Student Financial Aid: Schools' Experiences Using the National Student
Loan Data System (Letter Report, 09/24/98, GAO/HEHS-98-192).

Pursuant to a congressional request, GAO reviewed schools' use of the
National Student Loan Data System (NSLDS), focusing on: (1) the extent
to and purposes for which schools are using NSLDS; (2) problems these
schools are having and the benefits they are getting from using the
system; (3) why some schools are not using NSLDS; and (4) the extent to
which the Department of Education is taking or plans to take steps to
ensure that schools are fully using NSLDS.

GAO noted that: (1) postsecondary schools participating in federal
student financial aid programs are making limited use of NSLDS' online
and batch processing functions; (2) GAO estimated that almost half of
the schools are not using these system capabilities at all--3 years
after they first become available; (3) those that are using these
functions are not routinely using them for many of the tasks they are
capable of performing; (4) the one use made by the majority of schools
is to provide and update student financial data on the student status
confirmation report, which the Department now requires all schools to
perform; (5) GAO estimated that over half of the schools rarely or never
performed 7 of the 10 tasks that GAO identified for its survey using
NSLDS' online and batch processing functions; (6) in general, schools'
experiences using NSLDS have been relatively problem free; however, some
schools did experience problems with some aspects of the system; (7)
schools using NSLDS' online and batch processing functions had mixed
views on whether they led to improvements in their program
administration; (8) schools that did not use NSLDS' online and batch
processing functions cited a variety of reasons for not doing so; (9)
the most frequent reasons cited by these schools included relying on
alternative methods to obtain or submit data needed to administer
student aid programs; (10) of the schools that did not use the systems'
online and batch processing functions, many did not have plans to obtain
access to NSLDS, had plans to obtain access to the system but did not
know when, or were unsure when they would obtain access to the system in
the future; (11) in an effort to increase schools' use of NSLDS, the
Department has provided training assistance to schools and has worked to
ensure the accuracy of the system's data; (12) the Department recently
expanded its NSLDS customer service center and will offer NSLDS training
to users at its 11 regional training centers; (13) in addition, to
demonstrate its commitment to improving the reliability of data on its
postsecondary education programs, the Department has addressed the issue
of data integrity in its long-range strategic and annual performance
plans prepared in response to the Government Performance and Results Act
of 1993; (14) as part of this committment, the Department has initiated
efforts to identify and correct inaccurate data in NSLDS, and to
strengthen its working relationships with other data providers, such as
guaranty agencies.

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

 REPORTNUM:  HEHS-98-192
     TITLE:  Student Financial Aid: Schools' Experiences Using the 
             National Student Loan Data System
      DATE:  09/24/98
   SUBJECT:  Higher education
             Student financial aid
             Student loans
             Loan accounting systems
             Surveys
             Data integrity
             Technical assistance
             Strategic planning
             Data bases
             Schools
IDENTIFIER:  William D. Ford Federal Direct Loan Program
             Federal Family Education Loan Program
             Dept. of Education National Student Loan Data System
             Dept. of Education Perkins Student Loan Program
             Pell Grant
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to the Chairman, Subcommittee on Oversight and Investigations,
Committee on Education and the Workforce, House of Representatives

September 1998

STUDENT FINANCIAL AID - SCHOOLS'
EXPERIENCES USING THE NATIONAL
STUDENT LOAN DATA SYSTEM

GAO/HEHS-98-192

Schools' Use of NSLDS

(104890)


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

  FDLP - William D.  Ford Direct Loan Program
  FFELP - Federal Family Education Loan Program
  HEA - Higher Education Act
  ISIR - Institutional Student Information Record
  NSLDS - National Student Loan Data System
  OIG - Office of Inspector General
  PEPS - Postsecondary Education Participants System
  SAR - Student Aid Report
  SEOG - Supplemental Educational Opportunity Grant
  SSCR - Student Status Confirmation Report

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


B-280409

September 24, 1998

The Honorable Pete Hoekstra
Chairman, Subcommittee on Oversight
  and Investigations
Committee on Education and the Workforce
House of Representatives

Dear Mr.  Chairman: 

The Department of Education provides loans and grants to students to
help finance their postsecondary education.  The Department reported
that during fiscal year 1997, $43.3 billion in student financial aid
was awarded to 8.1 million recipients.  Concerns over unreliable data
in the Department's student loan database as well as its ability to
effectively manage its student loan programs led the Congress in 1986
to authorize the Secretary of Education to develop a national student
loan database.  Recognizing the complex nature of its multiple loan
and grant programs and the need for good data to ensure program funds
are awarded appropriately and loans are repaid promptly, the
Department responded by developing the National Student Loan Data
System (NSLDS). 

NSLDS has three main goals:  (1) improve the quality and
accessibility of student financial aid data, (2) reduce the burden of
administering the Department's student financial aid programs, and
(3) minimize fraud and abuse in these programs.  The Department
intended for NSLDS to be used by schools, lenders, third-party
servicers, and guaranty agencies\1 to help determine student
eligibility for aid, identify the status of borrowers' loans, update
student information, and serve as an overall financial aid history
file on program participants.  As of February 1997, the Department
requires all schools to use NSLDS to report, confirm, and update
enrollment dates and status of borrowers--key information in
determining student eligibility for federal aid. 

The need for improved controls in the Department's student financial
aid systems is well documented.  In 1995, we reported and testified
that the Department had, in general, ineffectively used available
student financial aid data to enforce compliance with federal
requirements.\2 For example, Department data indicated that
approximately 43,500 ineligible students had received more than
58,000 loans, totaling over $138 million.  Furthermore, according to
data in both the loan and grant systems, more than 101,000 students
who had loans and subsequently became ineligible for additional aid
may have received more than 139,000 Pell grants totaling
approximately $200 million.\3 In July 1997, we found that several
schools we visited chose to use paper transcripts to obtain student
financial aid histories because they considered NSLDS' electronic
data unreliable.\4 In addition, the Department could not obtain
complete, accurate, and reliable FFELP data necessary for reporting
on its financial position. 

As a result of continuing concerns about the Department's ability to
improve the reliability and efficiency of student financial aid
information and delivery systems and school officials' concerns about
unreliable electronic data, you asked that we report on schools' use
of NSLDS.  Specifically, you requested that we (1) determine the
extent to and purposes for which schools are using NSLDS; (2)
identify any problems these schools are having and the benefits they
are getting from using the system; (3) determine why some schools are
not using NSLDS; and (4) describe the extent to which the Department
is taking or plans to take steps to ensure that schools are fully
using NSLDS. 

To respond to your request, we surveyed a random sample of 600 of the
nearly 6,200 postsecondary schools that participated in federal
student financial aid programs as of August 1997 on their use of
NSLDS.  (For school responses to our survey, see app.  I.) We defined
use of NSLDS as accessing its on-line or batch processing functions
to perform specific tasks.\5 To focus on school personnel's direct
use of these functional capabilities, we instructed survey
respondents not to consider their use of student eligibility
information reports generated from the NSLDS database and sent by the
Department as accessing NSLDS.  Similarly, if schools only used the
National Student Loan Clearinghouse to process Student Status
Confirmation Reports (SSCR)\6 and did not use any of the system's
other functional capabilities, we did not consider those schools to
be NSLDS users.  We asked schools to have the person or persons on
their staff who are the most knowledgeable in using NSLDS to complete
the survey instrument and obtain any needed input from servicers. 

In developing and pretesting our survey instrument, we met with
officials from the Department of Education, the contractor
responsible for developing and maintaining NSLDS, and representatives
from the higher education community.  The survey instrument was
administered in November 1997; with an 83-percent response rate, our
survey results represent the universe of schools.  We did not verify
data provided by the schools; however, we did examine responses for
extreme values and inconsistencies.  The results we report are based
on experiences reported by schools and reflect the self-assessments
of the officials who completed the survey instrument.  We did not
make judgments about the importance of the tasks or interpret or draw
conclusions about the significance of these results or their
implications for NSLDS' implementation.  We conducted our review and
analyses between May 1997 and July 1998 in accordance with generally
accepted government auditing standards.  (For a more detailed
discussion of our scope and methodology, see app.  II.)


--------------------
\1 Guaranty agencies are state-designated agencies that guarantee
Federal Family Education Loan Program (FFELP) loans against default. 
Guaranty agencies are intermediaries between the Department and
lenders, insuring student loans made by lenders and making certain
that the lenders and schools meet program requirements. 

\2 Student Financial Aid:  Data Not Fully Utilized to Identify
Inappropriately Awarded Loans and Grants (GAO/HEHS-95-89, July 11,
1995) and Student Financial Aid:  Data Not Fully Utilized to Identify
Inappropriately Awarded Loans and Grants (GAO/T-HEHS-95-199, July 12,
1995). 

\3 Pell grants, authorized by the Higher Education Act of 1965, as
amended (HEA), are awarded to eligible students based in part on
their financial need and cost of attendance. 

\4 Student Financial Aid Information:  Systems Architecture Needed to
Improve Programs' Efficiency (GAO/AIMD-97-122, July 29, 1997). 

\5 Schools can access NSLDS through store-and-forward (batch
processing) or on-line.  Batch processing allows the school to
electronically send to and receive from NSLDS large data, or "batch,"
files through an electronic mailbox on the Department's title IV wide
area network.  The Department uses batch processing to send files,
reports, and transcripts to schools.  Schools can also obtain on-line
access through their computers to update or request information, such
as that found on a student's financial aid transcript. 

\6 SSCR is the primary means of verifying borrowers' loan privileges
and determining the federal government's monetary obligations. 
Schools enter SSCR data either directly through NSLDS or through the
National Student Loan Clearinghouse--a third-party servicer
established by guaranty agencies and lenders to simplify the SSCR
process--which submits information to NSLDS. 


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

Postsecondary schools participating in federal student financial aid
programs are making limited use of NSLDS' on-line and batch
processing functions.  We estimate that almost half of the schools
are not using these system capabilities at all--3 years after they
first became available.  Those that are using these functions are not
routinely using them for many of the tasks they are capable of
performing.  The one use made by the majority of schools is to
provide and update SSCR information, which the Department now
requires all schools to perform.  We estimate that more than half of
the schools rarely or never performed 7 of the 10 tasks that we
identified for our survey using NSLDS' on-line and batch processing
functions. 

In general, schools' experiences using NSLDS have been relatively
problem free; however, some schools did experience problems with some
aspects of the system.  For example, almost one-fourth had a problem
using NSLDS to correct or update SSCR information, such as enrollment
dates and borrower status.  Similarly, while most schools rarely or
never encountered data inaccuracies, 7 to 29 percent of schools found
occasional inaccuracies in several data fields, such as Social
Security number, last name, and date of birth, which are critical for
properly identifying students. 

Schools using NSLDS' on-line and batch processing functions had mixed
views on whether they led to improvements in their program
administration.  For example, half or more of the schools believed
NSLDS has improved the availability of student aid data (61 percent),
made student data easier to access (59 percent), and reduced the
amount of paper handled in administering student financial aid
programs (50 percent).  On the other hand, less than half believed
NSLDS reduced the time required for student financial aid
administration (31 percent) or reduced necessary staff (15 percent). 

Schools that did not use NSLDS' on-line and batch processing
functions cited a variety of reasons for not doing so.  The most
frequent reasons cited by these schools included relying on
alternative methods to obtain or submit data needed to administer
student aid programs--such as relying on the Clearinghouse to update
the enrollment status of their borrowers--and facing resource or
staff skill limitations, such as a lack of training.  Of the schools
that did not use the system's on-line and batch processing functions,
many did not have plans to obtain access to NSLDS, had plans to
obtain access to the system but did not know when, or were unsure
when they would obtain access to the system in the future. 

In an effort to increase schools' use of NSLDS, the Department has
provided training assistance to schools and has worked to ensure the
accuracy of the system's data.  The Department recently expanded its
NSLDS customer service center and will offer NSLDS training to users
at its 11 regional training centers.  In addition, to demonstrate its
commitment to improving the reliability of data on its postsecondary
education programs, the Department has addressed the issue of data
integrity in its long-range strategic and annual performance plans
prepared in response to the Government Performance and Results Act of
1993.  As part of this commitment, the Department has initiated
efforts to identify and correct inaccurate data in NSLDS, such as
identifying and eliminating duplicate loan records, and to strengthen
its working relationships with other data providers, such as guaranty
agencies. 


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

The Department of Education administers four major student financial
aid programs under title IV of HEA:  FFELP, the William D.  Ford
Direct Loan Program (FDLP), the Federal Pell Grant Program, and the
Federal Campus-Based Programs.\7 These programs together will make
available an estimated $47 billion to about 8 million individuals
during the 1998-99 academic year--about 80 percent of it in student
loans. 

Schools are responsible for obtaining and evaluating the financial
aid history of students to ensure that students are eligible for aid. 
During our review, we identified 10 tasks inherent to administering
federal student aid programs through NSLDS or other methods.  These
tasks are related to four general processes and functions operational
at the time of our survey that the Department made available to
schools through NSLDS to help them administer the student financial
aid programs more effectively.  (See table 1.)



                                     Table 1
                     
                      Relationship Among NSLDS' 4 Processes
                      and Functions and 10 Tasks Inherent to
                        Administering Federal Student Aid
                                     Programs

                                         Processes and functions
                          ------------------------------------------------------
                          Prescreen                              Obtain
                          for          Track        Update       financial aid
Tasks                     eligibility  borrowers    SSCRs        transcripts\a
------------------------  -----------  -----------  -----------  ---------------
Determine student's       X            X            X            X
enrollment status

Determine if student      X                                      X
reached aid limits
(annual, cumulative, or
both)

Determine if student has  X            X                         X
prior loan defaults

Provide or update         X
Perkins loan data\b

Provide or update SSCR                              X
data\c

Locate student borrowers               X                         X
to resolve problems

Identify loan status (or  X            X                         X
obtain information about
individual loans)

Identify lenders, third-  X            X                         X
party servicers, and
guaranty agencies

Obtain financial aid      X                                      X
transcripts (for
students who have
attended other schools)

Correct or update                                   X
student information
--------------------------------------------------------------------------------
\a Financial aid transcripts summarize all previous student financial
aid a student has received.  They are reviewed by school financial
aid administrators to determine, for example, the student's current
level of aid or identify any information that would prevent awarding
aid to an enrolled or enrolling student. 

\b The Federal Perkins Loan Program, one of three campus-based
programs, provides low-interest, long-term loans made through
institutional financial aid offices to help needy undergraduate and
graduate students pay postsecondary educational costs. 

\c Enrollment status reporting is critical for effectively
administering student financial aid loans because a borrower's
enrollment date and status determine his or her repayment date,
deferment privileges, and grace periods, as well as the government's
payment of interest subsidies. 

Once the school determines that a student is eligible, financial aid
funds are disbursed to the student according to program requirements. 

To support--as well as monitor--these student loan programs, the
Department has developed a number of automated processing systems,
including NSLDS.  The budget for these systems is expected to be
about $378 million for fiscal year 1999.  (See app.  III for
descriptions of the major student financial aid systems.) Prior to
NSLDS, the Department relied on a system--commonly referred to as the
guaranty agency tape dump--to collect selected information from
guaranty agencies on each federal student loan.  The tape dump,
developed in the late 1970s, was initially intended to be used by the
Department primarily as an annual source of data for analysis of
program trends.  According to the Department, it did not expect that
every guaranty agency would have historically collected all the data
requested because each agency's system was designed to meet the needs
of that individual guaranty agency.  The tape dump, according to the
Department, was not designed to be used, for example, to prevent
awarding loans to ineligible borrowers. 

Under 1986 HEA amendments, the Secretary of Education was authorized
to replace the tape dump and develop a computer system that would
make national student loan data accessible to guaranty agencies;
however, the Department could not require guaranty agencies to use
the database before approving new loans.  As a result, planning for
the development of the new NSLDS was delayed for several years, when
the Department was allowed, under the Omnibus Budget Reconciliation
Act of 1989, to require guaranty agencies to use the system in
determining student eligibility. 

The scope of the database expanded when 1992 HEA amendments required
the Department to integrate a national student loan database with
other financial aid data systems.  The 1992 amendments also stated
that the Secretary of Education, in establishing a national database,
should give priority to providing information on student enrollment
and status, current loan holders, and servicers.  In response to
these legislative mandates, in January 1993, the Department awarded a
5-year, $39 million contract to develop and maintain NSLDS.  As of
March 1998, the costs for developing, implementing, and maintaining
the system have totaled $96.5 million. 

According to the Department, NSLDS contained about 118 million loan
and grant records as of February 1998.  These records were provided
by guaranty agencies for the FFELP loans they guaranteed, by the
contractor that services FDLP loans for the Department, and by
schools for Pell grants and campus-based aid they awarded.  As figure
1 illustrates, a significant portion of data stored in NSLDS--about
70 percent--related to FFELP loans as of March 1998. 

   Figure 1:  Approximate Number
   and Percentage of Loan and
   Grant Records in NSLDS, by
   Student Aid Program, as of
   March 1998

   (See figure in printed
   edition.)

Source:  Department of Education. 

The Department's student financial aid data systems have suffered
from data quality problems.  For example, in 1996, the Department's
Office of Inspector General (OIG) found in its review of the
September 30, 1992, tape dump that the number of FFELP loans in
repayment was overstated by approximately 5.9 million loans.\8 OIG
also found that a significant number of loans that were incorrectly
recorded in the tape dump remained incorrect in NSLDS, affecting the
reliability of the new system.  As we reported in 1997, poor quality
and unreliable FFELP loan data remain in the Department's systems,
and inaccurate loan data were being entered into NSLDS.\9 As a
result, the Department cannot obtain complete, accurate, and reliable
FFELP data, which, according to its OIG, hinders the Department's
effort to monitor borrowers and properly award aid to those who are
eligible.  The Department acknowledges that its student financial aid
data systems have suffered from data quality problems and has
initiated a number of actions to address data accuracy and integrity
issues. 

Regardless of its weaknesses, NSLDS is the most comprehensive
departmental database on federal student loans that schools, lenders,
and guaranty agencies can use.  The Department has stated that it
expected NSLDS to significantly reduce schools' and students'
administrative burden of applying for and accounting for financial
aid.  With the exception of mid-year transfer students, the
Department has not required schools to obtain paper financial aid
transcripts since the 1996-97 award year.  The Department also
envisioned that NSLDS would simplify and enhance the process of
updating SSCR information for schools.  Prior to NSLDS, schools, for
their FFELP loans, received SSCR rosters from every guaranty agency
that guaranteed their student loans and had to manually verify and
resubmit the rosters to the guaranty agencies.  In addition, the
Department believes that NSLDS has led to considerable improvements
in identifying ineligible student aid recipients.  It estimates that
since the 1994-95 academic year, for example, NSLDS' improved default
matching capabilities may have prevented over $1 billion from being
awarded to ineligible students.\10


--------------------
\7 The Federal Campus-Based Programs--so named because each school is
allocated funds for the award year based on the anticipated financial
need of its student body--are (1) the Federal Work-Study Program, (2)
the Federal Perkins Loan Program, and (3) the Federal Supplemental
Educational Opportunity Grant (SEOG) Program. 

\8 Office of Inspector General, The Department Should Continue Its
Efforts to Improve the Accuracy of Its Student Loan Database,
ACN-A09-38058 (San Francisco, Calif.:  Department of Education, June
14, 1996). 

\9 High Risk Series:  Student Financial Aid (GAO/HR-97-11, Feb. 
1997). 

\10 The Department acknowledges that this estimate is probably high
because schools are able to determine that some defaults and Pell
grant overpayments identified in the matches have been resolved and
some schools have not made proper student eligibility override
determinations.  We did not verify the data used by the Department to
make this estimate. 


   SCHOOLS' USE OF NSLDS' ON-LINE
   AND BATCH PROCESSING FUNCTIONS
   IS LIMITED
------------------------------------------------------------ Letter :3

Our survey results indicate that a significant proportion--42
percent--of the schools participating in federal student financial
aid programs were not using NSLDS' on-line and batch processing
functions, which first became available in November 1994.  In
addition, of the schools that were using these functions, most were
not routinely using them for many of the tasks they were capable of
performing. 

Based on our survey results, we estimate that 58 percent of the 6,181
schools participating in student financial aid programs used NSLDS'
on-line or batch processing functions.  (See fig.  2.) Of the schools
that were not using these functions, they either had the ability to
use it and simply had not done so (19 percent) or did not have the
ability to use the system (23 percent). 

   Figure 2:  Schools' Use of
   NSLDS' On-Line and Batch
   Processing Functions

   (See figure in printed
   edition.)

Source:  GAO survey. 

When we asked schools about their use of NSLDS to accomplish the 10
tasks we identified as inherent to program administration, we found
that schools were not routinely using NSLDS for most of them.  As
shown in figure 3, more than half of the schools rarely or never
performed 7 of the 10 tasks using NSLDS. 

   Figure 3:  Percentage of
   Schools Using NSLDS' On-Line
   and Batch Processing Functions
   "Rarely" or "Never" for 10
   Surveyed Tasks

   (See figure in printed
   edition.)

Source:  GAO survey. 

The only task that a majority of schools routinely used NSLDS to
accomplish was providing or updating SSCR information, which since
February 1997, the Department has required schools--or third-party
servicers on their behalf--to perform on the system.  Specifically,
our survey data show that 69 percent of the schools always or most of
the time use NSLDS for SSCR processing.  An additional 11 percent
occasionally use NSLDS for SSCR processing, and another 5 percent
rarely use the system for this task.  Of the remaining schools--those
that never use NSLDS for SSCR processing--13 percent process SSCRs
through the Clearinghouse or other third-party servicers; the last 2
percent appear not to be meeting the requirement.  The schools' next
most common uses of NSLDS' on-line and batch processing functions
were for correcting or updating student information (33 percent) and
determining prior loan defaults (30 percent).  However, 43 and 47
percent of schools responded that they rarely or never use NSLDS for
these two tasks, respectively. 

The reason schools most frequently gave for rarely or never using
NSLDS to perform any of the 10 tasks was that they use the
Department's Student Aid Reports (SAR) or Institutional Student
Information Records (ISIR).\11 Schools can identify loan status and
determine student eligibility from information contained in SAR or
ISIR, including prior defaults, types of loans, default dates, and
outstanding balances.  Other reasons schools often gave are that they
used the Clearinghouse for SSCR processing, they did not know how to
use NSLDS, and that they had experienced problems using the system. 


--------------------
\11 ISIRs and SARs contain the same information in different formats. 
SAR is used to record the family's financial and other information as
reported by the student on the application for financial aid.  The
Department's Central Processing System generates SARs, which are
mailed directly to students.  The Central Processing System
electronically sends ISIRs to the schools students identify on their
financial aid applications. 


   SCHOOLS ENCOUNTERED FEW
   PROBLEMS OR DATA INACCURACIES
   USING NSLDS AND ARE SATISFIED
   WITH TRAINING AND CUSTOMER
   SERVICE
------------------------------------------------------------ Letter :4

When we asked schools that were using NSLDS' on-line and batch
processing functions whether they experienced problems, such as
transmitting or receiving data, or whether they encountered
inaccuracies in NSLDS data, many reported that they rarely or never
experienced problems or encountered inaccuracies in the data fields
we specified.  Many schools also responded that they are satisfied
with the support they received from training and customer service. 

Some schools, however, responded that they experienced problems more
frequently with certain functions, including correcting and updating
SSCR information or understanding error messages.  Some schools said
they occasionally found inaccuracies in data fields critical to
correctly identifying students--such as Social Security numbers, last
names, and date of birth--and other critical data fields, including
enrollment status.  While school responses indicate that these
problems may not be widespread, they suggest that NSLDS does not yet
offer the level of data accuracy expected by the Department. 


      SCHOOLS RARELY OR NEVER
      EXPERIENCED PROBLEMS WITH
      MOST ASPECTS OF NSLDS, BUT
      SOME PROBLEMS DID EXIST
---------------------------------------------------------- Letter :4.1

In general, schools' experiences with most aspects of using NSLDS
have been essentially problem free.  (See fig.  4.) For example, 85
percent of schools that use NSLDS rarely or never experienced
problems identifying multiple entries in the system of the same loans
or grants. 

   Figure 4:  Percentage of
   Schools Experiencing Problems
   Using NSLDS

   (See figure in printed
   edition.)

Source:  GAO survey. 

However, some schools did experience problems with some aspects of
the system.  For example, about 23 percent always or most of the time
had a problem correcting or updating SSCR information, a critical
tool for effectively administering student financial aid.  As we
reported in 1997, some FDLP student loan borrowers have not started
to repay their student loans after they are no longer enrolled in
school.\12 Reasons given were schools' failing to report enrollment
changes to the Department or the Department's failing to accurately
or promptly record the reported enrollment changes.  Failure to
record a borrower's enrollment changes or status may result in
borrowers not promptly repaying their student loans or in the loans
becoming delinquent, increasing the likelihood of defaulting. 

Although less than 10 percent of the schools found inaccuracies in
NSLDS data fields always or most of the time, 7 to 29 percent of the
schools occasionally found inaccuracies in these fields.  (See fig. 
5.) As we reported in 1997, recording a student's correct Social
Security number, which the Department considers its common student
identifier, is critical for ensuring that aid is awarded to the
correct individual and for identifying an individual's data
records.\13

   Figure 5:  Percentage of
   Schools Encountering Data
   Inaccuracies in NSLDS

   (See figure in printed
   edition.)

Source:  GAO survey. 

NSLDS as well as other Department information systems also use
combinations of unique elements--such as a student's date of birth
and the first two to three letters of the first or last name--to
identify, access, and update a specific student's record.  Therefore,
the Department's information systems depend on these data fields to
be as accurate as possible. 

School responses indicate that inaccuracies in the data fields do not
appear to be widespread, but they suggest that NSLDS does not yet
offer the level of data accuracy expected by the Department.  The
Department depends on guaranty agencies, loan servicers, and other
entities to provide many of these data and has initiated efforts to
address these problems. 


--------------------
\12 Reporting of Student Loan Enrollment Status (GAO/HEHS-97-44R,
Feb.  6, 1997). 

\13 GAO/AIMD-97-122, July 29, 1997. 


      SCHOOLS THAT USE NSLDS ARE
      MORE SATISFIED THAN
      DISSATISFIED WITH THE
      SUPPORT FROM TRAINING AND
      CUSTOMER SERVICE THEY
      RECEIVED
---------------------------------------------------------- Letter :4.2

In preparing schools for NSLDS' implementation, the Department sent
training materials to each school, made training sessions available
to them, and established a customer service center to respond to
questions and otherwise assist schools.  Seventy-nine percent of
schools using NSLDS said that they or their servicers received
training or training materials from the Department.  As figure 6
shows, 44 percent of the schools that received training or training
materials were satisfied with the training, about 30 percent were
neither satisfied nor dissatisfied, and 27 percent were dissatisfied. 

   Figure 6:  Satisfaction of
   Schools Using NSLDS With
   Department-Provided NSLDS
   Training

   (See figure in printed
   edition.)

Source:  GAO survey. 

About 80 percent of schools that use the system responded that they
requested assistance or information from the NSLDS Customer Service
Center.  As figure 7 shows, more than two-thirds of the schools that
received assistance responded that they were satisfied with the
assistance provided by the NSLDS Customer Service Center. 

   Figure 7:  Satisfaction of
   Schools Using NSLDS With the
   Department's NSLDS Customer
   Service Assistance

   (See figure in printed
   edition.)

Source:  GAO survey. 


   SCHOOLS HAD MIXED VIEWS ABOUT
   WHETHER NSLDS IMPROVED STUDENT
   FINANCIAL AID ADMINISTRATION
------------------------------------------------------------ Letter :5

Schools using NSLDS' on-line and batch processing functions had mixed
views about whether these capabilities led to improvements in their
ability to administer federal student financial aid programs. 
Although some schools identified several areas where NSLDS had
improved their program administration, other schools thought NSLDS
did not lead to improvements or make a noticeable difference in other
areas.  For example, half or more of the schools agreed or strongly
agreed that NSLDS had improved the availability of student data, made
student data easier to access, and reduced the amount of paper
handled in administering financial aid programs.  (See table 2.)
However, a significant percentage of schools disagreed or strongly
disagreed that NSLDS had reduced the staff required to administer
student financial aid programs (49 percent) or reduced the time
required to administer student financial aid (39 percent). 



                          Table 2
          
            Views of Schools That Used NSLDS on
              Administrative Benefits of NSLDS

                      Agree or       Neither   Disagree or
                      strongly     agree nor      strongly
NSLDS benefits           agree      disagree      disagree
----------------  ------------  ------------  ------------
Improved                   61%           23%           11%
 availability of
 student data
Made student                59            22            15
 data easier to
 access
Reduced                     50            23            25
 paperwork
Improved quality            47            31            16
 of student data
Improved                    46            31            17
 exchange of
 information
 with other
 schools
Improved                    42            34            19
 school's
 management and
 oversight of
 title IV
 programs
Improved ability            36            35            21
 to resolve loan
 problems
Reduced time                31            27            39
 required for
 title IV
 administration
Reduced fraud               30            41            16
 and abuse at
 school
Improved ability            21            49            19
 to resolve
 grant problems
Reduced staff               15            33            49
 required for
 title IV
 administration
----------------------------------------------------------
Note:  When asked to what extent they agreed or disagreed that NSLDS
benefited their school, some schools responded "don't know." We did
not include the percentages for this category; therefore, responses
may not add up to 100 percent. 

Source:  GAO survey. 


   SCHOOLS HAD A VARIETY OF
   REASONS FOR NOT USING NSLDS'
   ON-LINE AND BATCH PROCESSING
   FUNCTIONS
------------------------------------------------------------ Letter :6

Our survey results suggest that schools have a variety of reasons for
not using NSLDS' on-line and batch processing functions. 
Twenty-three percent of all schools lacked the ability to access the
system, and 19 percent of all schools had the ability to do so but
did not. 

We estimate that--at the time of our survey--more than 1,000 schools
lacked the ability to access the system; of these, a number planned
to obtain access within 8 months.  However, most had no plans to
obtain access to NSLDS, had plans to obtain access but did not know
when, or were unsure whether they would ever obtain access.  Most of
these schools were using the Clearinghouse or other third-party
servicers to perform tasks, but several hundred schools appeared not
to be accessing NSLDS either directly from their campus or through
the Clearinghouse or other third-party servicer.  Since these schools
did not have plans to obtain access, they may not be meeting a
requirement to have on-line access to NSLDS from their campuses as of
January 1998. 

While the schools that did not use NSLDS' on-line and batch
processing functions--either because they chose not to (19 percent)
or did not have the ability to (23 percent)--had a variety of reasons
for not using or having NSLDS, these reasons generally fell into one
of three categories:  use of alternative methods, such as the
Clearinghouse, for obtaining and processing SSCR information;
limitations in resources, personnel, or skills, such as lack of
training; and lack of confidence in data reliability.  (See fig.  8.)

   Figure 8:  Reasons Non-Users
   Said They Did Not Use NSLDS'
   On-Line and Batch Processing
   Functions

   (See figure in printed
   edition.)

Note:  "Other" reasons include "school will be closing" and "school
just applied for access to NSLDS and is awaiting Department approval
to use the system."

Source:  GAO survey. 


   THE DEPARTMENT HAS TAKEN STEPS
   TO PROMOTE GREATER SCHOOL USE
   OF NSLDS
------------------------------------------------------------ Letter :7

The Department has efforts under way to increase schools' use of
NSLDS.  While a small percentage of schools have not complied with
requirements for SSCR processing and on-line access, the Department
recognizes that to encourage schools to use NSLDS beyond these
requirements, it must promote the system and ensure the accuracy of
the system's data.  Therefore, the Department plans to provide
additional NSLDS training to users at sites throughout the country
and review and correct any inaccurate data in NSLDS.  Although these
actions were not developed specifically in response to the concerns
schools had cited in our survey, they address many of the issues
schools raised. 


      ACTIONS TO INCREASE
      COMPLIANCE WITH REQUIREMENTS
      FOR USING NSLDS ARE IN PLACE
---------------------------------------------------------- Letter :7.1

Schools are required to use NSLDS for SSCR processing.  They may do
so directly from their campuses or through the Clearinghouse or other
third-party servicers.  Our survey shows that 5 percent of all
schools are not meeting this requirement.\14 The Department has
independently identified schools that have not complied with the
requirement and taken a number of actions to increase compliance. 
For example, it has assessed fines ranging from $1,000 to $7,500
against 19 schools. 

As of January 1998, schools are also required to have on-line access
to NSLDS for their financial aid staff.  The Department has estimated
that 2 percent of all participating schools have not registered for
on-line access.  The contractor responsible for operating NSLDS is
contacting these schools at the Department's instruction to attempt
to get them to register. 


--------------------
\14 This includes 2 percent of the schools that use NSLDS for
purposes other than SSCR reporting and 8 percent of schools that do
not use NSLDS at all. 


      ADDITIONAL NSLDS TRAINING
      HAS BEEN SCHEDULED
---------------------------------------------------------- Letter :7.2

The Department plans to offer more training to increase participation
among those schools that do not access NSLDS.  According to
Department officials, the Department already offers a variety of
training opportunities and has no plans to alter the types of
training it offers but will increase the number of training sessions. 
In February 1998, the Department began offering a series of workshops
aimed at helping schools automate their financial aid offices.  These
workshops include components on using NSLDS.\15 More recently, the
Department began offering computer-based training sessions solely on
how to use NSLDS.  The Department expects that between August and
October 1998, it will have offered 39 NSLDS sessions--each
accommodating 40 participants--at its 11 regional training centers. 
If there is demand and sufficient funding, the Department plans to
offer more training sessions in spring 1999.  For school personnel
who cannot travel to a training site, the officials said schools can
use a self-paced computer program included with the NSLDS users'
manual. 


--------------------
\15 Department officials had expected that by the end of July 1998,
the Department would have held workshops at 100 sites around the
country, training 7,000 to 8,000 lender, guaranty agency, and school
personnel; however, on July 23, the Department announced it had
canceled or was canceling more than 20 of these workshops due to low
enrollment. 


      DATA ACCURACY AND INTEGRITY
      ARE BEING ADDRESSED
---------------------------------------------------------- Letter :7.3

To ensure the overall accuracy of NSLDS data, the Department has
initiated "serious and aggressive" data integrity efforts to review
and correct any inaccurate or incomplete data, according to
Department officials.  These efforts include providing detailed
technical instructions to data originators and providers (schools,
lenders, and guaranty agencies); focusing on correcting inaccurate
data; and addressing the issue of data quality in its annual
performance plan. 

The Department gave the major data providers detailed technical
instructions that specify their responsibilities and the data they
are required to provide.  According to Department officials, data
providers must correct data problems as quickly as possible,
especially since they are the ones that usually identify the
problems.  Correcting data problems may require system changes or
major data entry and research.  The Department provides technical
update bulletins on an as-needed basis, training sessions, and
on-site technical reviews.  Further, the Department tracks whether
providers have corrected the data problems and sends monthly
management reports to all providers.  If a significant problem occurs
when two or more data providers have conflicting information each
believes is correct, the dispute must be formally adjudicated within
the Department. 

According to Department officials, to increase both voluntary use of
and school satisfaction with NSLDS, schools must have confidence in
the system's data.  The Department's strategy focuses on increasing
the accuracy of loan data as well as strengthening relationships with
the financial aid community.  For example, the Department's data
improvement efforts focus on deleting duplicate loans and loans with
a "zero balance" (that is, loans that have been paid in full);
identifying loans not in NSLDS that should be; resolving differences
between NSLDS and lender databases; and conducting outreach efforts,
such as holding regular workshops with other data providers.  The
Department is now taking these measures to address problems with
FFELP and Perkins loan data.  Department officials said that problems
with FDLP data have been largely corrected. 

To further demonstrate its commitment to improving the reliability of
data on its financial aid programs, the Department has addressed the
issue of data integrity in its long-range strategic and annual
performance plans prepared in response to the Government Performance
and Results Act of 1993.\16 For example, in its fiscal year 1999
performance plan--its first annual plan--the Department acknowledges
that its student financial aid delivery system has suffered from data
quality problems that are severe enough to cause it to fail to
receive an unqualified audit opinion.  In its June 15, 1998, audit
report, the Department's OIG reported that, in its opinion, the
consolidated financial statements present fairly, in all material
respects, the financial position of the Department as of September
30, 1997.\17 However, the reliability of data in NSLDS is still a
material internal control weakness.  Specifically, the audit report
states that the Department's ability to continue to prepare auditable
loan estimates for its financial statements depends on establishing a
reliable store of up-to-date historical loan data.  The audit report
notes that because of questionable data in NSLDS, the estimated
liability for loan guarantees was based on data received from 10
large guaranty agencies, as opposed to NSLDS. 

According to the Department's performance plan, steps are being taken
to improve the efficiency and quality of its student aid data.  These
include

  -- improving data accuracy by receiving individual student loan
     data directly from lenders rather than through guaranty agencies
     and by expanding efforts to verify the data reported to NSLDS
     and

  -- preparing a system architecture for the delivery of federal
     student aid by December 1998 that will help integrate the
     multiple student aid databases with NSLDS based on student-level
     data to improve the availability and quality of information on
     student aid applicants and recipients. 


--------------------
\16 The Results Act is the primary legislative framework through
which federal agencies are being required to set strategic goals,
measure performance, and report on the degree to which their goals
were met. 

\17 Office of Inspector General, Final Audit Reports:  Fiscal Year
1997 Annual Financial Statement, ACN-17-70002 (Washington, D.C.: 
Department of Education, June 15, 1998). 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

The Department of Education provided written comments on a draft of
this report in a letter from the Assistant Secretary for
Postsecondary Education, dated August 14, 1998.  Overall, the
Department felt that the draft report was not balanced or written
fairly and that it did not accurately portray the full extent of
schools' use of NSLDS.  The comments expressed three specific
concerns:  (1) inappropriate emphasis placed on negative aspects of
the survey results, (2) numerically distorted survey results, and (3)
outdated audit report information in the background section that had
no relation to the scope of the audit. 

In addressing the Department's comments, we revised the report, as
appropriate, to clarify that our review focused on schools' use of
NSLDS' on-line and batch processing functions rather than all school
uses of NSLDS.  To address the specific concerns about our
presentation of survey results, we made other revisions, where
appropriate, to better ensure that our results were presented
objectively and fairly.  However, we continue to focus the reader's
attention on those responses that show the extent to which schools
were encountering shortcomings in using NSLDS' on-line and batch
processing functions.  Finally, to address the concern about our use
of previous audit reports, we added information on more recent
activities, such as the Department's efforts to improve the quality
of NSLDS data.  However, we retained a discussion of previous audit
reports because we believe it is needed to establish the historical
context and significance for creating NSLDS. 

Department officials discussed these and other concerns and provided
other comments in meetings with our staff on July 31 and August 7,
1998.  (See app.  IV for a discussion of these comments and our
responses and a reprint of the Assistant Secretary's August 14, 1998,
letter.)


---------------------------------------------------------- Letter :8.1

We are sending copies of this report to the Secretary of Education,
appropriate congressional committees, and other interested parties. 
We also will make copies available to others on request. 

If you have any questions about this report, please contact me at
(202) 512-7104.  Major contributors to this report are listed in
appendix V. 

Sincerely yours,

Carlotta C.  Joyner
Director, Education and
 Employment Issues


RESPONSES TO QUESTIONS FROM GAO
SURVEY ON NSLDS USE
=========================================================== Appendix I

This appendix contains schools' responses to our survey.  All numbers
are percentages, except for those in questions 2, 4, 7, 14, 27, 31,
and 33.  Percentages shown are based on the number of respondents
answering each question.  Percentages may not always add to 100
percent due to rounding. 



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


SCOPE AND METHODOLOGY
========================================================== Appendix II

To address our objectives, we (1) surveyed a random sample of schools
that participate in financial aid programs, (2) reviewed pertinent
documents, and (3) spoke with Department of Education officials and
members of the higher education community. 

To obtain information on schools' use of NSLDS, including problems
encountered and benefits derived, we surveyed a random sample of
postsecondary colleges and universities that participated in federal
student financial aid programs as of August 1997.  During the
development of the survey instrument, Department officials informed
us that all schools use NSLDS because data they receive on SARs and
ISIRs are generated from the system; therefore, we focused our survey
questions on the use of NSLDS' on-line and batch processing
functions.  In addition, we reviewed NSLDS documents obtained from
the Department and prior GAO reports. 

To identify measures undertaken by the Department to ensure schools'
use of the system, we interviewed officials from the Department as
well as staff from Raytheon/E-Systems, the contractor responsible for
developing and maintaining NSLDS.  We discussed NSLDS' current
operation, including the functions available for school use, the
number of schools that have access to NSLDS, methods by which
information is transmitted into NSLDS, and how users gain access to
this information. 

To aid in designing our survey instrument, we also contacted members
of the higher education community.  We interviewed officials from the
American Association of State Colleges and Universities, the American
Council on Education, the Coalition of Higher Education Assistance
Organizations, the National Association of Independent Colleges and
Universities, the National Association of Student Financial Aid
Administrators, and the National Student Loan Clearinghouse. 

We conducted our study between May 1997 and July 1998 in accordance
with generally accepted government auditing standards. 

SURVEY DESIGN

To determine the extent of schools' use of NSLDS' on-line and batch
processing functions, we developed a survey instrument and sent it to
a randomly selected sample of postsecondary schools.  The survey
covered a variety of topics, including descriptive background data on
each school, types of access to NSLDS, and actual experiences using
NSLDS' on-line and batch processing functions.  For example, we asked
schools to identify themselves as public or private and to provide
the size of their student body.  We also asked the schools that
reported having access to NSLDS whether they used the system
themselves or through a third-party servicer.  In addition, we asked
schools to provide information on their use of various features of
NSLDS, including benefits derived and problems encountered.  To
determine the purposes for which schools use NSLDS, we reviewed NSLDS
manuals and pretested our instrument with 12 schools. 


      SURVEY UNIVERSE AND RESPONSE
------------------------------------------------------ Appendix II:0.1

We drew our sample of 600 postsecondary schools from 6,181 schools
listed in the Department's automated Postsecondary Education
Participants System (PEPS) as of August 1997.\18 We mailed our
instrument to the 600 schools in November 1997.  We did a follow-up
mailing in December 1997 and again in January 1998. 

Of the 600 schools, we determined that 11 were ineligible for our
survey--because they no longer participated in federal student aid
programs, were high schools rather than postsecondary schools, or had
closed--resulting in an adjusted sample of 589 schools.  Of these,
490 schools returned completed, usable survey instruments, which
yielded a school response rate of 83 percent. 

Our analyses are based on the 490 responses from 83 percent of the
eligible schools sampled.  All data are self-reported, and we did not
independently verify their accuracy. 


--------------------
\18 According to Department officials, PEPS is the Department's
database on the universe of schools participating in federal student
financial aid programs. 


      NSLDS PROCESSES AND
      FUNCTIONS INCLUDED IN OUR
      SURVEY
------------------------------------------------------ Appendix II:0.2

We included in our survey four of the six NSLDS processes and
functions available to schools (listed in the Department's NSLDS
users' manual, NSLDS:  The Paperless Link):  prescreening for
eligibility, borrower tracking, SSCR data, and financial aid
transcripts.  We did not include the remaining two
functions--overpayment data and report selection--because the first
was not available at the time of our review and the latter duplicated
the four processes and functions included in our survey. 

  -- Prescreening for eligibility:  The prescreening function allows
     schools to receive data on prior student financial aid
     recipients, enabling schools to determine the eligibility of
     financial aid applicants before funds are awarded and thereby
     reduce defaults. 

  -- Borrower tracking:  This on-line NSLDS function is generally
     used by loan servicers and guaranty agencies attempting to
     locate a borrower who has defaulted on a student loan.  NSLDS
     provides data on other organizations (such as schools and
     lenders) associated with the borrower, which servicers and
     agencies can contact to obtain the borrower's current address. 

  -- SSCR:  Schools are required to use this function to confirm and
     report the enrollment status of students who receive federal
     loans. 

  -- Financial aid transcript:  NSLDS' financial aid transcript
     function summarizes all previous title IV financial aid a
     student has received.  Histories are received on students
     currently attending or transferring to an institution. 
     Financial aid transcripts are reviewed by a financial aid
     administrator to determine current levels of aid, whether there
     is any information that would prevent awarding aid for the first
     time, or to continue aid to an enrolled or enrolling student. 

  -- Overpayment:  NSLDS' overpayment function--added since our
     survey--will enable schools to notify NSLDS that a student owes
     a refund of an overpayment on a Pell grant, State Student
     Incentives Grant, or SEOG grant, as well as a Perkins Loan.  An
     overpayment notification to NSLDS notifies the entire student
     financial aid community because the actual overpayment data
     appear on all financial aid transcripts that are requested
     through NSLDS and through prescreenings of ISIRs and SARs. 

  -- Report selection:  Reports and extracts are produced by NSLDS on
     both a regularly scheduled and on-request basis.  Schools may
     query the system regarding the existence of reports, extracts,
     or both and may gain access to them via an on-line display or a
     file deposited to their wide area network mailbox. 


      IDENTIFIED ADMINISTRATIVE
      TASKS INCLUDED IN THE SURVEY
------------------------------------------------------ Appendix II:0.3

During the pretest, we identified 10 tasks inherent to administering
federal student aid programs.  Schools can perform these tasks by
using NSLDS' on-line and batch processing functions directly or using
other methods.  We also identified operations during which schools
might encounter problems using the system and data fields in which
they might encounter inaccuracies. 

In the survey, we asked users of the system to identify the tasks for
which they used NSLDS and any problems they had encountered,
including what kinds of inaccuracies, if any, they had found in the
data. 


      RESPONDENT USE OF NSLDS
------------------------------------------------------ Appendix II:0.4

For the purposes of this survey, we defined use of NSLDS as accessing
it through on-line or batch processing functions designed to perform
specific tasks.  To focus on the direct use of these functional
capabilities by school personnel, we instructed school officials
completing the survey instrument not to consider their use of student
eligibility information reports generated from the NSLDS database and
sent by the Department to be a use of the system.  Similarly, if
schools only used the Clearinghouse for processing SSCR reports and
did not use any of the system's other functional capabilities, we did
not consider those schools to be NSLDS users. 

We directed our survey to the officials at the selected schools whom
we determined to be the most knowledgeable about NSLDS use and
federal student financial aid programs.  To identify the appropriate
respondent at each sample school, we sent a letter to the Director of
Financial Aid, which both alerted the school to our survey and
requested that the school return a postcard with the name and address
of the appropriate recipient, if different from the financial aid
office. 

SAMPLING AND NONSAMPLING ERRORS

All sample surveys are subject to sampling errors, that is, the
extent to which the results differ from what would be obtained if the
whole population had received the survey instrument.  Since the whole
population does not receive the instrument in a sample survey, the
true size of this difference cannot be known.  However, it can be
estimated from the responses to the survey. 

Using the number of respondents and the amount of variability in the
data, we were able to estimate sampling errors for our survey.  (See
table II.1.)



                               Table II.1
                
                       Estimated Sampling Errors

                                                                Margin
                                                                    of
                                                                 error
                                                                (perce
Sampling error                                                     nt)
--------------------------------------------------------------  ------
Schools participating in student financial aid programs            ï¿½ 4
Schools responding that they use NSLDS                             ï¿½ 6
Schools responding that they do not use NSLDS                      ï¿½ 7
----------------------------------------------------------------------
Note:  Estimates are at 95-percent confidence level. 

In addition to sampling errors, surveys are subject to other types of
systematic error or bias that can affect results.  Bias can affect
both response rates and the way respondents answer particular
questions.  We cannot assess the magnitude of the effect of bias, if
any, on our survey results.  Rather, possibilities of bias can only
be identified and accounted for when interpreting results.  One
possible source of bias in our survey is inherent in all self-ratings
and self-reports.  Bias inherent in self-rating and self-reporting
may impact survey results because integrity of the data depends upon
respondents providing honest and accurate answers to survey
questions.  The results of this report are affected by the extent to
which respondents accurately reported their school's use or non-use
of NSLDS. 

We took several steps to minimize the impact of nonsampling errors. 
First, we examined responses for extreme values and inconsistencies. 
In a few cases, respondents had reported numbers incorrectly, and in
these cases, we corrected the data or, if correction was not
possible, we rejected the data known to be in error. 


Appendix III MAJOR STUDENT
FINANCIAL AID SYSTEMS
========================================================== Appendix II

There are six major student financial aid systems that provide
various types of information on student loans. 


      CAMPUS-BASED PROGRAMS SYSTEM
------------------------------------------------------ Appendix II:0.5

The Campus-Based Programs System supports all data tracking and
reporting functions associated with campus-based programs.  This
system uploads and edits data received from participating schools;
calculates tentative and final school awards, notifying schools of
their award levels; allocates funds; and reconciles school accounts. 
This system contains no student-level information; it uses only
summary data by school. 


      CENTRAL PROCESSING SYSTEM
------------------------------------------------------ Appendix II:0.6

The Central Processing System supports student financial aid
applications and the determination of Pell grant eligibility; matches
other databases for applicant eligibility; makes corrections to the
records; and produces statistical analysis tables, student data
rosters, and tapes for schools and state agencies. 


      DIRECT LOAN SERVICING SYSTEM
------------------------------------------------------ Appendix II:0.7

The Direct Loan Servicing System services FDLP loans while the
borrower is in school, in deferment status, or in repayment. 


      NATIONAL STUDENT LOAN DATA
      SYSTEM
------------------------------------------------------ Appendix II:0.8

The National Student Loan Data System performs prescreening of
student financial aid program applications, performs student status
confirmation reporting, and tracks borrowers.  The system contains
information regarding loans made, insured, or guaranteed under title
IV and selected Pell grant information.  Its purposes are to (1)
ensure that accurate and complete data on student loan indebtedness
and institutional lending practices are available, (2) screen
applications to identify prior loan defaults and grant overawards,
(3) provide a database to research and identify trends and patterns,
(4) support audits and program reviews, and (5) calculate default
rates. 


      PELL GRANT RECIPIENT AND
      FINANCIAL MANAGEMENT SYSTEM
------------------------------------------------------ Appendix II:0.9

This system receives, evaluates, and processes student payment data
and serves as the basis for obligations to schools. 


      POSTSECONDARY EDUCATION
      PARTICIPANTS SYSTEM
----------------------------------------------------- Appendix II:0.10

This system maintains data on school participation in student
financial aid programs (such as eligibility, certification, address,
and program participation); supports institutional reviewers and
related activities; acts as the official source of information
regarding schools and their associated school codes for all
Department of Education systems; and supports the annual default rate
calculation process for FFELP and FDLP. 


COMMENTS FROM THE DEPARTMENT OF
EDUCATION AND OUR RESPONSES
========================================================== Appendix IV

On July 31 and August 7, 1998, we met with Department of Education
officials to obtain their comments on a draft of this report.  In
general, the Department commented that the survey results show NSLDS
in a favorable light and that it is used universally for important
operational purposes and without significant difficulty.  The
Department believes the report should convey such results. 

Our meetings were supplemented by an August 14, 1998, letter from the
Assistant Secretary for Postsecondary Education, which appears at the
end of this appendix.  Our responses to the comments raised in this
letter are provided in the body of the report.  Summaries of the
comments Department officials provided during our meetings and our
responses to these comments follow. 

1.  The Department said all schools use NSLDS, and that we were
incorrect in reporting that almost half of the schools do not use
NSLDS at all.  In addition to accessing NSLDS through the on-line and
batch processing functions, they said all schools (1) routinely use
NSLDS for prescreening because the SAR or ISIR data they receive for
this purpose are generated by NSLDS and (2) must submit SSCR data to
NSLDS regardless of whether they do so themselves or use a servicer,
such as the Clearinghouse.  When such a servicer provides this
service, it submits SSCR data to NSLDS in lieu of the schools doing
so directly.  Department officials said we failed to point out in our
discussion of the 23 percent of schools that do not access the system
at all that (1) this is due to limitations or constraints the schools
face that are beyond the Department's control and (2) schools use
servicers or a third party to access NSLDS in order to meet SSCR
processing requirements or receive ISIRs.  The officials believe that
the report should have stated that all schools use NSLDS (including
those that access NSLDS indirectly) and schools do not need to use
batch or on-line functions (except SSCR) if they are satisfied that
the information provided by SARs or ISIRs in prescreening meets their
needs. 

GAO's Response:  We recognize that all schools use SAR and ISIR data
generated by NSLDS and that in this way, all schools use NSLDS.  Our
review focused on schools' use of NSLDS' on-line and batch processing
functions, and we have revised the report to further clarify this
focus.  As pointed out elsewhere in our report, the topic of NSLDS
use was discussed extensively during our study design--with schools,
Department officials, and others in the education community.  To help
school representatives in completing the survey, we explained on page
1 of the instrument that we defined "access" as on-line and batch
processing functions (see app.  I). 

With regard to the number of schools we identified that were not
accessing NSLDS at all, we have revised the report to clarify that
these are schools that do not use the on-line or batch functions
either themselves or through a servicer.  As figure 8 shows,
limitations and constraints at schools were among the more frequently
cited reasons given by schools for not having this access.  Figure 8
also shows that using servicers and the Clearinghouse for SSCR
processing were frequently given as reasons for schools not using
NSLDS' on-line and batch processing functions.  We did not
independently obtain evidence on the extent of the Department's
ability to influence these factors and did not draw conclusions about
the Department's role.  Our report, however, discusses how the
surveyed schools are meeting the SSCR reporting requirement. 

With regard to the Department's comment that schools may not need
on-line and batch processing functions when SAR or ISIR data meet
their needs, we did not ask schools for details about their use of
these or any other sources of financial aid data for determining
student eligibility.  It was our intent to find out from schools the
extent of their use of NSLDS' on-line or batch processing
capabilities to perform a variety of tasks, such as borrower tracking
or updating student information.  We did not draw conclusions about
the reported level of schools' use of these capabilities. 

2.  The Department said that favorable responses we received to the
survey instrument were not adequately reported.  Department officials
believe that there were numerous instances throughout our draft
report where we deemphasized favorable results that reflect the
majority of schools and emphasized corresponding unfavorable results. 
They cited as an example a statement in the draft that about
one-fourth of schools had problems using NSLDS to correct or update
SSCR information, and 20 to 30 percent of schools encountered
occasional or frequent inaccuracies in certain data fields.  They
believe a more accurate portrayal of the survey results would be to
state that 75 percent of the schools did not have any problems using
NSLDS for SSCR processing and that 70 to 80 percent rarely or never
encountered data inaccuracies. 

GAO's Response:  Overall, we believe our draft report reasonably
presented the results of our analysis.  However, we made minor
revisions, where appropriate, to further clarify our objectives and
ensure the fairest possible presentation of our results.  For
example, in our discussion of schools' views of the training and
training materials they received, we added a statement that the
Department mailed training materials to every school.  However, we
decided to continue to focus the reader's attention on those
responses that show the extent to which schools encountered
shortcomings in using NSLDS' on-line and batch processing functions
because we believe it is important to note potential problem areas so
the source of the problems may be explored and improvements to the
program or operation can be identified. 

3.  Department officials said that the tone and balance of the draft
were not true to the schools' survey answers.  For example, they said
we cite a number of previous reports and studies illustrating data
problems that they believe are outdated and are not relevant to the
objectives of our review.  If we feel strongly that these studies
should be cited, the officials said we should also report more recent
statistics on default matching, as well as numerous increases in the
functional uses of NSLDS that have been developed.  In addition, they
believe the draft did not adequately address the work undertaken or
under way to improve the quality of NSLDS data. 

GAO's Response:  Our purpose in discussing prior GAO and OIG audit
reports is to provide general background information, establish a
historical context and significance for creating NSLDS, and discuss
the basis for congressional interest in our review.  As Department
officials suggested, we have included more recent statistics on the
Department's use of NSLDS for conducting student loan default matches
and expanded our discussion about its efforts to improve the quality
of student financial aid data. 

4.  Department officials said that the draft report did not fully
discuss the purposes and advantages of NSLDS, and that our treatment
of original NSLDS goals needed clarification.  They believe that it
is important for us to distinguish between the original purpose of
NSLDS and current efforts to expand its functionality.  They
suggested that the report should note that NSLDS was set up as a
research database and that the Department is expanding its use to
help improve the accuracy and availability of student aid data.  More
importantly, they said these improvements will ensure better
accountability for student financial aid monies. 

GAO's Response:  The report identifies the three main goals of NSLDS
as they were presented and distributed to schools in the NSLDS users'
guide.  Our study focused on schools' use of NSLDS' on-line and batch
processing functions; we did not review the full spectrum of NSLDS'
purposes and functions.  The Department's explanation of these new
functions is informative, and it appears that the Department is
enhancing NSLDS to take advantage of many of its expanded
capabilities.  But many of these new functions were either not
available to schools at the time we administered our survey
instrument, were not related to on-line and batch processing
functions, or were not designed to be used by schools. 

5.  Department officials were concerned about the development and
timing of the survey instrument we administered.  They said that it
is not clear which school official completed the survey and that this
is important because different offices in a school may use NSLDS for
different purposes.  Also, since the instrument was administered in
late 1997, they believe some of the results may now be outdated and
inaccurate.  Finally, they believe that the report should note that
the instrument was administered at a time when NSLDS was only 3 years
old and that some portion of schools were not using it as a result of
the time lags that occur in getting all schools to adapt to and
welcome its use. 

GAO's Response:  We made a concerted effort to direct the survey
instrument to those school officials who were most knowledgeable
about their school's use of NSLDS' on-line and batch processing
functions.  As discussed in appendix II, before we mailed the survey
instrument, we sent a postcard to the student financial aid
administrator of each school in our sample asking him or her to
provide the name, title, and complete address of the school official
who was best suited to complete the survey.  About half of the
schools provided this information; for those that did not, we mailed
the survey to the director of student financial aid.  In a letter
accompanying the survey instrument, we further requested schools to
ensure that those persons most knowledgeable about using NSLDS be
involved in responding to the survey instrument. 

The purpose of the survey was to record schools' use of NSLDS at the
time the survey was administered in November 1997, and this is noted
in the report.  We recognize that not all schools had access to NSLDS
when they completed the survey instrument; as our results showed,
many schools were still in the process of obtaining access.  We would
expect that, at this writing, more schools would be accessing NSLDS'
on-line and batch processing functions, but the report is intended to
assess schools' use of NSLDS' on-line and batch processing
capabilities and document the extent of this use at a point in time. 
We do not, nor did we intend to, draw any conclusions as to whether
the level of school use is sufficient or indicative of the long-term
utility of NSLDS as an administrative tool for federal student aid
programs. 



(See figure in printed edition.)

Appendix V GAO CONTACTS AND STAFF ACKNOWLEDGMENTS

GAO CONTACTS

Joseph J.  Eglin, Jr., Assistant Director, (202) 512-7009
Joel R.  Marus, Evaluator-in-Charge, (202) 512-7287

STAFF ACKNOWLEDGMENTS

In addition to those named above, the following individuals made
important contributions to this report:  Paula Denman and Joan A. 
Denomme, Senior Evaluators; Carolyn S.  Blocker and Edward H. 
Tuchman, Evaluators; and Deborah L.  Edwards and Wayne M.  Dow,
Senior Social Science Analysts. 

RELATED GAO PRODUCTS

The Results Act:  Observations on the Department of Education's
Fiscal Year 1999 Annual Performance Plan (GAO/HEHS-98-172R, June 8,
1998). 

Department of Education:  Information Needs Are at the Core of
Management Challenges Facing the Department (GAO/T-HEHS-98-124, Mar. 
24, 1998). 

Student Financial Aid Information:  Systems Architecture Needed to
Improve Programs' Efficiency (GAO/AIMD-97-122, July 29, 1997). 

Department of Education:  Multiple, Nonintegrated Systems Hamper
Management of Student Financial Aid Programs (GAO/T-HEHS/AIMD-97-132,
May 15, 1997). 

Reporting of Student Loan Enrollment Status (GAO/HEHS-97-44R, Feb. 
6, 1997). 

High Risk Series:  Student Financial Aid (GAO/HR-97-11, Feb.  1997). 

Department of Education:  Status of Actions to Improve the Management
of Student Financial Aid (GAO/HEHS-96-143, July 12, 1996). 

Student Financial Aid:  Data Not Fully Utilized to Identify
Inappropriately Awarded Loans and Grants (GAO/T-HEHS-95-199, July 12,
1995). 

Student Financial Aid:  Data Not Fully Utilized to Identify
Inappropriately Awarded Loans and Grants (GAO/HEHS-95-89, July 11,
1995). 

Federal Family Education Loan Information System:  Weak Computer
Controls Increase Risk of Unauthorized Access to Sensitive Data
(GAO/AIMD-95-117, June 12, 1995). 

High Risk Series:  Student Financial Aid (GAO/HR-95-10, Feb.  1995). 

Department of Education:  Management Commitment Needed to Improve
Information Resources Management (GAO/IMTEC-92-17, Apr.  20, 1992). 


*** End of document. ***