GAO Recommends: 

GAO makes no recommendations in this report. 

View [hyperlink, http://www.gao.gov/products/GAO-13-567]. For more 
information, contact Melissa Emrey-Arras at (617) 788-0534 or 
EmreyArrasM@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

A Small Percentage of Schools Received a Majority of VA Payments, 
Primarily through the Post-9/11 GI Bill: 

Highly VA-Funded Schools Enrolled a Lower Percentage of Low-Income 
Students and Had More Positive Outcomes than Other Schools: 

Among Highly VA-Funded Schools, Student Characteristics and Outcomes 
Varied at Public, Nonprofit, and For-Profit Schools: 

Agency Comments and Our Evaluation: 

Appendix I: Scope and Methodology: 

Appendix II: For-Profit Schools and the 90-10 Requirement: 

Appendix III: School Characteristics at Highly VA-funded Schools 
Compared to Other VA-Funded Schools: 

Appendix IV: Comparisons of School Characteristics across Public, 
Nonprofit, and For-Profit Schools: 

Appendix V: Comments from the Department of Veterans Affairs: 

Appendix VI: Comments from the Department of Education: 

Appendix VII: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: VA Education Programs Available to Eligible Veterans: 

Table 2: Highly VA-Funded Schools at Different Post-9/11 Payment 
Levels: 

Table 3: Fiscal Year 2011 Post-9/11 GI Bill Funding, Numbers of 
Veterans, and Funding per Veteran: 

Figures: 

Figure 1: Beneficiaries of Selected VA Education Benefit Programs, 
Fiscal Years 2007-2013: 

Figure 2: Fiscal Year 2011 VA Education Payments by Program: 

Figure 3: Fiscal Year 2011 Tuition and Fee Payments by Sector for All 
VA Education Programs and for Post-9/11 GI Bill: 

Figure 4: Fiscal Year 2011 Post-9/11 GI Bill Payments, Numbers of 
Veterans, Payments per Veteran, and School Year 2011-2012 Average 
Tuition, at Highly VA-Funded Schools, by Sector: 

Figure 5: Percentage of Low-Income, Minority, Part-Time, and Older 
Students at Highly VA-Funded and Other Schools, School Year 2011-12: 

Figure 6: School Year 2011-12 Retention, Overall Graduation, 4-and 2-
Year Graduation, and 3-Year Student Loan Default Rates for Highly VA-
Funded Schools and Other Schools: 

Figure 7: Comparisons of Selected Student Characteristics and Outcomes 
for Different Definitions of Highly VA-Funded Schools: 

Figure 8: Percentage of Different Student Populations at Highly VA-
Funded Public, Nonprofit, and For-Profit Schools, School Year 2011-12: 

Figure 9: 90/10 Rates of Highly VA-Funded and Other VA-Funded Schools, 
School Year 2011-12: 

Figure 10: Number of Students at Highly VA-Funded and Other VA-Funded 
Schools, School Year 2010-2011: 

Figure 11: Comparison of Participation in VA's Yellow Ribbon Program 
and Offering Distance Education at Highly VA-Funded and Other VA-
Funded Schools, School Year 2011-12: 

Figure 12: School Year 2011-12 Highest Degree Offering at Highly VA-
Funded and Other VA-Funded Schools: 

Figure 13: Retention, Overall Graduation, 4-and 2-Year Graduation, and 
3-Year Student Loan Default Rates across Highly VA-Funded Schools, by 
Sector, School Year 2011-12, before Controlling for Differences in 
Student and School Characteristics: 

Figure 14: School Year 2010-11 Enrollment Levels at Highly VA-Funded 
Schools, by Sector: 

Figure 15: School Year 2011-12 Comparison of VA's Yellow Ribbon 
Program Participation and Distance Education at Highly VA-Funded 
Schools, by Sector: 

Figure 16: School Year 2011-12 Highest Degree Offering at Highly VA-
Funded Schools, by Sector: 

[End of figure] 

GAO:
United States Government Accountability Office: 
441 G St. N.W. 
Washington, DC 20548: 

July 2013: 

The Honorable Patty Murray Chairman: 
Budget Committee: 
United States Senate: 

The Honorable Bernard Sanders: 
Chairman: 
Committee on Veterans' Affairs: 
United States Senate: 

The Honorable John D. Rockefeller IV: 
United States Senate: 

The Department of Veterans Affairs (VA) provided over $11 billion in 
postsecondary education benefits through multiple programs to almost 1 
million veterans and their beneficiaries in fiscal year 2012, up from 
about $3 billion in 2008. Education benefits for veterans 
significantly increased with the passage of the Post-9/11 Veterans 
Educational Assistance Act of 2008 (Post-9/11 GI Bill).[Footnote 1] 
Other veterans are still receiving benefits for prior service through 
older programs like the Montgomery GI Bill. Participation in the Post-
9/11 GI Bill program is expected to increase by over 10 percent from 
2011 to 2013 in part due to veterans returning from service in Iraq 
and Afghanistan. Given the growth of the Post-9/11 GI Bill program, VA 
education programs and schools that receive significant VA payments 
have been subject to increased scrutiny. At the same time, little 
systemic information is available on veterans' education outcomes. 
[Footnote 2] In light of these trends, you asked us to review data on 
VA's education programs. For this report, we examined: 

1. What is the distribution of VA education payments among schools? 

2. How do student characteristics and outcomes at highly VA-funded 
schools compare to those of other VA-funded schools? 

3. How do student characteristics and outcomes compare at public, 
nonprofit, and for-profit schools that are highly VA-funded? 

To address these questions, we reviewed relevant federal laws and 
regulations and interviewed VA and Department of Education (Education) 
officials. For question one, we analyzed the most recent data 
available from VA at the time of our study--fiscal years 2010 and 2011 
data on enrollment and payment amounts by school for all students 
receiving VA education benefits. Specifically, we compared payment 
amounts across all VA education programs and types of schools. We also 
identified "highly VA-funded schools," specifically, those schools 
that received at least $2 million in payments from the Post-9/11 GI 
Bill program from fiscal year 2010 through fiscal year 2011.[Footnote 
3] 

To compare student characteristics and outcomes for questions two and 
three, we conducted descriptive statistical analyses using Post-9/11 
GI Bill payment data and school years 2010-2011 and 2011-2012 data on 
school characteristics reported by schools through Education's 
Integrated Postsecondary Education Data System (IPEDS). Because data 
on student veterans are limited, the student characteristics and 
outcomes in our analyses are for school populations as a whole. 
[Footnote 4] We compared highly VA-funded schools with other VA-funded 
schools and, within the group of highly VA-funded schools, we compared 
schools across the public, nonprofit, and for-profit sectors. We also 
conducted sensitivity analyses to test differences in student 
characteristics and outcomes using narrower definitions of highly VA-
funded schools ($5 million, $10 million, and $15 million in total Post-
9/11 GI Bill payments from fiscal years 2010 through 2011). In 
addition, for question three, we conducted multivariate regression 
analyses at the $2 million level to compare student outcomes in 
different sectors while statistically controlling for differences in 
student characteristics that could affect those outcomes.[Footnote 5] 
With this analysis, for example, we could determine whether schools in 
one sector were significantly more likely to have higher graduation 
rates than schools in another sector, after controlling for other 
factors. We assessed the reliability of the VA and Education data by 
interviewing relevant officials about how these data were compiled and 
reviewing documentation about the specific data systems. As a result 
of this assessment, we concluded that the VA and Education data were 
sufficiently reliable for our reporting purposes. 

The student outcome measures we used have certain limitations. For 
instance, the IPEDS graduation rates include only full-time, first-
time students and exclude students who attend part-time or transfer to 
another school. For schools with a bachelor's degree as the highest 
degree offered, the IPEDS retention rates--the percentage of students 
returning to the same school from one year to the next--include only 
students seeking a bachelor's degree. See appendix I for a more 
detailed discussion of our scope and methodologies. 

We conducted this performance audit from February 2012 through July 
2013 in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

Background: 

VA has been providing veterans educational assistance benefits since 
1944.[Footnote 6] These benefits have been put in place over time in 
order to compensate for compulsory service, encourage voluntary 
service, avoid unemployment, provide equitable benefits to all who 
served, or promote military retention. Table 1 lists the current major 
benefit programs available to eligible individuals who pursue 
education or training. Eligibility requirements vary among programs 
and are based on several factors, including when and for how long a 
veteran served in the military. In some cases, a veteran can be 
eligible for multiple education programs and may have to choose which 
program to use. 

Table 1: VA Education Programs Available to Eligible Veterans: 

Program: Post-9/11 Veterans Educational Assistance Program (Post-9/11 
GI Bill) 38 U.S.C. §§ 3301-3334; 
Description: Provides benefits to veterans who served on active duty 
for at least 90 days after September 10, 2001. The program provides 
individuals who served on active duty for 36 months with the full cost 
of attendance at a public school and up to a maximum amount at 
nonprofit and for-profit schools ($18,077 in academic year 2012-13). 
Some veterans may also receive additional payments for housing 
expenses and books. Under certain circumstances, veterans can transfer 
their Post-9/11 GI Bill benefits to spouses and children; 
Participants (FY 2012): 646,302; 
Payments (FY 2012): $8,453 million (includes both tuition and housing). 

Program: Montgomery GI Bill-Active Duty; 
38 U.S.C. §§ 3001-3036; 
Description: Provides a fixed monthly allowance primarily to veterans 
who enter active duty after June 30, 1985. In fiscal year 2013, the 
benefit is $1,564 per month for full-time study; 
Participants (FY 2012): 118,549; 
Payments (FY 2012): $932 million. 

Program: Montgomery GI Bill-Selected Reserve; 
10 U.S.C. §§ 16131-16137; 
Description: Provides a monthly benefit payment for Reservists, 
including the National Guard, who agree to serve for 6 years. In 
fiscal year 2013, the benefit is $356 per month for full-time study; 
Participants (FY 2012): 60,393; 
Payments (FY 2012): $157 million. 

Program: Survivors' and Dependents' Educational Assistance Program; 
38 U.S.C. §§ 3500-3566; 
Description: Provides education and training opportunities to eligible 
dependents of certain veterans who were disabled or died during active 
duty service. In fiscal year 2013, beneficiaries could receive up to 
$987 per month based on their enrollment status; 
Participants (FY 2012): 87,707; 
Payments (FY 2012): $455 million. 

Program: Reserve Educational Assistance Program; 
10 U.S.C. §§ 16131-16166; 
Description: Provides benefits to Reservists with at least 90 days 
consecutive active-duty service after September 10, 2001. In fiscal 
2013, the benefit is $1,251 per month for full-time study, with at 
least 2 years of consecutive active-duty service; 
Participants (FY 2012): 19,774; 
Payments (FY 2012): $76 million. 

Program: Vocational Rehabilitation and Employment Program (VR&E); 
38 U.S.C. §§ 3100-3121; 
Description: Provides services to veterans with service-connected 
disabilities. These services help servicemembers and veterans achieve 
their employment goals and can include education benefits and 
counseling; 
Participants (FY 2012): 61,755; 
Payments (FY 2012): $791 million (including subsistence payments). 

Program: Total; 
Participants (FY 2012): 994,480[A]; 
Payments (FY 2012): $10.9 billion. 

Source: GAO analysis of data provided by VA and relevant VA program 
documents. 

Notes: Data for the VR&E program are for those veterans who received 
retraining as part of their rehabilitation services. VR&E provides for 
veterans' tuition, fees, books, and supplies. Veterans enrolled in 
training service are eligible to receive a subsistence allowance to 
assist them with living expenses. 

[A] Participants would be counted twice in this total if they received 
benefits under more than one program. 

FY = fiscal year. 

[End of table] 

The education benefits available to veterans and their dependents were 
significantly expanded with the enactment of the Post-9/11 GI Bill. 
Since the Post-9/11 GI Bill became effective August 1, 2009, the 
number of veterans and beneficiaries receiving VA education benefits 
has increased by almost two-thirds (see figure 1). Participation in VA 
education benefit programs is expected to increase as the number of 
Post-9/11 veterans grows to over 5 million by 2020. 

Figure 1: Beneficiaries of Selected VA Education Benefit Programs, 
Fiscal Years 2007-2013: 

[Refer to PDF for image: stacked vertical bar graph] 

Fiscal year: 2007; 
Other VA education programs: 100%; 
Post-9/11 GI Bill: 0. 

Fiscal year: 2008; 
Other VA education programs: 100%; 
Post-9/11 GI Bill: 0. 

Fiscal year: 2009; 
Other VA education programs: 94%; 
Post-9/11 GI Bill: 6%. 

Fiscal year: 2010; 
Other VA education programs: 54%; 
Post-9/11 GI Bill: 46%. 

Fiscal year: 2011; 
Other VA education programs: 40%; 
Post-9/11 GI Bill: 60%. 

Fiscal year: 2012; 
Other VA education programs: 32%; 
Post-9/11 GI Bill: 68%. 

Fiscal year: 2013 (estimated); 
Other VA education programs: 35%; 
Post-9/11 GI Bill: 65%. 

Source: GAO Analysis of VA’s Fiscal Year 2011 Annual Benefits Report, 
interim data provided by VA for publication in an updated version of 
its fiscal year 2012 Annual Benefits Report (which was not available 
when GAO issued its report} and fiscal year 2013 Congressional Budget 
Submission. 

Note: Other VA education programs include the Montgomery GI Bill-
Active Duty, the Montgomery GI Bill-Selected Reserve, Survivors' and 
Dependents' Educational Assistance Program, and the Reserve 
Educational Assistance Program, but not the VR&E program. 

[End of figure] 

Administration of the Post-9/11 GI Bill: 

The Post-9/11 GI Bill provides a new way to deliver VA education 
benefits. VA education programs established prior to the Post-9/11 GI 
Bill typically provided a fixed monthly payment to a veteran or other 
eligible beneficiary to be used for tuition, books, or living expenses 
while in school. The Post-9/11 GI Bill, however, provides for tuition 
and fee payments to be made directly to the school and authorizes a 
separate monthly housing allowance to the student, as well as a book 
allowance.[Footnote 7] 

Other Sources of Student Aid: 

To help cover the costs of their postsecondary education, veterans may 
also be eligible for grants and loans available from federal student 
aid programs administered by Education, such as Pell Grants and 
William D. Ford Federal Direct Loans.[Footnote 8] Veterans may use 
student loans to pay for living expenses not covered by VA education 
payments. VA education payments are not considered when calculating 
eligibility for federal student aid and do not affect the amount of 
aid a veteran can receive from Education.[Footnote 9] Student veterans 
may also be eligible for state and institutional aid (scholarships 
from state governments or schools, for example). 

To maintain their eligibility for federal student aid programs 
administered by Education, for-profit schools must obtain at least 10 
percent of their total revenues from sources other than federal 
student aid programs (those authorized by Title IV of the Higher 
Education Act of 1965)--this is known as the 90/10 rule.[Footnote 10] 
Because VA education payments are not authorized by this law, they are 
counted as nonfederal student aid and thus toward the 10 percent 
requirement; as a result, enrolling veterans helps for-profit schools 
meet the 90/10 rule See appendix II for more information on the 90-10 
rule. 

Types of Schools Offering Postsecondary Education: 

A wide variety of schools offer different types of postsecondary 
education programs. Schools also have different campus structures, 
including a single campus, multiple campuses, or no physical campus if 
a school offers only online programs. Correspondingly, veterans may 
use VA education benefits at different types of schools to pursue a 
variety of programs: 

* 4-year and above schools: Colleges and universities that typically 
offer bachelor's and higher level degrees in programs that cover 4 or 
more years. 

* 2-year schools: Community colleges and other schools that typically 
offer associate's degrees, but can also offer certificate programs. 

* Less than 2-year schools: Vocational and technical schools can 
receive VA education payments that offer certificate programs, but 
typically not degrees, in programs lasting less than 2 years. 

The following types of schools may offer programs of any length noted 
above: 

* Public schools: operated and funded by state or local governments, 

* Nonprofit schools: owned and operated by nonprofit organizations 
whose net earnings do not benefit any shareholder or individual, and: 

* For-profit schools: privately owned and net earnings can benefit a 
shareholder or individual. 

While many colleges and universities are eligible to receive payments 
from both VA and Education, other types of entities, while not 
eligible for federal student aid, are eligible for VA education 
funding. For example, providers of apprenticeship opportunities, on-
the job training programs, and entrepreneurship courses may be 
eligible to receive VA education funding. Consequently, eligible 
veterans can use VA education funding for programs that may not be 
eligible for federal student aid administered by Education. However, 
there is significant overlap among the schools receiving funding from 
VA or Education. 

A Small Percentage of Schools Received a Majority of VA Payments, 
Primarily through the Post-9/11 GI Bill: 

In fiscal year 2011, about 5 percent of VA-funded schools received 
over 60 percent of VA education payments devoted to tuition and fees. 
[Footnote 11] That year, 654 "highly VA-funded schools" received more 
than $3.8 billion in total VA education payments.[Footnote 12] The 
highly VA-funded schools--which each received at least $2 million (and 
as much as $113 million) in total payments from the Post-9/11 GI bill 
from fiscal years 2010 through 2011--consisted of 293 public, 147 
nonprofit, and 214 for-profit schools. Other VA-funded schools 
received about $2.4 billion in total VA education payments in fiscal 
year 2011.[Footnote 13] Highly VA-funded schools enrolled a greater 
number of students on average compared to other schools. In school 
year 2010-11, for example, 34 percent of highly VA-funded schools had 
more than 20,000 students, while only 2 percent of other VA-funded 
schools did. 

The majority of VA's education payments were distributed through the 
Post-9/11 GI Bill program, while the Montgomery GI Bill-Active Duty 
provided the second highest amount of payments (see figure 2). Use of 
the Post-9/11 GI Bill program increased more than any other VA 
education program from fiscal years 2010 through 2011. In fiscal year 
2011, Post-9/11 GI Bill payments and number of student veterans served 
increased by over 50 percent. In contrast, enrollments and total 
payments for the Montgomery GI Bill-Active Duty decreased by 25 
percent from fiscal years 2010 through 2011. VA officials said that 
veterans with a choice between the Post-9/11 GI Bill and other 
programs may be more apt to choose the Post-9/11 GI Bill because of 
its more generous benefits. 

Figure 2: Fiscal Year 2011 VA Education Payments by Program: 

[Refer to PDF for image: pie-chart] 

Post-9/11 GI Bill: $3,593,661,426; 56%; 
Montgomery GI Bill-Active Duty: $1,485,625,036; 23%; 
Vocational Rehabilitation and Employment: $633,139,947; 10%; 
Survivors’ and Dependents’ Educational Assistance Program: 
$368,573,334; 6%; 
Montgomery GI Bill-Selected Reserve: $161,825,738; 3%; 
Reserve Educational Assistance Program: $113,700,858; 2%. 

Source: GAO analysis of VA data. 

Note: The Post-9/11 GI Bill and Vocational Rehabilitation and 
Employment payment amounts listed here include tuition and fee 
payments made to the school, but not additional housing allowance 
payments. 

[End of figure] 

When we looked at all VA programs, we found that veterans used almost 
half of total nationwide VA education payments devoted to tuition and 
fees at public schools, about one-fifth at nonprofit schools, and 
about a third at for-profit schools in fiscal year 2011 (see figure 
3). However, the breakdown of Post-9/11 GI Bill funding was a little 
different--for-profit schools collectively received about the same 
percentage of tuition and fee payments from the Post-9/11 GI Bill 
program as public schools. 

Figure 3: Fiscal Year 2011 Tuition and Fee Payments by Sector for All 
VA Education Programs and for Post-9/11 GI Bill: 

[Refer to PDF for image: 2 pie-charts] 

All VA Education Programs including Post-9/11 GI Bill: 
Public schools: 45%; 
Nonprofit schools: 21%; 
For profit schools: 34%. 

Post-9/11 GI Bill: 
Public schools: 38%; 
Nonprofit schools: 25%; 
For profit schools: 37%. 

Source: GAO analysis of VA data. 

Notes: Payment amounts for the Post-9/11 GI Bill and the Vocational 
Rehabilitation and Employment programs include only tuition payments 
made directly to schools, but not additional housing allowance or 
other payments. For all other VA education programs, because it is not 
possible to determine what portion of payments was used for tuition 
payments to schools, payments include the total payments made to 
veterans, which may exceed actual tuition costs in some cases. 

[End of figure] 

When looking specifically at highly VA-funded schools (those receiving 
$2 million or more in Post-9/11 GI Bill funds from fiscal years 2010 
through 2011), we found that for-profit schools received the most 
total Post-9/11 GI Bill payments for tuition and fees. Highly VA-
funded for-profit schools also had the highest Post-9/11 GI Bill 
payments per veteran, while public schools enrolled the most veterans 
and had the lowest Post-9/11 GI Bill payments per veteran (see figure 
4).[Footnote 14] 

Figure 4: Fiscal Year 2011 Post-9/11 GI Bill Payments, Numbers of 
Veterans, Payments per Veteran, and School Year 2011-2012 Average 
Tuition, at Highly VA-Funded Schools, by Sector: 

[Refer to PDF for image: 4 horizontal bar graphs] 

FY 2011 Post-9/11 GI Bill payment amount: 
Public: $871,385,605; 
Nonprofit: $560,513,009; 
For-profit: $1,004,571,372; 
All: $2,436,469,986. 

FY 2011 Number of veterans receiving benefits: 
Public: 174,137; 
Nonprofit: 60,619; 
For-profit: 98,983; 
All: 333,739. 

FY 2011 Average payment amount per veteran: 
Public: $5,004; 
Nonprofit: $9,246; 
For-profit: $10,149. 

School Year 2011-2012 Tuition: 
Public: $15,804. 
Nonprofit: $$24,416; 
For-profit: $19,160. 

Source: GAO analysis of VA and IPEDS data on highly VA-funded schools. 

Notes: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from FY 2010 through FY 
2011. 

When we tested narrower definitions of highly VA-funded schools, we 
saw similar results in the percentage of veterans at schools receiving 
at least $5 million in Post-9/11 payments from FY 2010 through FY 
2011. However, at the $10 million payment level, for-profit schools 
did not enroll the highest proportion of student veterans, in contrast 
to the $2 million level. Rather, nonprofit schools had the highest 
proportion of student veterans at the $10 million level, while public 
schools had the lowest proportion. The number of schools in the $15 
million group was too small to conduct a reliable sector analysis. 

FY = fiscal year. 

[End of figure] 

Highly VA-Funded Schools Enrolled a Lower Percentage of Low-Income 
Students and Had More Positive Outcomes than Other Schools: 

Our analysis showed that the overall student population at highly VA-
funded schools had some different characteristics and generally had 
higher retention and graduation rates compared to students enrolled at 
other VA-funded schools.[Footnote 15] Highly VA-funded schools--those 
that received at least $2 million in Post-9/11 GI Bill payments from 
fiscal years 2010 through 2011--generally enrolled a lower percentage 
of low-income students in school year 2011-2012 than other VA-funded 
schools (see figure 5). They also enrolled proportionately fewer part-
time students, a slightly lower percentage of minority students, and a 
higher proportion of veterans than other schools. 

Figure 5: Percentage of Low-Income, Minority, Part-Time, and Older 
Students at Highly VA-Funded and Other Schools, School Year 2011-12: 

[Refer to PDF for image: vertical bar graph] 

Student characteristics: Low-income; 
Highly VA-funded schools: 39%; 
Other VA-funded schools: 56%. 

Student characteristics: Minority; 
Highly VA-funded schools: 38%; 
Other VA-funded schools: 41%. 

Student characteristics: Part-time; 
Highly VA-funded schools: 32%; 
Other VA-funded schools: 42%. 

Student characteristics: Over age 25; 
Highly VA-funded schools: 20%; 
Other VA-funded schools: 19%. 

Student characteristics: Veterans as a percentage of school enrollment; 
Highly VA-funded schools: 5%; 
Other VA-funded schools: 2%. 

Source: GAO analysis of VA and IPEDS data. 

Notes: We use the term "low-income" to describe students receiving 
federal Pell grants, a need-based federal student aid program for low-
income students. 

The information on veterans as a percentage of student enrollment is 
from school year 2010-2011. 

[End of figure] 

Highly VA-funded schools are those that received at least $2 million 
in total Post-9/11 GI Bill payments from fiscal year 2010 through 
fiscal year 2011. 

Highly VA-funded schools generally had better retention and overall 
graduation rates, with similar student loan default rates as other VA-
funded schools in school year 2011-2012 (see figure 6). When we looked 
specifically at the graduation rates of 4-year schools, highly VA-
funded schools generally had higher graduation rates. However, at 2-
year schools, highly VA-funded schools had slightly lower graduation 
rates than other schools.[Footnote 16] 

Figure 6: School Year 2011-12 Retention, Overall Graduation, 4-and 2-
Year Graduation, and 3-Year Student Loan Default Rates for Highly VA-
Funded Schools and Other Schools: 

[Refer to PDF for image: vertical bar graph] 

Retention rate: 
Highly VA-funded schools: 79%; 
Other VA-funded schools: 68%. 

Overall graduation rate: 
Highly VA-funded schools: 55%; 
Other VA-funded schools: 45%. 

4-year graduation rate: 
Highly VA-funded schools: 59%; 
Other VA-funded schools: 52%. 

2-year graduation rate: 
Highly VA-funded schools: 33%; 
Other VA-funded schools: 34%. 

3-year student loan default rate: 
Highly VA-funded schools: 22%; 
Other VA-funded schools: 22%. 

Source: GAO analysis of VA and IPEDS data. 

Notes: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

Overall graduation rates include rates for both 2-year and 4-year 
schools. 

Education calculates student loan default rates to include defaults 
that occur within 3 years after a student borrower was required to 
begin repaying loans. In some cases, a school with several related 
campuses may report a single default rate for all of the campuses 
together. In these cases, we assigned that single rate to each campus. 

IPEDS outcome data have certain limitations. For example, the 
graduation rate includes only first-time, full-time students, while 
part-time and transfer students are excluded from the calculation. 

[End of figure] 

Our results were generally similar when we tested narrower definitions 
of highly VA-funded schools. Specifically, when we looked at schools 
receiving at least $5 million and $10 million (rather than $2 million) 
in combined fiscal year 2010 and 2011 Post-9/11 GI Bill payments, 
student characteristics and outcomes were generally similar to the 
results at the $2 million level (see figure 7).[Footnote 17] Student 
characteristics and outcomes differed somewhat for a small group of 
schools receiving at least $15 million in Post-9/11 GI Bill payments. 
These 38 schools had a higher percentage of low-income students and 
lower overall graduation rates than other schools, in contrast to the 
results at the $2 million level. However, the results for these 38 
schools are sensitive to the inclusion/exclusion of specific schools. 
For example, the overall graduation rate is similar to the other 
levels when the largest school is excluded. 

Figure 7: Comparisons of Selected Student Characteristics and Outcomes 
for Different Definitions of Highly VA-Funded Schools: 

[Refer to PDF for image: illustrated table] 

How highly VA-funded schools compare to other VA-funded schools: 

Percent low-income students: 
Funding: 
$2 million: lower; 
$5 million: lower; 
$10 million: lower; 
$15 million: higher. 

Percent minority students: 
Funding: 
$2 million: lower; 
$5 million: tied; 
$10 million: higher; 
$15 million: higher. 

Percent part-time students: 
Funding: 
$2 million: lower; 
$5 million: lower; 
$10 million: lower; 
$15 million: lower. 

Overall graduation rate: 
Funding: 
$2 million: higher; 
$5 million: higher; 
$10 million: higher; 
$15 million: lower. 

Retention rate: 
Funding: 
$2 million: higher; 
$5 million: higher; 
$10 million: higher; 
$15 million: higher. 

Source: GAO analysis of FY 2010 and 2011 VA data and school year 2011-
2012 IPEDS data. 

Notes: The 2-year graduation rate differed at the $15 million level as 
well. In contrast to the results at the $2 million level, highly VA-
funded schools at the $15 million level had a higher 2-year graduation 
rate than other VA-funded schools. Likewise, the 4-year graduation 
rate differed at the $15 million level. In contrast to the results at 
the $2 million level, highly VA-funded schools at the $15 million 
level had a lower 4-year graduation rate than other VA-funded schools. 

We use the term "low-income" to describe students who received federal 
Pell grants, a need-based federal student aid program for low-income 
students. 

[End of figure] 

Among Highly VA-Funded Schools, Student Characteristics and Outcomes 
Varied at Public, Nonprofit, and For-Profit Schools: 

Within the highly VA-funded group of schools, for-profit schools had 
higher proportions of low-income and minority students, along with 
mixed outcomes in retention, graduation, and default rates, compared 
to public and nonprofit schools.[Footnote 18] Nonprofit schools, 
meanwhile, had the lowest percentage of low-income students and public 
schools had the lowest percentage of minority students (see figure 8). 
[Footnote 19] 

Figure 8: Percentage of Different Student Populations at Highly VA-
Funded Public, Nonprofit, and For-Profit Schools, School Year 2011-12: 

[Refer to PDF for image: vertical bar graph] 

Student characteristics by sector: Low-income; 
Public schools: 36%; 
Nonprofit schools: 30%; 
For-Profit schools: 74%. 

Student characteristics by sector: Over age 25; 
Public schools: 16%; 
Nonprofit schools: 23%; 
For-Profit schools: 39%. 

Student characteristics by sector: Minority; 
Public schools: 36%; 
Nonprofit schools: 37%; 
For-Profit schools: 51%. 

Student characteristics by sector: Part-time; 
Public schools: 32%; 
Nonprofit schools: 29%; 
For-Profit schools: 30%. 

Student characteristics by sector: Veterans as a percentage of school 
enrollment; 
Public schools: 4%; 
Nonprofit schools: 5%; 
For-Profit schools: 8%. 

Source: GAO analysis of VA and IPEDS data on highly VA-funded schools.  

Notes: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

The information on veterans as a percentage of student enrollment is 
from school year 2010-2011. 

We use the term "low-income" to describe students who received federal 
Pell grants, a need-based federal student aid program for low-income 
students. 

[End of figure] 

After controlling for differences in school and student 
characteristics using multiple regression analysis, we found that 
student outcomes varied across schools in different sectors.[Footnote 
20] 

* Retention Rates: Highly VA-funded for-profit schools had lower 
retention rates than public and nonprofit schools. The average for-
profit school had retention rates about 6 points lower than similar 
public schools. Public and nonprofit schools had similar retention 
rates.[Footnote 21] 

* Graduation Rates: Highly VA-funded for-profit and nonprofit schools 
had similar graduation rates, which were higher, on average, than 
graduation rates at public schools.[Footnote 22] Graduation rates at 
for-profit schools were about 6 percentage points higher than those of 
public schools.[Footnote 23] 

* Default Rates: Highly VA-funded for-profit schools had higher 
student loan default rates than public and nonprofit schools.[Footnote 
24] For-profit schools had default rates that were, on average, about 
6.5 percentage points higher than public schools' default rates. 
Public schools' default rates were, on average, slightly higher than 
nonprofit schools' rates.[Footnote 25] 

Agency Comments and Our Evaluation: 

We provided a draft copy of this report to VA and Education for review 
and comment. VA's comments are reproduced in appendix V, and 
Education's comments are reproduced in appendix VI. 

We are sending copies of this report to relevant congressional 
committees, the Secretaries of VA and Education, and other interested 
parties. In addition, this report will be available at no charge on 
GAO's Web site at [hyperlink, http://www.gao.gov]. 

If you or your staffs have any questions about this report, please 
contact me at (617) 788-0534 or EmreyArrasM@gao.gov. Contact points 
for our Offices of Congressional Relations and Public Affairs may be 
found on the last page of this report. GAO staff who made key 
contributions to this report are listed in appendix VII. 

Signed by: 

Melissa Emrey-Arras: 
Director, Education, Workforce, and Income Security Issues: 

[End of section] 

Appendix I: Scope and Methodology: 

To address our objectives, we reviewed relevant federal laws and 
regulations, interviewed VA and Education officials, and analyzed 
fiscal years 2010 and 2011 data from VA on enrollment and payment 
amounts by school for all students receiving benefits from one of VA's 
education programs. Specifically, we compared payment levels across VA 
education programs and types of schools and identified schools with 
high payment levels. We then conducted descriptive statistical 
analyses using VA payment data and school characteristic data reported 
by schools through Education's Integrated Postsecondary Education Data 
System (IPEDS) to compare the student outcomes and other 
characteristics of highly VA-funded schools with other schools 
receiving VA funding. In addition, we conducted multivariate 
statistical analyses to examine sector level differences in student 
outcomes at highly VA-funded schools. 

We conducted our review between February 2012 and July 2013 in 
accordance with generally accepted government auditing standards. 
Those standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe 
that the evidence obtained provides a reasonable basis for our 
findings based on our audit objectives. 

Data Sources: 

We used the following data sources for our analyses. 

VA Data: 

We obtained data from VA on enrollment and payment amounts by schools 
for students receiving benefits from the following VA education 
programs in fiscal years 2010 or 2011: 

* Post-9/11 Veterans Educational Assistance Program: 

* Montgomery GI Bill--Active Duty: 

* Montgomery GI Bill--Selected Reserve: 

* Survivors' and Dependents' Educational Assistance Program: 

* Reserve Educational Assistance Program: 

* Vocational Rehabilitation and Employment Program: 

We assessed the reliability of the VA payment and enrollment data by 
interviewing VA officials about how these data were compiled and 
reviewing documentation about VA's data systems. As a result of this 
assessment, we concluded that VA data were sufficiently reliable for 
our reporting purposes. 

IPEDS: 

IPEDS is a system of interrelated surveys conducted annually to gather 
information from every college, university, and technical and 
vocational institution that participate in federal student financial 
aid programs. The Higher Education Act of 1965[Footnote 26] requires 
institutions that participate in federal student aid programs to 
complete surveys conducted as part of IPEDS, which requires schools to 
report data on enrollments, program completions, graduation rates, 
finances, institutional prices, and student financial aid.[Footnote 
27] We used IPEDS data for the 2010-2011 and 2011-2012 school years to 
identify the characteristics of schools that received VA education 
payments, such as sector, enrollment, highest degree offering, student 
characteristics, graduation rate, and retention rate. We assessed the 
reliability of IPEDS data by discussing variables with IPEDS officials 
and reviewing past GAO uses of IPEDS data. As a result of this 
assessment, we concluded that IPEDS data were sufficiently reliable 
for our reporting purposes. 

Student Loan Default Rates: 

We obtained data on student loan default rates from the National 
Student Loan Data System (NSLDS), which is Education's central 
database for federal student aid grants and loans. We used the 
official fiscal year 2009 3-year cohort default rate. This is 
calculated for all borrowers of federal student loans who were 
required to begin repaying these loans in fiscal year 2009 and is 
calculated by determining how many of these borrowers defaulted on 
their loans in fiscal years 2009, 2010, or 2011. We assessed the 
reliability of the default rate data by interviewing Education 
officials about how the NSLDS data are compiled and used. As a result 
of this assessment, we concluded that default rate data were 
sufficiently reliable for our reporting purposes. 

90/10 Rates at For-Profit Schools: 

To maintain their eligibility for federal student aid programs (those 
authorized by Title IV of the Higher Education Act of 1965[Footnote 
28]), for-profit schools must obtain at least 10 percent of their 
total revenues from sources other than federal student aid programs--
this is known as the 90/10 rule.[Footnote 29] The percentage of a for-
profit school's total revenues obtained from federal student aid 
programs is known as the "90/10 rate." Our analysis of 90/10 rates is 
presented in appendix II. 

We obtained data on 90/10 rates at for-profit schools from Education's 
eZ-Audit data system, which collects information from the annual 
audited financial statements of all schools receiving federal student 
aid funds. We used 90/10 rates that were reported on schools' audited 
financial statements for school fiscal years that ended between July 
1, 2010 and June 30, 2011. We assessed the reliability of the 90/10 
rates by reviewing past GAO uses of 90/10 data and discussing the 
current data system with Education officials. As a result of this 
assessment, we concluded that the eZ-Audit data were sufficiently 
reliable for our reporting purposes. 

Data Analysis: 

Objective 1: We analyzed VA's data to compare payment and enrollment 
levels across VA's different education programs. Comparing payments 
received by schools across VA education programs is complicated by the 
different payment mechanisms used by VA education programs. Most VA 
education programs provide a fixed monthly payment to the student 
veteran who can use it to pay for tuition and other education 
expenses. For the analyses in this objective, we attributed total 
payments made to a veteran under these programs to the school; 
however, to the extent that the total payments exceed a school's 
tuition, this analysis may overestimate VA payment amounts received by 
schools with lower tuition. The Post-9/11 GI Bill and the Vocational 
Rehabilitation and Employment (VR&E) programs are different in that 
they make tuition and fee payments directly to the school a veteran is 
attending. Our calculations for those programs include tuition and fee 
payments made to the school, but not additional housing allowance or 
other payments, unless otherwise stated. The total number of schools 
in our analysis was increased by including payments for educational 
programs funded by the VR&E program. The VR&E program includes a 
variety of educational providers, such as computer training companies, 
law enforcement academies, and private tutoring services. We also 
analyzed VA's data to compare payment and enrollment levels at 
different types of schools (e.g., associate's, bachelor's, and non-
college degree programs and public, nonprofit, and for-profit schools). 

Defining Highly VA-Funded Schools: We also used the VA payment data to 
define highly VA-funded schools. We defined highly VA-funded schools 
using Post-9/11 GI Bill payments because it is the largest VA 
education program in terms of total student enrollment and tuition and 
fee payments and because tuition and fee payments made to the school 
can be identified. Specifically, we defined highly VA-funded schools 
as those that received at least $2 million in total tuition and fee 
payments from the Post-9/11 GI Bill for fiscal years 2010 through 
2011. We used this cut-off because this group of schools represented 
the majority of Post-9/11 GI Bill funding (60 percent) and included 
sufficient numbers of schools in different sectors to allow us to 
compare sector results using a multiple regression analysis. 

We conducted extensive sensitivity analyses to assess the consistency 
of our results for more narrow definitions of highly VA-funded 
schools. Specifically, we repeated our analyses using a cut-off of $5 
million, $10 million, and $15 million for highly VA-funded schools. 
Table 3 provides information on the number and type of schools 
included at each funding level. 

Table 2: Highly VA-Funded Schools at Different Post-9/11 Payment 
Levels: 

Public Schools; 
$2 Million: 
Number of schools: 293; 
Percentage: 45%; 
$6 Million: 
Number of schools: 101; 
Percentage: 42%; 
$10 Million: 
Number of schools: 20; 
Percentage: 27%; 
$16 Million: 
Number of schools: 8; 
Percentage: 21%. 

Nonprofit Schools; 
$2 Million: 
Number of schools: 147; 
Percentage: 22%; 
$6 Million: 
Number of schools: 57; 
Percentage: 24%; 
$10 Million: 
Number of schools: 22; 
Percentage: 29%; 
$16 Million: 
Number of schools: 8; 
Percentage: 21%. 

For-Profit Schools; 
$2 Million: 
Number of schools: 214; 
Percentage: 33%; 
$6 Million: 
Number of schools: 81; 
Percentage: 34%; 
$10 Million: 
Number of schools: 33; 
Percentage: 44%; 
$16 Million: 
Number of schools: 22; 
Percentage: 58%. 

Total; 
$2 Million: 
Number of schools: 654; 
Percentage: 100%; 
$6 Million: 
Number of schools: 239; 
Percentage: 100%; 
$10 Million: 
Number of schools: 75; 
Percentage: 100%; 
$16 Million: 
Number of schools: 38; 
Percentage: 100%. 

Source: GAO analysis of VA data. 

[End of table] 

Our findings at the $5 million and $10 million levels were generally 
consistent with our findings at the $2 million level, although some 
differences were not statistically different. This may be due to the 
fact that analyses at the higher dollar thresholds were based on fewer 
observations. Some results differed at the $15 million dollar level; 
however, this group consisted of a very small number of schools and 
statistical testing on such a small subset of schools may be less 
reliable and more subject to the influence of large outliers than at 
the $2 million, $5 million, or $10 million levels. The results for 
these 38 schools are sensitive to the inclusion/exclusion of specific 
schools. For example, the overall graduation rate is similar to the 
other levels when the largest school is excluded. Therefore, the 
conclusions that can be drawn from statistical analyses at the $15 
million level are more limited than at the $2 million level. Further, 
because of the limited number of schools at the $10 million and $15 
million levels, we could not confirm whether sector differences we 
observe at these payment levels would persist after controlling for 
other factors with regression modeling. 

Objectives 2 and 3: For these objectives, we first had to merge the VA 
payment and enrollment data with IPEDS data on student and school 
characteristics. We matched all of the highly VA-funded schools with 
the appropriate school in the IPEDS database. However, we could not 
match all of the other VA-funded schools for a variety of reasons. For 
example, some VA-funded schools do not participate in federal student 
aid programs and are therefore not included in IPEDS. In addition, the 
data for VA's Vocational Rehabilitation and Employment program 
included payments for tutoring and employment services provided by 
entities other than schools (and that were not relevant to our 
analyses). It is also possible that we were unable to match some 
schools because the school name in the VA data did not match closely 
enough to the school name listed in IPEDS. However, these non-matched 
schools represented a small percentage of the total cases--less than 1 
percent of the total Post-9/11 enrollees and tuition payments. 

We then calculated a variety of student and school characteristics, 
such as the percentage of students who are low-income,[Footnote 30] 
minority, or over the age of 25; and graduation and retention rates. 
[Footnote 31] Education's data do not identify student veterans, so 
these characteristics are for a school's total student population and 
are not veteran specific. We compared student and school 
characteristics at (1) highly VA-funded schools with other schools 
receiving VA education payments and (2) public, non-profit, and for-
profit schools in the highly VA-funded group. For our descriptive 
analysis of student-level characteristics (such as the percentage of 
low-income or minority students or graduation rates), we calculated 
percentages using the student population (rather than the number of 
schools) as the denominator. 

Descriptive Statistics: We conducted various descriptive statistical 
analyses to determine whether highly VA-funded schools had different 
school and student characteristics and outcomes than other schools 
that received payments from VA education benefits. We conducted 
similar analyses to determine whether specific kinds of highly VA-
funded schools had different school and student characteristics and 
outcomes than others. We examined t-tests of means and proportions for 
continuous and categorical variables, comparing both highly VA-funded 
schools to all other schools, and, among highly VA-funded schools, 
sector comparisons. We aggregated across schools within each sector to 
develop implicitly weighted means and proportions for these 
comparisons; accordingly, large outliers may have an influence on the 
results, particularly when looking at smaller subsets of schools. 
Additionally, these kinds of t-tests assume that there are a 
sufficient number of schools in each sector to reliably calculate a 
distribution around the statistic of interest. We did not conduct 
statistical testing across sectors among schools receiving $15 million 
or more in VA payments--when there are as few as 8 schools per sector--
because this assumption does not apply. All differences reported are 
significant at the 95 percent level unless otherwise noted. This 
indicates that there is less than a 5 percent probability that we 
would have gotten such a result by chance if there were really no 
difference in the mean or proportion between variables. These analyses 
examine differences prior to controlling for the effects of other 
variables. 

Multiple Regression Analysis: To develop meaningful comparisons when 
comparing different types of schools, it is important to account for 
differences in student and institutional characteristics as much as 
possible. We used a school-level analysis for the regression analysis. 
We conducted additional analyses of our comparison of highly VA-funded 
schools in different sectors to control for differences in student and 
institutional characteristics. We were not able to conduct a similar 
analysis for our comparison of highly VA-funded schools with other VA-
funded schools because some of the other VA-funded schools did not 
report IPEDS information; however, these schools represent a small 
percentage of total Post-9/11 GI Bill payments and enrollment. 
Multiple regression analysis is a method for exploring how a dependent 
variable is related to a number of independent variables, while 
controlling for other factors that could have an impact on the value 
of the dependent variable. We used ordinary least squares regression 
modeling to assess whether observed differences in student outcomes at 
highly VA-funded schools in different sectors remained after 
controlling for differences in student characteristics. 

We tested multiple model specifications to ensure that our estimates 
were stable across different models, as well as different functional 
forms for the dependent variables to ensure consistency of 
interpretation. While we report the results from our analysis of all 
highly VA-funded schools, we also estimated the models using the 
subset of schools that received $5 million or more in VA payments. 
Except where noted below with respect to the retention rate analysis, 
the results were consistent across specifications and populations, in 
terms of the direction, relative magnitude, and significance of the 
coefficients. Results reported are significant at the 95 percent level. 

For our regression analysis of the proportion of veterans for highly 
VA-funded schools, we used the log of the veteran-student ratio 
(veterans as a percent of all students) to mitigate high skew, the 
impact of outliers, and to ensure that the data better conformed with 
the statistical assumptions that underlie linear regression modeling. 
We controlled for a variety of characteristics in addition to sector, 
including school enrollment in five categories, whether a school 
offered bachelor's or graduate degrees, whether a school offered 
distance education, the school's specialty (information technology, 
health care, or business compared to other schools), tuition, and 
student characteristics.[Footnote 32] Student characteristics included 
the percent of students on Pell Grants (as a proxy for the percent of 
low-income students), the percent of part-time students, and the 
percentages of students who were minority, female, or over the age of 
25. Our model showed that after controlling for other factors there 
was no significant difference in the log of the veteran-student ratios 
across sectors. As tuition increased, the percentage of the student 
body who were veterans generally decreased. When we examined an 
interaction term for sector and tuition, we found that the veteran-
student ratios decreased significantly as tuition increased for public 
and nonprofit schools, but that veteran-student ratio at for-profit 
schools did not appear to fall as tuition increased. That is, as 
tuition increased, the percentage of the student body who were 
veterans at highly VA-funded public and nonprofit schools decreased 
after controlling for school and student characteristics, whereas at 
highly VA-funded for-profit schools, higher tuition rates were not 
associated with significant declines in the proportion of veterans 
attending a school. Our final model with separate terms for sector and 
tuition included 646 schools with complete data on veteran-student 
ratio and all independent variables, and had an R2 of .66, meaning 
that variation in the independent variables predicted approximately 
two-thirds of the variation in the log of the veteran-student ratio. 

When analyzing graduation and retention rates, we adapted our set of 
independent variables to accommodate potential confounds resulting 
from the fact that schools serve students who are pursuing different 
levels of degrees.[Footnote 33] Instead of controlling for the highest 
degree offered, our regression models control for the primary focus of 
the school by including intercepts for schools that primarily serve 
associates' students, schools that do not primarily serve bachelor's 
students, and for schools that offered greater than a bachelors' 
degree. We also included a control for the proportion of veterans in 
the school population, as well as the balance of the student and 
school controls used in our model of the proportion of veterans at a 
school. We excluded from the regression analyses of graduation and 
retention rates those few highly VA-funded schools that reported 
serving only graduate students or not offering degrees, and schools 
that had graduation rates more than 10 percentage points higher than 
their retention rates.[Footnote 34] After excluding these records, 
records with missing or problematic graduation or retention rate data, 
and other records with missing data on independent variables, our 
regression analyses included 608 and 582 schools respectively for the 
analyses of graduation and retention rate. 

We found that, on average, highly VA-funded nonprofit and for-profit 
schools had significantly higher graduation rates than public schools, 
after adjusting for other factors. Nonprofit schools' graduation rates 
were, on average, about 6.5 percentage points higher than those of 
public schools, and the graduation rate of for-profits was on average 
about 6.0 percentage points higher than those of public schools; the 
difference between nonprofit schools' rates and those of for-profit 
schools was not statistically significant. Across sectors, schools 
with higher tuition tended to have significantly higher graduation 
rates than their lower-priced counterparts, controlling for school and 
student characteristics. Our final model with separate terms for 
sector and tuition had an R2 of .74, which suggests that the variation 
in the independent variables predicted approximately three quarters of 
the variation in the graduation rate. Our regression analysis of 
retention rates showed that controlling for other factors, public and 
nonprofit schools had similar retention rates, while on average, for-
profit schools had significantly lower retention rates than public and 
nonprofit schools. After controlling for school and student 
characteristics, for-profit schools had retention rates that were on 
average approximately 6 points lower than that of public schools with 
similar characteristics. Additionally, although schools with higher 
tuitions tended to have higher retention rates than their lower-priced 
counterparts, an interaction term to explore this relationship showed 
that it did not hold for for-profit schools. In other words, although 
higher priced public and nonprofit schools had higher retention rates 
than low-priced public schools, there was no statistically significant 
difference between low priced public school's retention rates and 
those of higher-priced for-profit schools. Our final model with 
separate terms for sector and tuition had an R2 of .60, meaning that 
the variation in our independent variables predicted approximately 60 
percent of the variation in the retention rate. 

The student loan default rate is reported at the Office of 
Postsecondary Education Identification (OPEID-6) level that may link 
to a central campus or school, rather than individual campuses; 
consequently we used a collapsed dataset to assess differences in 
default rate at the central campus or school level.[Footnote 35] The 
dataset collapsed together highly VA-funded schools based on the OPEID-
6 identifier, which left approximately 560 schools and chains for 
analysis; due to missing data on independent variables our models of 
default rate include 546 cases. Although the majority of our control 
variables overlap with those used in other models, we did not include 
in our default rate analysis variables that were unique to a site or 
campus, such as whether a school offered distance education. For 
student characteristic variables, each individual variable reflects 
the number of students in the category across all sites or campuses 
over the total number of students across sites or campuses. Our 
tuition measure is a weighted average across campuses, weighted by 
total enrollment. The size variable is based on the total enrollment 
at all campuses. Our regression analysis shows that among schools and 
chains, highly VA-funded for-profit schools and chains had 
significantly higher student loan default rates than nonprofit and 
public schools. The default rate of nonprofits was slightly but 
significantly below that of public schools, controlling for other 
factors. Overall, highly VA-funded for-profit schools and chains had 
default rates that were, on average, about 6.5 percentage points 
higher than their public counterparts, whereas the default rate at 
nonprofits was approximately 1 percentage point lower than that for 
public schools, after adjusting for tuition, size, and student 
characteristics. When we limited our analysis to schools receiving $5 
million or more in VA payments, we found that while for-profit schools 
and chains still appeared to have higher default rates than public 
schools, the default rate of nonprofits was not significantly 
different from that of public schools. Our final model of the default 
rate at all highly VA-funded central campuses and institutions had an 
R2 of about .70, which suggests that variation in the independent 
variables in the model predicted approximately 70 percent of the 
variation in the default rate. 

Limitations of the Analysis: 

The student characteristics and outcomes included in our analyses are 
for all students enrolled at a school and are not specific to student 
veterans. While veterans typically represent a small percentage of a 
school's overall student population, we analyzed the characteristics 
and outcomes of the overall student populations because veteran-
specific data do not exist. 

The student outcomes measures used in our analyses have certain 
limitations. For example, the IPEDS graduation rates include only full-
time, first-time students and thus exclude a significant number of 
other students (e.g., those who attend part-time or transfer to 
another school). For schools with a bachelor's degree as the highest 
degree offered, the IPEDS retention rates include only students 
seeking a bachelor's degree, even if most students at the school 
receive associate's degrees. Finally, Education tracks only student 
loan default rates for a 3-year period after a borrower is scheduled 
to begin repaying the student loan. Defaults that occur after this 
three-year period are not counted in a school's default rate. 

Lastly, schools do not always report their IPEDS graduation and 
retention rate data using the same unit of analysis as their default 
and 90/10 rates. Education allows a school to report data on related 
campuses in several ways. For example, a school with five related 
campuses may report a separate graduation rate for each campus, but a 
single 90/10 rate or a single default rate for the five campuses 
together. In these cases, we assigned the single 90/10 rate or default 
rate to each related campus for our descriptive analysis. For our 
regression modeling of default rates we collapsed data to the central 
campus or institution level. 

[End of section] 

Appendix II: For-Profit Schools and the 90-10 Requirement: 

To maintain their eligibility for federal student aid programs (those 
authorized by Title IV of the Higher Education Act of 1965, as 
amended, such as Pell Grants and William D. Ford Federal Direct 
Loans), for-profit schools must obtain at least 10 percent of their 
total revenues from sources other than federal student aid programs--
this is known as the 90/10 rule.[Footnote 36] Because VA education 
payments are not authorized by this legislation, it is counted as non-
federal student aid and thus toward the 10 percent requirement; as a 
result, enrolling veterans helps for-profit schools meet the 90/10 
rule. The percentage of a for-profit school's total revenues obtained 
from federal student aid programs is known as the "90/10 rate." If a 
for-profit school does not comply with the 90/10 rule for 1 fiscal 
year, the school's eligibility to participate in the federal student 
aid program becomes provisional for the next 2 years.[Footnote 37] If 
noncompliance occurs for 2 consecutive fiscal years, the school 
becomes ineligible to participate in the federal student aid program 
for at least 2 fiscal years. 

Schools with high levels of revenue from sources other than federal 
student aid (such as Post-9/11 GI Bill payments) could be expected to 
have a lower proportion of overall revenue from federal student aid 
programs and therefore lower 90/10 rates. This is what we found when 
we compared 90/10 rates at highly VA-funded and other VA-funded for-
profit schools. Specifically, a somewhat higher percentage of highly 
VA-funded schools had 90/10 rates at 85 percent or below, compared to 
other for-profit schools (see figure 9). In other words, highly VA-
funded for profit schools had lower 90/10 rates in the aggregate than 
other for profit schools, meaning that they received proportionally 
less federal student aid than other for-profit schools. [Footnote 38] 

Figure 9: 90/10 Rates of Highly VA-Funded and Other VA-Funded Schools, 
School Year 2011-12: 

[Refer to PDF for image: vertical bar graph] 

Percentage of schools at 90/10 rate 85 percent or below: 

Highly VA-funded schools: 90%; 
Other VA-funded schools: 82%. 

Source: GAO analysis of VA and IPEDS data. 

Note: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

[End of section] 

Appendix III: School Characteristics at Highly VA-funded Schools 
Compared to Other VA-Funded Schools: 

This appendix presents additional data comparing school 
characteristics at highly VA-funded schools and other VA-funded 
schools. 

Figure 10: Number of Students at Highly VA-Funded and Other VA-Funded 
Schools, School Year 2010-2011: 

[Refer to PDF for image: stacked vertical bar graph] 

Highly VA-funded schools: 
More than 40,000 students: 10%; 
20,001 to 40,000 students: 24%; 
10,001 to 20,000 students: 21%; 
1,501 to 10,000 students: 33%; 
351 to 1,500 students: 11%; 
0 to 350 students: 0%. 

Other VA-funded schools: 
More than 40,000 students: 0%; 
20,001 to 40,000 students: 2%; 
10,001 to 20,000 students: 0%; 
1,501 to 10,000 students: 32%; 
351 to 1,500 students: 33%; 
0 to 350 students: 26%. 

Source: GAO analysis of VA and IPEDS data. 

Note: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

Table 3: Fiscal Year 2011 Post-9/11 GI Bill Funding, Numbers of 
Veterans, and Funding per Veteran: 

Highly VA-Funded Schools; 
FY 2011 Post-9/11 GI Bill Funding: 2,436,469,986; 
FY 2011 Post-9/11 GI Bill Student Veterans: 333,739; 
FY 2011 Funding Per Veteran: $7,300.53; 
School Year 2011-2012 Average Tuition: $18,829. 

Other; 
FY 2011 Post-9/11 GI Bill Funding: 1,139,785,636; 
FY 2011 Post-9/11 GI Bill Student Veterans: 222,549; 
FY 2011 Funding Per Veteran: $5,121.50; 
School Year 2011-2012 Average Tuition: $16,689. 

Source: GAO Analysis of VA and IPEDS data. 

Notes: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from FY 2010 through FY 
2011. 

Post-9/11 GI Bill funding per veteran is influenced by several 
factors, including tuition rates, program offerings, and use of the 
Yellow Ribbon program. 

Tuition may be higher at the highly VA-funded schools because they all 
offer at least an associate's degree, while some other VA-funded 
schools offer only non-degree programs; tuition tends to increase for 
more advanced degrees. 

FY = fiscal year. 

[End of table] 

Figure 11: Comparison of Participation in VA's Yellow Ribbon Program 
and Offering Distance Education at Highly VA-Funded and Other VA-
Funded Schools, School Year 2011-12: 

[Refer to PDF for image: vertical bar graph] 

School characteristics: Yellow Ribbon; 
Highly VA-Funded Schools: 95%; 
Other VA-funded schools: 17%. 

School characteristics: Distance Education; 
Highly VA-Funded Schools: 84%; 
Other VA-funded schools: 52%. 

Source: GAO Analysis of VA and IPEDS data. 

Notes: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

Veterans attending nonprofit or for-profit schools (or public schools 
as out-of-state students) may receive additional benefits to cover 
tuition costs that exceed their Post-9/11 GI Bill benefit through VA's 
Yellow Ribbon Program. Under this program, participating schools enter 
into a voluntary agreement with VA to pay a portion of the tuition and 
fees that exceed an individual's Post-9/11 GI Bill benefit and VA 
matches the school's contribution. 

Distance education refers to schools that offer any courses via the 
internet, video, or other forms outside the classroom. 

[End of figure] 

Figure 12: School Year 2011-12 Highest Degree Offering at Highly VA-
Funded and Other VA-Funded Schools: 

[Refer to PDF for image: stacked vertical bar graph] 

School type: Highly VA-Funded Schools; 
Graduate degree: 67%; 
Bachelor's degree: 19%; 
Associate's degree: 15%; 
Less than Associate's degree: 0. 

School type: Other VA-funded schools; 
Graduate degree: 29%; 
Bachelor's degree: 14%; 
Associate's degree: 36%; 
Less than Associate's degree: 21%. 

Source: GAO Analysis of VA and IPEDS data. 

Note: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

[End of section] 

Appendix IV: Comparisons of School Characteristics across Public, 
Nonprofit, and For-Profit Schools: 

This appendix presents additional data comparing student outcomes and 
school characteristics at highly VA-funded for-profit, nonprofit, and 
public schools. 

Figure 13: Retention, Overall Graduation, 4-and 2-Year Graduation, and 
3-Year Student Loan Default Rates across Highly VA-Funded Schools, by 
Sector, School Year 2011-12, before Controlling for Differences in 
Student and School Characteristics: 

[Refer to PDF for image: vertical bar graph] 

Retention rate: 
Public schools: 79%; 
Nonprofit schools: 84%; 
For-Profit schools: 60%. 

Overall graduation rate: 
Public schools: 54%; 
Nonprofit schools: 69%; 
For-Profit schools: 49%. 

4-year graduation rate: 
Public schools: 59%; 
Nonprofit schools: 69%; 
For-Profit schools: 39%. 

2-year graduation rate: 
Public schools: 15%; 
Nonprofit schools: n/a; 
For-Profit schools: 64%. 

3-year student loan default rate: 
Public schools: 8%; 
Nonprofit schools: 6%; 
For-Profit schools: 28%. 

Source: GAO Analysis of VA and IPEDS data on highly VA-funded schools. 

Note: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

Figure 14: School Year 2010-11 Enrollment Levels at Highly VA-Funded 
Schools, by Sector: 

[Refer to PDF for image: stacked vertical bar graph] 

Public schools: 
More than 40,000 students: 18%; 
20,001 to 40,000 students: 44%; 
10,001 to 20,000 students: 28%; 
1,501 to 10,000 students: 10%; 
351 to 1,500 students: 0%; 
0 to 350 students: 0%. 

Nonprofit schools: 
More than 40,000 students: 3%; 
20,001 to 40,000 students: 17%; 
10,001 to 20,000 students: 30%; 
1,501 to 10,000 students: 48%; 
351 to 1,500 students: 3%; 
0 to 350 students: 0%. 

For-profit schools: 
More than 40,000 students: 4%; 
20,001 to 40,000 students: 3%; 
10,001 to 20,000 students: 5%; 
1,501 to 10,000 students: 56%; 
351 to 1,500 students: 30%; 
0 to 350 students: 1%. 

Source: GAO analysis of VA and IPEDS data on highly VA-funded schools. 

Note: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

Figure 15: School Year 2011-12 Comparison of VA's Yellow Ribbon 
Program Participation and Distance Education at Highly VA-Funded 
Schools, by Sector: 

[Refer to PDF for image: vertical bar graph] 

School characteristics: Yellow Ribbon; 
Public schools: 94%; 
Nonprofit schools: 95%; 
For-profit schools: 96%. 

School characteristics: Distance Education; 
Public schools: 97%; 
Nonprofit schools: 81%; 
For-profit schools: 67%. 

Source: GAO analysis of VA and IPEDS data on highly VA-funded schools. 

Notes: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

Veterans attending nonprofit or for-profit schools (or public schools 
as out-of-state students) may receive additional benefits to cover 
tuition costs that exceed their Post-9/11 GI Bill benefit through VA's 
Yellow Ribbon Program. Under this program, participating schools enter 
into a voluntary agreement with VA to pay a portion of the tuition and 
fees that exceed an individual's Post-9/11 GI Bill benefit and VA 
matches the school's contribution. 

Distance education refers to schools that offer any courses via the 
internet, video, or other forms outside the classroom. 

Figure 16: School Year 2011-12 Highest Degree Offering at Highly VA-
Funded Schools, by Sector: 

[Refer to PDF for image: stacked vertical bar graph] 

School type: Public; 
Graduate degree: 77%; 
Bachelor's degree: 5%; 
Associate's degree: 17%. 

School type: Nonprofit; 
Graduate degree: 95%; 
Bachelor's degree: 5%; 
Associate's degree: 0. 

School type: For-profit; 
Graduate degree: 32%; 
Bachelor's degree: 47%; 
Associate's degree: 21%. 

Source: GAO analysis of VA and IPEDS data on highly VA-funded schools. 

Note: Highly VA-funded schools are those that received at least $2 
million in total Post-9/11 GI Bill payments from fiscal year 2010 
through fiscal year 2011. 

[End of figure] 

[End of section] 

Appendix V: Comments from the Department of Veterans Affairs: 

Department of Veterans Affairs: 
Washington DC 20420: 

July 12, 2013: 

Ms. Melissa Emrey-Arras: 
Director:
Education, Workforce, and Income Security Issues: 
U.S. Government Accountability Office: 
441 G Street, NW: 
Washington, DC 20548: 

Dear Ms. Emrey-Arras: 

The Department of Veterans Affairs (VA) has reviewed the Government 
Accountability Office's (GAO) draft report, "VA Education Benefits: 
Student Characteristics and Outcomes Vary across Schools" (GAO-13-
567). VA generally agrees with GAO's findings. 

The enclosure contains technical comments related to the draft report. 
VA appreciates the opportunity to comment on your draft report. 

Sincerely, 

Signed by: 

Jose D. Riojas: 
Interim Chief of Staff: 

Enclosure: 

[End of section] 

Appendix VII: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Melissa Emrey-Arras, (617) 788-0534 or emreyarrasm@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, individuals making key 
contributions to this report were Michelle St. Pierre, Assistant 
Director; Mariana Calderon, Nathan Gottfried, Jennifer Gregory, Andrew 
Nelson, and Tranchau Nguyen. In addition, key support was provided by 
Deborah Bland, Will Colvin, Kirsten Lauber, Theresa Lo, John Mingus, 
Mimi Nguyen, Anna Maria Ortiz, Michelle A. Sager, George Scott, Kate 
van Gelder, Mark Ward, and Craig Winslow. 

[End of section] 

Footnotes: 

[1] Pub. L. No. 110-252, tit. V, 122 Stat. 2323, 2357-86. 

[2] VA has initiated several recent efforts to collect new outcome 
information on student veterans, which should eventually shed light on 
the extent to which veterans are achieving successful postsecondary 
outcomes. See GAO, VA Education Benefits: Efforts Needed to Improve 
Program Management and Provide More Information to Student Veterans, 
[hyperlink, http://www.gao.gov/products/GAO-13-338] (Washington, D.C.; 
May 22, 2013). 

[3] We used the Post-9/11 GI Bill program for the designation of 
"highly VA-funded school" because it provides for tuition and fee 
payments to be made directly to schools, whereas most other VA 
education programs do not make tuition and fee payments separately. 
The Post-9/11 GI Bill program is also the largest and fastest-growing 
VA education program. 

[4] Veterans typically comprise a small proportion of a school's total 
enrollment, so veterans would generally have little impact on those 
characteristics and outcomes for the school population as a whole. 

[5] We could not conduct regression analyses for question two to 
compare highly VA-funded schools and other VA-funded schools because 
some other VA-funded schools were not included in Education's IPEDS 
data and therefore data on their characteristics were not available. 
For question three, we also conducted regression analyses at the $5 
million level to identify any significant differences compared to the 
$2 million regression results. Regression analyses were not possible 
at the $10 or $15 million level because of the limited number of 
schools in some sectors. 

[6] Servicemen's Readjustment Act of 1944, Pub. L. No. 78-346, 58 
Stat. 284. 

[7] VA's Vocational Rehabilitation and Employment program also makes 
tuition and fee payments directly to a school and pays a separate 
monthly housing allowance to students. This program offers a variety 
of services to help veterans with service-connected disabilities and 
employment barriers obtain stable and suitable employment, including 
educational benefits. 

[8] For the purposes of this report, we define federal student aid 
programs as financial aid programs authorized under Title IV of the 
Higher Education Act, 20 U.S.C. §§ 1070-1099d and 42 U.S.C. §§ 2751-
2756b. 

[9] 20 USC § 1087vv(j). See also GAO, Higher Education: Veteran 
Students Received Similar Amounts of Title IV Aid As Nonveterans but 
More Total Aid with GI Benefits, [hyperlink, 
http://www.gao.gov/products/GAO-08-741], (Washington, D.C.: June 20, 
2008). 

[10] 20 U.S.C. § 1094(a)(24). 

[11] For this analysis, we used tuition and fee payments made directly 
to schools for the Post-9/11 GI Bill and the Vocational Rehabilitation 
and Employment program, but excluded payments made to student veterans 
for books and housing expenses, unless otherwise stated. For other VA 
education programs that pay a fixed monthly amount to veterans, we 
attributed to the school the total VA payments made to veterans, which 
may overestimate VA payments received by schools with tuition rates 
that are less than total VA payments. 

[12] The schools in this highly VA-funded group each received at least 
$2 million in total funding from fiscal years 2010 through 2011 from 
the Post-9/11 GI Bill. 

[13] The Vocational Rehabilitation and Employment program includes a 
variety of postsecondary education providers, including computer 
training companies and law enforcement academies, which are included 
in total schools. 

[14] The average VA payment amount per veteran is influenced by 
several factors, including tuition rates, program offerings, and use 
of the Yellow Ribbon GI Education Enhancement (Yellow Ribbon) Program. 
Under this program, participating private schools or public schools 
charging out-of-state tuition enter into a voluntary agreement with VA 
to pay a portion of the tuition and fees that exceed an individual's 
Post-9/11 GI Bill benefit and VA matches the school's contribution. 38 
U.S.C. § 3317. 

[15] Our analyses of student and school characteristics are based on 
data for all students at a school and are not specific to veteran 
students. Our analyses are also limited to schools that reported data 
to the IPEDS system; some VA-funded schools are not required to report 
data to IPEDS. However, all highly VA-funded schools reported data to 
IPEDS and therefore were included in our analyses. Furthermore, the 
other VA-funded schools that did not report information to IPEDS 
represented a small percentage of the total cases--less than 1 percent 
of the total Post-9/11 GI Bill enrollees and tuition and fee payments. 
Retention rate is a measure of the rate at which students persist in 
their educational program at an institution, expressed as a percentage. 

[16] See appendices II and III for additional analysis of highly VA-
funded and other VA-funded schools. 

[17] The $5 million level created a highly VA-funded group of 239 
schools and the $10 million level created a group of 75 schools. 

[18] See appendix IV for additional analysis of highly VA-funded 
public, nonprofit, and for-profit schools. 

[19] Previous GAO analysis found similar differences in student 
characteristics among public, nonprofit, and for-profit schools that 
participate in federal student aid programs. See GAO, Postsecondary 
Education: Student Outcomes Vary at For-Profit, Nonprofit, and Public 
Schools, [hyperlink, http://www.gao.gov/products/GAO-12-143] 
(Washington, D.C.: Dec. 7, 2011). 

[20] We controlled for a number of school and student characteristics, 
such as the size of the school, the proportion of low-income and 
minority students, and whether a school offered bachelor or graduate 
degrees. 

[21] IPEDS retention and graduation rates are not directly related as 
they are calculated for different groups of students. IPEDS retention 
rates are calculated for first-time degree or certificate-seeking 
students from the previous fall who either re-enrolled or successfully 
completed their program by the current fall, except for 4-year 
schools, where they are calculated for first-time bachelors degree-
seeking undergraduates from the previous fall who are again enrolled 
in the current fall. By contrast, IPEDS graduation rates are 
calculated for students entering a school as full-time, first-time, 
degree/certificate-seeking undergraduate students in a particular year. 

[22] In a previous report, we reviewed existing studies on students' 
outcomes that controlled for differences in student characteristics at 
public, nonprofit, and for-profit schools participating in federal 
student aid programs. We found that two studies showed that students 
at for-profit schools had higher graduation rates for certificate 
programs, similar graduation rates for associate's degree programs, 
and lower graduation rates for bachelor's degree programs than 
students at public and nonprofit schools. The studies we reviewed 
generally used a different data source for graduation rates than this 
report; they focused on individual student graduation rates, rather 
than aggregate school-level graduation rates reported in IPEDS. See 
[hyperlink, http://www.gao.gov/products/GAO-12-143]. Our regression 
analysis of highly VA-funded schools controlled for schools offering 
programs of different lengths; however, methodological challenges with 
IPEDS graduation rates limited our ability to separately analyze 
graduation rates for different types of programs. 

[23] Across sectors, schools with higher tuition tended to have higher 
retention and graduation rates than their lower-priced counterparts. 
Our model also showed that, as tuition increased, the percentage of 
the student body who were veterans decreased at highly VA-funded 
public and nonprofit schools. However, at highly VA-funded for-profit 
schools, higher tuition rates were not associated with significant 
declines in the proportion of veterans attending a school, after 
controlling for differences in school and student characteristics. 

[24] In some cases, the student loan default rate is reported at a 
central campus level, rather than individual campuses. Accordingly, we 
used a collapsed dataset to assess differences in default rate at the 
central campus level. 

[25] In a previous report, we reviewed existing studies on students' 
outcomes that controlled for differences in student characteristics at 
public, nonprofit, and for-profit schools participating in federal 
student aid programs. In that report, we reviewed the results from two 
studies that show students at for-profit schools had higher default 
rates than those at 4-year public schools, but had mixed results 
compared to students at other schools--specifically, one study shows 
that students at for-profit schools had higher default rates than 
those at 4-year nonprofit schools and 2-year nonprofit and public 
schools, while the other study did not show any statistically 
significant differences. See [hyperlink, 
http://www.gao.gov/products/GAO-12-143]. 

[26] 20 U.S.C. §§ 1070-1099d and 42 U.S.C. 2751-2756b. 

[27] 20 U.S.C. § 1094(a)(17). 

[28] 20 U.S.C. §§ 1070-1099d and 42 U.S.C. §§ 12751-2756b. 

[29] 20 U.S.C. § 1094(a)(24). 

[30] We use the term "low-income" to describe students who received 
federal Pell grants, a need-based federal student aid program for low-
income students. 

[31] Retention rate is a measure of the rate at which students persist 
in their educational program at an institution, expressed as a 
percentage. For 4-year institutions, this is the percentage of first-
time bachelors (or equivalent) degree-seeking undergraduates from the 
previous fall who are again enrolled in the current fall. For all 
other institutions this is the percentage of first-time 
degree/certificate-seeking students from the previous fall who either 
re-enrolled or successfully completed their program by the current 
fall. 

[32] Enrollment was broken into five categories, based roughly on the 
10th percentile, the 25th percentile, the 75th percentile and the 90th 
percentile. Depending on the model, tuition was broken into five 
categories based roughly on the 10th, 25th, 75th and 90th percentiles, 
or into three categories as part of an interaction term with sector 
based loosely on the 10th and 90th percentile cutoffs. 

[33] Schools that have a 4-year component but that serve primarily 
associates' degree or other students may report only the 4-year 
graduation rate, even if 4-year students are a relatively small 
portion of the school's student body. 

[34] Although in theory a school cannot graduate a larger proportion 
of students than it retains, in practice the use of different student 
cohorts in the denominator of graduation and retention rate 
calculations can result in minor misalignments across years. 

[35] The OPEID-6 is the identification number used by Education's 
Office of Postsecondary Education to identify schools that have 
program participation agreements so that its students are eligible to 
participate in federal student financial assistance programs under 
Title IV regulations. 

[36] 20 U.S.C. 1094(a)(24). 

[37] 34 C.F.R. § 1094(d)(2). 

[38] For more information on 90/10 rates, see GAO, For-Profit Schools: 
Large Schools and Schools that Specialize in Healthcare Are More 
Likely to Rely Heavily on Federal Student Aid, [hyperlink, 
http://www.gao.gov/products/GAO-11-4] (Washington, D.C.: Oct. 4, 2010). 

[End of section] 

GAO’s Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO’s website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, go to [hyperlink, http://www.gao.gov] and select 
“E-mail Updates.” 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO’s 
website, [hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information. 

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

Congressional Relations: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, DC 20548. 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, DC 20548. 

[End of document]