oversight

Georgia Department of Education's Controls Over Performance Data Entered in EDFacts

Published by the Department of Education, Office of Inspector General on 2010-04-07.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                                              UNITED STATES DEPARTMENT OF EDUCATION
                                                     OFFICE OF INSPECTOR GENERAL


                                                                                                                                Audit Services
                                                                                                                                   Region IV

                                                                 April 7, 2010


Mrs. Kathy Cox
State Superintendent of Schools
Georgia Department of Education
2066 Twin Towers East
Atlanta, GA 30334

Dear Mrs. Cox:

This final audit report, Control Number ED-OIG/A04J0003, presents the results of our audit titled
Georgia Department of Education’s Controls Over Performance Data Entered in EDFacts. The
objectives of the audit were to 1) determine whether the Georgia Department of Education (GaDOE)
and Clayton County Public School (CCPS) District established adequate systems of internal control to
provide accurate education data to EDFacts, and 2) evaluate GaDOE’s use of program reviews as a
monitoring tool for local educational agencies (LEA). Our audit covered selected CCPS EDFacts data
for the 2006-2007 school year and program reviews performed by GaDOE for the 2006-2007 and
2007-2008 school years.

For a complete list of acronyms/abbreviations used in this report, see Appendix A.


                                                           BACKGROUND


The U.S. Department of Education’s (Department) EDFacts system 1 is a central repository that
consolidates Kindergarten through 12th grade (K-12th) education information collected from 52 State
Educational Agencies (SEA). SEAs submit required data files through the Education Data Exchange
Network (EDEN) Submission System (ESS) 2 component of EDFacts as required by 34 C.F.R. § 80.40
and Federal Register Volume 72. The internet-based collection process simplifies reporting and
improves the timeliness of the K-12th grade information that is required for annual and final grant
reporting, specific program mandates, and the Government Performance and Results Act. EDFacts
data are used for planning, policy, and management at the Federal, State, and local levels.

1
  The system is managed by the Office of Planning, Evaluation, and Policy Development (OPEPD) and was previously
called the Education Data Exchange Network (EDEN).
2
  The EDFacts System has three components – the ESS collects required data submission; the EDEN Survey Tool collects
supplementary data submission, such as data for the Civil Rights Data Collection and the Indian Education Formula Grant
Program Application; and the EDEN staging database holds newly submitted data.


The Department of Education's mission is to promote student achievement and preparation for global competitiveness by fostering educational excellence
                                                             and ensuring equal access.
Final Report
ED-OIG/A04J0003                                                                      Page 2 of 23
The EDFacts initiative was funded in 2003, operational by 2004, and mandated for use by SEAs
starting with the 2006-2007 school year. According to the EDFacts Fiscal Year (FY) 2011 Office of
Management and Budget (OMB) Exhibit 300 submitted to OMB in September 2009, the Department
has spent approximately $78.6 million on the EDFacts system from its inception through FY 2009.

GaDOE oversees public education throughout the State of Georgia, with more than 2,500 schools and
1.6 million students. It ensures that laws and regulations pertaining to education are adhered to and
State and Federal funds appropriated for education are properly allocated to its 182 LEAs. For the
2006-2007 school year, GaDOE allocated more than $375 million of Title I funds to its LEAs.

GaDOE has submitted performance and other data through ESS since the 2003-2004 school year
(reporting 2002-2003 school year data). GaDOE requires LEAs to submit specific data to the Georgia
Statewide Student Information System (GSSIS). Using the Department’s EDEN file specification
documents as guidance, GaDOE extracts selected data from GSSIS and submits it to EDFacts.

CCPS, located in Jonesboro, is the sixth largest school district in Georgia, serving more than 50,000
students in 33 elementary schools, 13 middle schools, and 8 high schools. The chart below lists
Department funded programs that CCPS participates in and the amount it was allocated for each
program for the 2006-2007 school year.

                       CCPS Allocation by Department Program
                             Name of       Amount Allocated to
                           Department      CCPS for 2006-2007
                        Funded Program        School Year
                       Title I-A                   $18,134,586
                       Title II-A                    $2,149,169
                       Title II-B                       $24,206
                       Title II-D                      $453,491
                       Title III-A                     $676,999
                       Title IV-A                      $413,219
                       Title V-A                       $137,786
                       Reading First                 $2,168,529


                                        AUDIT RESULTS


We found that neither CCPS nor GaDOE established adequate systems of internal control to ensure
that accurate, reliable, and complete data were entered in EDFacts. As a result, CCPS and GaDOE
reported inaccurate or unsupported data, including dropouts, graduates, and discipline incidents.
Specifically, CCPS and GaDOE underreported dropouts and discipline incidents. In addition, we
found that GaDOE’s use of program reviews as a monitoring tool for LEAs was insufficient to ensure
the accuracy of data reported.

Without sufficient controls to ensure the accuracy of data, CCPS was providing inaccurate or
unsupported data to GaDOE for reporting to EDFacts. Based on the findings at CCPS, GaDOE could
Final Report
ED-OIG/A04J0003                                                                                       Page 3 of 23
be receiving the same type of inaccurate data from its remaining LEAs. The Department relies on the
reported data in EDFacts and could be making planning, policy, and management decisions based on
inaccurate or unreliable data.

In its comments to the draft report, GaDOE did not concur with either of the two findings nor did it
concur with draft Recommendations 1.4 through 1.7 and Recommendations 2.1 and 2.2. GaDOE
stated that the report findings were based on “program requirements” when no such requirements exist
in Federal law, regulation, or guidance. GaDOE claimed that its practices and internal controls met
Federal regulations and guidance. GaDOE did not state whether or not it concurred with draft
Recommendations 1.1 through 1.3 and Recommendation 2.3.

GaDOE’s comments did not provide sufficient information nor did it provide additional documentation
in support of its nonconcurrence with the Findings and draft Recommendations 1.4, 1.7, 2.1, and 2.2.
We deleted draft Recommendations 1.5 and 1.6 based on GaDOE actions subsequent to the scope of
our audit, which effectively addressed the recommendations. Draft Recommendation 1.5
recommended that GaDOE clearly communicate responsibilities for GSSIS’ triggered edit codes
requiring resolution; and draft Recommendation 1.6 recommended that GaDOE define violent
incidents using the U.S. Department of Education – Education Data Exchange Network (EDEN) N030-
Discipline Incidents File Specifications, Version 4.0, SY 2007-08 as guidance. For the current school
year, GaDOE guidance requires that data submission errors related to our finding 3 be resolved by the
school districts (draft Recommendation 1.5); and GaDOE provided definitions of violent incidents in
its 2007-2008 Consolidated State Performance Report (CSPR) (draft Recommendation 1.6).

In response to comments from the Department’s Program Offices, we added a new Recommendation
1.6 pertaining to GaDOE’s lack of monitoring over LEA reported discipline incidents not classified as
PDS offenses.

GaDOE’s comments are summarized after each finding in the report. The full text of its comments to
the draft report is included as an attachment to the report.

Summary of Changes to Recommendations

Based on GaDOE actions subsequent to the scope of our audit, we deleted draft Recommendations 1.5
and 1.6. As a result of the deleted Recommendations, draft Recommendation 1.7 is now 1.5. Based on
the Department’s Program Offices’ comments, we added a new Recommendation 1.6.




3
  Submission errors that are triggered when students are reported as transferred from one Georgia school to another but not
included in the list of registered students in any Georgia school, which previously only triggered warnings that were not
required to be resolved.
Final Report
ED-OIG/A04J0003                                                                                  Page 4 of 23

FINDING NO. 1 - Inadequate Systems of Internal Control Over Reported Data
We found that CCPS did not have sufficient controls to ensure that accurate, reliable, and complete
data were reported to GaDOE; and GaDOE did not have sufficient controls to ensure the accuracy,
reliability, and completeness of the LEA reported data it submitted to EDFacts. As a result, both
CCPS 4 and GaDOE provided inaccurate, incomplete, or unsupported data related to dropouts,
graduates, and discipline incidents. Specifically, we found that
    •   CCPS did not maintain sufficient documentation supporting dropout, graduate, or discipline
        incident classifications; and reported inaccurate and incomplete dropout and discipline data to
        GaDOE.
    •   GaDOE lacked a sufficient system of internal controls to ensure that it reported to EDFacts all
        of the data classified by its LEAs as reportable; and to ensure that its LEAs reported accurate,
        reliable, and complete information.

Clayton County Public Schools
CCPS did not provide or maintain sufficient documentation supporting the number of students reported
to GaDOE as dropouts and graduates. In addition, CCPS underreported dropout data and it reported
inaccurate, incomplete, and unsupported discipline data.

Insufficient Documentation Supporting Classifications

CCPS did not provide sufficient documentation supporting the number 5 of students it reported to
GaDOE as dropouts and graduates nor could it provide documentation of its evaluation of students
with consecutive days of unexcused absence for classification as dropouts. Further, CCPS could not
identify Department policy 6 governing the identification of dropouts.

We reviewed a sample of 9 of 177 CCPS dropouts in grades 7 through 12 reported to EDFacts for the
2006-2007 school year and found that CCPS could not provide documentation or information
supporting the identification of those 9 students as dropouts. 7 In addition, in our review of 20 of the
2,025 CCPS graduates reported in EDFacts, CCPS provided documentation supporting that 18 students
met graduation requirements, but no documentation was provided for the remaining 2. According to
the Georgia Retention Schedule for Local Government Records, students’ official records, which
include transcripts and standardized test scores, are required to be maintained permanently. Students’
transcripts and standardized test scores provide support that graduates met graduation requirements.

CCPS provided a manual entitled, Clayton County Public Schools Comprehensive Attendance Plan to
explain its process for identifying dropouts for the 2006-2007 school year. However, the manual only
discusses the dropout classification for students who accumulate 10 or more consecutive days of

4
  LEAs, including CCPS, report the data to GaDOE. GaDOE then submits the reported data for all of its LEAs to EDFacts.
5
  Student names are not reported to EDFacts, only student counts.
6
  Can be found in Part I of the Consolidated State Performance Report, and the U.S. DEPARTMENT OF EDUCATION -
Education Data Exchange Network (EDEN) N032 – Dropout File Specifications, Version 3.0, SY 2006-2007.
7
  The codes used to report these 9 students to GaDOE as dropouts include O-Adult Education/Postsecondary, E-Expelled,
R-Removed for Lack of Attendance, and I-Incarcerated.
Final Report
ED-OIG/A04J0003                                                                                  Page 5 of 23
unexcused absences. The manual provides a process for withdrawing these students and assigning
them a withdrawal code that represents a dropout. The process requires
    •   Numerous phone calls to all telephone numbers in the student’s file;
    •   Various letters sent to last known home addresses;
    •   Referral of the student to a school social worker; and
    •   Confirmation sent to the school principal or designee from the school social worker stating that
        the student’s whereabouts are unknown.

According to the manual, upon completion of its dropout determination process, CCPS is required to
withdraw students classified as dropouts using a code “R-Removed for Lack of Attendance” or
“U-Unknown.” Both of the withdrawal codes identify those students as dropouts in the State of
Georgia, 8 and all contact attempts are required to be maintained in the student’s permanent records.
According to the Georgia Retention Schedule for Local Government Records, students’ official
records are required to be maintained permanently.

We found that CCPS did not follow its policies and procedures for evaluating and identifying dropouts.
We reviewed all 15 CCPS students with consecutive days of unexcused absences of 50 days or more
who were not withdrawn or reported as dropouts. 9 CCPS could not provide any documentation
supporting its evaluation of any of the 15 students reviewed. Without the proper documentation
supporting the decision to report or not report a student as a dropout, CCPS may not be identifying and
reporting all dropouts and may be reporting students as dropouts who do not meet the definition.

In addition, we found that CCPS’ manual does not give a comprehensive definition of dropouts nor did
it refer to the Department’s definition. According to the U.S. DEPARTMENT OF EDUCATION -
Education Data Exchange Network (EDEN) N032 – Dropout File Specifications, Version 3.0,
SY 2006-2007,
        A dropout is defined as a student who:
                 a) was enrolled in school at some time during the school year and was not enrolled on
                    October 1 of the following school year, or
                 b) does not enroll in school by October 1 of the school year although expected (i.e.,
                    not reported as a dropout for the previous school year), and has not graduated or
                    completed a state or district-approved educational program, and does not meet any
                    of the following 3 conditions, (1) transferred to another public school district,
                    private school, or state- or district-approved educational program, (2) out on a
                    school-recognized absence due to suspension or illness, or (3) dead.



8
  According to the Georgia Department of Education - FY 2007 FTE Data Collection Data File Layout, the following
withdrawal codes represent dropouts: B-Marriage, E-Expelled, F-Financial Hardship/Job Related, I-Incarcerated,
L-Low Grades/School Failure, M-Military/Student enlisted in Military, O-Adult Education/Postsecondary,
P-Pregnant/Parent, R-Removed for Lack of Attendance, S-Serious Illness/Accident, and U-Unknown.
9
  CCPS and GaDOE provided the information for the 2006-2007 school year, listing a total of 309 students with
consecutive unexcused absences of 10 or more days, 15 of which had consecutive unexcused absences of 50 days or more.
Final Report
ED-OIG/A04J0003                                                                                      Page 6 of 23
The Federal guidance further states that grades 7 through 12 dropout numbers for the 2006-2007
school year are to be reported in EDFacts.

According to CCPS, as part of its controls over classification of dropouts, it periodically provides a list
of reported dropouts to the schools that reported them, requesting that the schools verify that the
students are actually dropouts. However, the periodic reports to schools on dropouts do not ensure that
CCPS is identifying all dropouts, only that the dropouts that have been identified have been verified by
the school. Because CCPS could not provide documentation to support 1) the identification and
classification of 9 students reviewed as dropouts, and 2) the evaluation of an additional 15 students
with 50 consecutive days of unexcused absences to determine the dropout status, we determined that
CCPS cannot rely on the periodic reports to confirm that students identified as dropouts are actually
dropouts. CCPS lacks sufficient controls to ensure complete reporting of dropouts; and it also may be
reporting students as dropouts who do not meet the definition of a dropout. In addition, as discussed in
the following section, CCPS has not been identifying and reporting all dropouts.

Underreported CCPS Dropout Data

CCPS did not identify and report all dropouts to GaDOE for reporting to EDFacts for the 2006-2007
school year because it lacked sufficient controls. As a result, its dropout data in EDFacts are
understated and, therefore, unreliable.

We compared CCPS District Office’s list of 508 students identified as dropouts and no-shows for the
2006-2007 school year to GaDOE’s list of 215 students 10 that supports the CCPS dropout count
reported in EDFacts. We found that 293 students on CCPS’ list were not included on GaDOE’s list. 11
Of the 293 CCPS dropouts and no-shows in question, we found that 102 of them 12 met the
Department’s definition 13 of a dropout (definition detailed in previous section - CCPS, Insufficient
Documentation Supporting Classifications), and therefore should have been reported to GaDOE and
EDFacts as dropouts. Specifically, the Department’s guidance 14 provides that any student who meets
the definition of a dropout according to the 2005-2006 guidance but is not reported as a dropout for the
2005-2006 school year meets the definition of a dropout for the 2006-2007 school year as long as that
student does not meet the other three conditions described in the guidance and has not graduated.

Of the 102 students who met the Department’s definition of a dropout, CCPS reported 3 of them to
GaDOE as grades 7 through 12 dropouts. 15 However, these three students did not appear on GaDOE’s
list of dropouts that supports the 2006-2007 CCPS dropout count in EDFacts. Although the remaining
99 students were all no-shows, CCPS reported all 99 students to GaDOE with a withdrawal



10
   Of the 215 students, 177 were in grades 7 through 12 and 38 in grade levels below 7th.
11
   The remaining 215 students on CCPS’ list were included on GaDOE’s list of dropouts.
12
   Of the 102 students, 56 were in grades 7 through 12, required to be reported to EDFacts, and 46 were in grade levels below
7th, reporting optional. However, GaDOE opted to report dropouts from all grade levels for the 2006-2007 school year.
13
   See page 4 of this report for the Department’s guidance on the identification of dropouts.
14
   U.S. DEPARTMENT OF EDUCATION - Education Data Exchange Network (EDEN) N032 – Dropout File
Specifications, Version 3.0, SY 2005-2006 and Version 3.0, SY 2006-2007.
15
   The withdrawal/dropout codes CCPS reported for the three students for the October Full-Time Equivalent (FTE) Count
for the 2006-2007 school year were “U-Unknown” or “O-Adult/Postsecondary Education.”
Final Report
ED-OIG/A04J0003                                                                                         Page 7 of 23
code “T” - Transfer to another public school system in Georgia or under the jurisdiction of the
Department of Juvenile Justice, for the October Full-Time Equivalent (FTE) Count for the 2006-2007
school year without any support to justify this designation.

CCPS did not follow GaDOE’s policies and procedures in reporting the 99 students. Specifically,
CCPS did not report the students correctly as “no-shows.” According to the Georgia Department of
Education - FY 2007 FTE Data Collection Data File Layout, “no-show refers to a student who started
the enrollment process but subsequently did not attend the school. This is indicated by an appropriate
withdrawal code and a withdrawal date of 06/16/2006.”

CCPS coded all 99 no-shows with the “T” withdrawal code without any notification to indicate that the
students had either transferred or were planning to transfer to another public school in Georgia or were
under the jurisdiction of the Department of Juvenile Justice. CCPS could not provide any
documentation to support the students’ transfer status. Examples of documentation that would indicate
support for use of the “T” withdrawal code could include 1) notice provided by the parent or student
informing the school/CCPS district office that the student was transferring to another public school in
Georgia, 2) withdrawal forms, or 3) a request from another public school in Georgia for the student’s
transcripts/permanent records. In addition, GaDOE could not locate a 2006-2007 enrollment record
for any of the 99 students reported as a transfer with withdrawal code “T.”

Without notification indicating that the 99 students transferred or were planning to transfer to another
public school in Georgia, CCPS should not have reported them to GaDOE with a withdrawal code “T”
indicating transfer. These students should have been reported with a withdrawal code of
“U – Unknown,” which is classified as a dropout in the State of Georgia. Because CCPS reported the
dropouts as transfers, the dropout count reported in EDFacts was understated for the 2006-2007 school
year by 99 students 16 – 53 students from grades 7 through 12 and 46 students from grade levels below
7. Since the number of dropouts is used to calculate the dropout rate, and the number of reported
CCPS dropouts is understated, CCPS' dropout rate for 2006-2007 is understated as well.

Inaccurate and Unsupported CCPS Discipline Data

We found that CCPS’ reported disciplinary data were not sufficiently supported, CCPS inaccurately
coded and reported discipline incidents, and it did not report all reportable discipline incidents. CCPS’
discipline data issues occurred because it did not monitor or verify discipline incident records at
schools to ensure that the incidents were properly labeled, except for offenses that would designate a
school as a Persistently Dangerous School (PDS). As a result, CCPS’ 2006-2007 discipline data
reported in EDFacts were not accurate, reliable, or complete.

GaDOE’s GSSIS limits its defined discipline incident codes to report only certain discipline incidents
to the State. In addition, GaDOE reports only a select group of the GaDOE defined discipline incident
codes in its discipline incident counts 17 reported to EDFacts. Discipline incidents not reported to

16
   The total dropout count is actually underreported by 102 students, the 99 reported as transfers plus the 3 (from grades 7
through 12) that CCPS did report to GaDOE as dropouts, but were not included on GaDOE’s list of dropouts.
17
   GaDOE does not report its defined discipline incident codes to EDFacts; rather, it categorizes its discipline incident codes
into the following five categories defined by the Department, and then reports the incident counts for each of the five
categories to EDFacts: Illicit Drug Related, Alcohol Related, Weapons Possession, Violent Incident With Physical Injury
(VIOWINJ), and Violent Incident Without Physical Injury (VIOWOINJ).
Final Report
ED-OIG/A04J0003                                                                                      Page 8 of 23
GaDOE are not reported to EDFacts. According to the Georgia Department of Education FY 2007
Student Record Data Collection Data File Layout-Discipline File Layout,
     •   Discipline incidents required to be reported to GaDOE include – alcohol, arson, battery,
         burglary, computer trespass, disorderly conduct, drugs except alcohol, fighting, homicide,
         kidnapping, larceny/theft, motor vehicle theft, robbery, sexual battery, sexual harassment, sex
         offenses, threat/intimidation, tobacco, trespassing, vandalism, weapons-knife, weapons-other,
         other discipline incident, weapons-handgun, weapons-rifle, and serious bodily injury.

     •   Discipline actions required to be reported to GaDOE include – corporal punishment, in-school
         suspension (ISS), out-of-school suspension (OSS), expulsion, suspended from riding bus,
         assigned to Crossroads Alternative School, assigned to Other Alternative School, juvenile or
         court system referral, other discipline action for a serious incident, and removed from class at
         teacher’s request.

According to the U.S. Department of Education – Education Data Exchange Network (EDEN) N030-
Discipline Incidents File Specifications Version 3.0, SY 2006-2007, discipline incidents required to be
reported to EDFacts are “…infractions by school-aged students…for illicit drugs, alcohol, weapons
possession, and violence (as those infractions are defined by the state).” GaDOE did not define violent
incidents in its 2006-2007 Consolidated State Performance Report (CSPR). However, consistent with
the U.S. Department of Education – Education Data Exchange Network (EDEN) N030-Discipline
Incidents File Specifications, Version 4.0, SY 2007-2008, GaDOE provided a spreadsheet specifying
the following five categories of discipline incidents and the discipline incident codes for each of the
five categories to be reported to EDFacts for the 2006-2007 school year.
     -   Illicit Drug Related (code 07 - Drugs, except alcohol and tobacco)
     -   Alcohol Related (code 01 - Alcohol)
     -   Weapons Possession (codes 22 – Weapons-knife; 23 – Weapons-other; 25 – Weapons-
         handgun; 26 – Weapons-rifle/shotgun)
     -   Violent Incident With Physical Injury (VIOWINJ) (codes 27 – Serious bodily injury;
         03 – Battery; 14 – Sexual battery)
     -   Violent Incident Without Physical Injury (VIOWOINJ) (code 08 – Fighting)

CCPS reported 3,422 battery, sexual battery, and fighting discipline incidents to GaDOE to include in
its count of VIOWINJ and VIOWOINJ incidents reported to EDFacts for the 2006-2007 school year.
We randomly selected 39 incidents 18 – 5 incidents from the battery category, 5 incidents from the
sexual battery category, and 29 incidents from the fighting category. In our review of the 39 discipline
incidents selected, we found that for 21 of the incidents, CCPS either had insufficient or no
documentation to support the classification reported; or CCPS incorrectly classified and reported the
discipline incidents. Specifically,

     •   CCPS either did not provide sufficient support or provided no support for the classification of
         14 of the 39 incidents reviewed – 2 were battery, 10 were fighting, and 2 were sexual battery.
18
  We reviewed approximately 1 percent of all reported battery, fighting, and sexual battery incidents, selecting random
samples from each category in proportion to the total number of incidents in each category.
Final Report
ED-OIG/A04J0003                                                                                     Page 9 of 23
         Examples of the support provided for the incident classification included a Student Incident
         Maintenance Screenshot from CCPS’ student information system and a Clayton County
         Discipline Record that displays only the incident code without any description or narrative to
         explain what occurred in the incident. Without supporting documentation we were unable to
         determine whether the 14 incidents were correctly reported to GaDOE as not being PDS
         offenses. 19 In addition, according to the Georgia Retention Schedule for Local Government
         Records, student discipline and suspension records are required to be maintained for 7 years, or
         until age 22, whichever is shorter. 20
     •   CCPS provided documentation in support of its discipline incident classification for 7 of the 39
         incidents reviewed; however, the support provided indicated that the 7 incidents were coded
         and reported incorrectly to GaDOE and EDFacts. Of the seven incorrectly coded discipline
         incidents, two incidents did not meet the definition of the GaDOE incident code assigned
         (battery and fighting) nor did the incidents fit into any of the GaDOE defined discipline
         incident codes. In accordance with GaDOE requirements, the two incidents are not reportable
         to GaDOE and, therefore, not reportable to EDFacts. 21 Although the remaining five incidents
         did not meet the definition of the GaDOE incident code assigned (battery and sexual battery),
         the incidents met other GaDOE defined discipline incident codes (other discipline incident,
         fighting, and sexual harassment) and should have been reported to GaDOE with the appropriate
         code. However, the codes that should have been reported for these five incidents were not
         included in the five categories of discipline incidents for which GaDOE reports a count to
         EDFacts. As such, the incidents should not have been reported to EDFacts. 22 None of the
         seven discipline incidents were reported as PDS offenses and the documentation provided
         supported that the incidents were not PDS offenses.

     •   CCPS provided support for the remaining 18 of the 39 discipline incidents reviewed, indicating
         that those incidents were not PDS offenses and were reported correctly to EDFacts.

We also found that CCPS did not provide clear guidance and monitor the use of the Option 7/Exclude
code, which allowed schools’ discretion in use of the code. CCPS coded 4,134 discipline incidents,
using an “Option7/Exclude” code to exclude those incidents for reporting purposes. However, we
found that 4,110 of those incidents should have been reported to GaDOE, and of those that should have
been reported to GaDOE, 148 were consistent with the Department’s examples of violent incidents and
should have been reported to EDFacts. 23


19
   PDS offenses include the following: aggravated battery, aggravated child molestation, aggravated sexual battery,
aggravated sodomy, armed robbery, arson-first degree, kidnapping, murder, rape, voluntary manslaughter, 2 percent or
more of student population or 10 (whichever is greater) involved with non-felony drugs, felony drugs, felony weapons, and
terroristic threats.
20
   According to 34 C.F.R. § 80.42, retention and access requirements for records, “all financial and programmatic
records, supporting documents, statistical records, and other records of grantees or subgrantees which are: (i)
Required to be maintained by the terms of this part, program regulations or the grant agreement, or (ii) Otherwise
reasonably considered as pertinent to program regulations or the grant agreement…. must be retained for three
years…”
21
   One of the two incidents was included in CCPS’ VIOWINJ count reported in EDFacts, and the other was included in
CCPS’ VIOWOINJ count.
22
   The five incidents were included in CCPS’ VIOWINJ count reported to EDFacts.
23
   Our determination is based on the discipline incident code and action CCPS assigned to the incidents.
Final Report
ED-OIG/A04J0003                                                                       Page 10 of 23
According to the Clayton County Public Schools – Administrative Disciplinary Guidelines and
Resource Manual, 2006-2007, “Option 7/Exclude, which prevents an incident from being accumulated
for state reporting, should be used in a relatively consistent manner throughout the school district as it
relates to disciplinary offenses that result in in-school suspension [ISS]….” However, GaDOE defines
all ISS actions as reportable, as such, all discipline incidents resulting in ISS should have been reported
to the State. CCPS staff explained that the intent of “Option 7/Exclude” was to 1) correct an error in
which a student was assigned ISS but did not serve it; and 2) exclude from reporting certain offenses
for which ISS was not given to students.

Of the 148 offenses that CCPS should have reported to GaDOE to report to EDFacts based on the
discipline incident code and discipline incident action assigned, we selected a random sample of 30 for
more detailed analyses. We found that 24 of the 30 incidents should have been reported to GaDOE or
to both GaDOE and EDFacts. For the remaining six incidents, CCPS used the “Option 7/Exclude”
code in accordance with the explanation provided by the CCPS District Office.

Although it had not defined violent incidents, GaDOE reported violent incident counts on its
2006-2007 CSPR and provided violent incident counts to EDFacts for that same school year. Of the
24 incidents determined to have been reportable, 16 were consistent with incidents reported to GaDOE
and EDFacts in the same 2006-2007 school year and included – fighting; bullying; horse-
playing/wrestling; and pinching. The remaining eight incidents were serious enough to be considered
violent, however, GaDOE did not have defined discipline incident codes for CCPS to report the 8
incidents, which included – pushing a student to ground, then hitting him; kneeing, kicking, and hitting
female students in the buttocks; slapping; punching; and choking. Of the 24 reportable incidents, 8 of
them should have been reported to GaDOE, and the remaining 16 incidents should have been reported
to both GaDOE and EDFacts.

The inaccurate coding and unsupported CCPS discipline incidents, and the failure to report all CCPS
discipline incidents required to be reported to EDFacts, indicate that CCPS has insufficient controls
over the accuracy of reported discipline data. Except for discipline incidents classified as PDS
offenses, CCPS did not monitor or verify discipline incident records at schools to ensure that the
incidents are properly labeled.

Georgia Department of Education
GaDOE reported inaccurate CCPS dropout and discipline data to EDFacts for the 2006-2007 school
year because it lacked sufficient controls to ensure that 1) it reported to EDFacts all dropouts correctly
identified by LEAs, and 2) LEA-reported dropout and discipline data were accurate. As a result,
CCPS’ 2006-2007 dropout data in EDFacts is understated, and its EDFacts reported discipline data are
inaccurate and incomplete, and therefore, unreliable.


Underreported Dropout Data

GaDOE did not report all dropouts to EDFacts for the 2006-2007 school year. As discussed in the
CCPS section above (CCPS, Underreported CCPS Dropout Data), we identified 102 CCPS students
who met the Department’s definition of dropout but were not included in the dropout count in EDFacts
Final Report
ED-OIG/A04J0003                                                                                     Page 11 of 23
for the 2006-2007 school year. As a result, GaDOE’s dropout data in EDFacts were underreported for
the 2006-2007 school year by the 102 CCPS students that we found met the Department’s definition. 24

Of the 102 students that we identified as having met the Department’s definition of a dropout, CCPS
reported 3 to GaDOE as grades 7 through 12 dropouts. 25 However, GaDOE did not include them in
the dropout count reported to EDFacts. For the remaining 99 students, GaDOE did not include those
students in the dropout count reported to EDFacts because CCPS erroneously reported them to GaDOE
as transfers 26 instead of dropouts, and GaDOE did not have sufficient controls to detect such erroneous
data. As a result, GaDOE underreported the CCPS dropout count in EDFacts by a total of 102
students 27 – 56 students from grades 7 through 12th and 46 students from grade levels below 7th.
Because the number of dropouts is used to calculate the dropout rate, CCPS' dropout rate for the 2006-
2007 school year is understated as well.

GaDOE’s failure to report all dropouts to EDFacts indicates weak controls over the accuracy and
reporting of dropout data. GaDOE’s Accountability Office stated that it does not monitor or verify
LEA reported data. Rather it accepts the LEAs’ information as accurate until told otherwise, with the
exception of verifying that reported dropouts have not re-enrolled in school. For every student
reported to GaDOE as a dropout, the Accountability Office checks for the students’ re-enrollment into
another school in Georgia. The Accountability Office makes the necessary changes to the student's
dropout status if the student has re-enrolled in another school. However, this control does not ensure
that all dropouts are identified and reported.

In addition, GaDOE’s Data Collections Office stated that it relies on business rules and edits
embedded 28 in the processing system of the GSSIS to ensure the accuracy of data reported to EDFacts.
However, while the business rules and edits may validate the data submitted by LEAs, they cannot
always be relied on to ensure the accuracy of the data. For example,
          GSSIS’ edit code W2562 is triggered when a student is reported to GaDOE with a
          withdrawal code of “T” and the student ID is not reported at any other school in the
          system or any other system in the State. Because GaDOE could not locate a 2006-
          2007 enrollment record for any of the 99 students reported as a transfer with
          withdrawal code “T” (as discussed in the CCPS section above), the edit code W2562
          should have been triggered. The code would have indicated a potential for inaccurate
          data. According to GaDOE, prior to the 2009-2010 school year, a warning would
          have been sent to the LEA, indicating that an online explanation was required.

Effective for the 2009-2010 school year, edit code W2562 changed to E2567. As a result, an error
report will be sent to the LEA for the current school year when the edit code E2567 is triggered,
indicating that a correction to the data must be made before it can be successfully uploaded to GSSIS.

24
   See page 4 for the definition.
25
   The 9 were reported as dropouts using GaDOE defined withdrawal codes for dropouts – “U-Unknown” or “O-
Adult/Postsecondary Education” for the October Full-Time Equivalent (FTE) Count for the 2006-2007 school year.
26
   Reported with a withdrawal code “T - Transfer to another public school system in Georgia or under the jurisdiction of the
Department of Juvenile Justice.”
27
   The 99 reported as transfers plus the 3 (from grades 7 through 12th) reported as dropouts to GaDOE, but not reported to
EDFacts.
28
   Examples of the embedded controls include edits checking for letters inputted into numeric fields; appropriate formatting
in fields; values recorded in required fields; and duplication of a student’s ID number.
Final Report
ED-OIG/A04J0003                                                                                   Page 12 of 23
However, had there been a clear, communicated requirement for LEAs or GaDOE to resolve records
with edit code W2562 prior to the current school year, the 99 students may have been properly
reported as dropouts.

GaDOE’s Data Collections Office also stated that the final step in the data reporting process for LEAs
is for local superintendents to attest to the accuracy and completeness of the data reported to GaDOE
and to attest that all explanatory notes regarding warnings of potentially discrepant data are accurate.
However, because GaDOE does not periodically check the accuracy of LEA reported data, it cannot
assure itself that the superintendents’ attestation is reliable in terms of ensuring the accuracy and
completeness of the reported data.

Inaccurate and Unsupported Discipline Data

GaDOE reported inaccurate and unsupported discipline data to EDFacts for the 2006-2007 school year
because 1) it lacked sufficient controls to detect erroneous LEA reported discipline data, and 2) it had
not clearly defined violent incidents, thereby limiting LEAs’ ability to identify and report all violent
incidents. As a result, GaDOE/CCPS’ discipline data in EDFacts are inaccurate, incomplete, and
unreliable.

As discussed in the CCPS section above (CCPS, Inaccurate and Unsupported CCPS Discipline Data),
from our sample of 39 CCPS battery, fighting, and sexual battery discipline incidents reported to
GaDOE and EDFacts for the 2006-2007 school year, 7 of the incidents were reported to GaDOE with
incorrect discipline incident codes. However, because GaDOE did not monitor or verify the accuracy
of LEA reported discipline data, 29 GaDOE reported the incorrectly coded discipline incidents to
EDFacts, thereby causing data in EDFacts to be inaccurate. 30 For 14 of the 39 sampled incidents,
CCPS either did not provide sufficient support or provided no support. Because GaDOE did not
ensure that its LEAs maintained support for discipline incidents for the 7 years required by the Georgia
Retention Schedule for Local Government Records, or even the 3 years required by 34 C.F.R.§ 80.42,
the discipline data it reported to EDFacts are unsupported and not reliable. 31

Also, as discussed in the CCPS section above (CCPS, Inaccurate and Unsupported CCPS Discipline
Data), we found that from our sample of 30 discipline incidents that CCPS excluded from being
reported to GaDOE using an Option 7 Exclude code, 24 should have been reported to GaDOE or both
GaDOE and EDFacts. However, because GaDOE does not have controls to ensure that its LEAs
report all reportable discipline incidents, the number of reported discipline incidents in GaDOE’s
GSSIS and in EDFacts is understated.

According to the U.S. Department of Education – Education Data Exchange Network (EDEN) N030-
Discipline Incidents File Specifications, Version 3.0, SY 2006-2007, discipline incidents required to be
reported to EDFacts are “…infractions by school-aged students…for illicit drugs, alcohol, weapons

29
   The only exception is that GaDOE does monitor discipline incidents that would designate a school as a Persistently
Dangerous School (PDS).
30
   Two of the seven incorrectly coded incidents should not have been reported to GaDOE or EDFacts. The other five
incorrectly coded incidents should have been reported to GaDOE with a different discipline incident code and should not
have been reported to EDFacts at all.
31
   For the remaining 18 of the 39 discipline incidents reviewed, CCPS provided adequate documentation to support the
incidents.
Final Report
ED-OIG/A04J0003                                                                         Page 13 of 23
possession, and violence (as those infractions are defined by the state).” We found that GaDOE had
not clearly defined violent incidents but rather provided categories and examples of what would be
considered a violent incident and limited its reporting of incidents to EDFacts to the categories
provided. As a result, we found that CCPS did not report all of its discipline incidents that were
serious enough to be considered violent because GaDOE did not have defined discipline incident codes
for CCPS to report such incidents (as detailed in a previous section of this report – CCPS, Inaccurate
and Unsupported CCPS Discipline Data).

Without defined incident codes that include all violence, or an overall code for those incidents that
meet the definition of violent but do not fit it to any prescribed classification, violent incidents go
unreported in EDFacts. In addition, GaDOE limited its LEAs’ ability to identify and report all violent
incidents while at the same time providing LEAs a great amount of discretion in classifying incidents
within the categories provided. For example, an LEA may consider an incident as violent with no
clear category of reporting and decide to report it as violent in any category, just so that it is reported.
However, another LEA may exclude the incident because there is no category for reporting. As a
result, discipline data between LEAs within GaDOE may be inconsistent.

Recommendations

We recommend that the Assistant Secretary for Planning, Evaluation, and Policy Development instruct
GaDOE to require CCPS to –

1.1 Establish and implement systems of internal control to ensure that it identifies and properly reports
    all dropouts, and maintain the supporting documentation in accordance with prescribed record
    retention requirements;

1.2 Maintain official student records, documenting whether graduates have met all graduation
    requirements, in accordance with prescribed record retention requirements; and

1.3 Establish and implement systems of internal control to ensure that it identifies and properly reports
    all discipline incidents required to be reported to the State and EDFacts, and maintain the
    supporting documentation in accordance with prescribed record retention requirements.

We recommend that the Assistant Secretary for Planning, Evaluation, and Policy Development instruct
GaDOE to –

1.4 Establish and implement systems of internal control to ensure that it identifies and properly
    reports all dropouts to EDFacts; and

1.5 Follow up on issues identified at CCPS, take appropriate action to correct the data, and advise the
    Department of actions taken.

We recommend that the Assistant Secretary for Planning, Evaluation, and Policy Development, in
conjunction with the Assistant Deputy Secretary for Safe and Drug-Free Schools, instruct GaDOE to –

1.6 Establish and implement systems of internal control to ensure that non-PDS discipline incidents
    are accurately classified at the local level.
Final Report
ED-OIG/A04J0003                                                                       Page 14 of 23
GaDOE Comments

In its comments to the draft audit report, GaDOE did not concur with the finding and draft
Recommendations 1.4 through 1.7. GaDOE objected to the report findings, stating that they were
based on “program requirements” that did not exist in Federal law, regulation or guidance. GaDOE
maintained that the auditors used criterion that had no relationship to EDFacts program regulations.

Specifically, GaDOE stated that it did not concur with Finding 1 for reasons summarized in its non-
concurrence of the following Recommendations.
    1.4   GaDOE stated it does not concur with Recommendation 1.4 because its internal controls
          over identifying and properly reporting all dropouts to EDFacts meet Federal regulations and
          no additional corrective action is needed.
    1.5   GaDOE stated it does not concur with draft Recommendation 1.5 (which has been deleted)
          because data resulting in errors or warnings are sent back to the school district, and all errors
          must be corrected before the LEA’s superintendent signs off. GaDOE stated that it believes
          its internal controls over resolution of edit codes are in compliance with Federal regulations
          and no additional corrective action is needed.
    1.6   GaDOE stated it does not concur with draft Recommendation 1.6 (which has been deleted)
          because data collected in the 2006-2007 school year should not have to follow guidance
          issued for the 2007-08 school year since such guidance did not exist at the time the data was
          collected.
    1.7   GaDOE stated it does not concur with draft Recommendation 1.7 (currently
          Recommendation 1.5) because the data collected is archived and used immediately for
          various purposes; and GaDOE does not allow any changes to the archived data. GaDOE
          stated that it does not agree that additional corrective action is needed.

GaDOE did not state whether or not it concurred with draft Recommendations 1.1 through 1.3.


OIG Response

GaDOE did not provide sufficient information nor did it provide additional documentation in support
of its nonconcurrence with the Finding, Recommendation 1.4, and draft Recommendation 1.7 (which
is Recommendation 1.5 in this report). As a result, we did not change the finding. We did, however,
delete draft Recommendations 1.5 and 1.6 based on GaDOE’s corrective actions subsequent to the
scope of our audit. GaDOE guidance for the current school year (2009-2010) requires that data
submission errors related to our finding on dropouts be resolved by the school districts, which
effectively addressed draft Recommendation 1.5. In addition, GaDOE provided definitions of violent
incidents in its 2007-2008 CSPR, which effectively addressed draft Recommendation 1.6. In addition,
in response to the Department’s comments, we added a new Recommendation 1.6 to address GaDOE’s
lack of sufficient monitoring of LEA reported discipline incidents not classified as PDS offenses,
which allowed inaccurate information to be reported in EDFacts.
Final Report
ED-OIG/A04J0003                                                                      Page 15 of 23
As stated in the report, 34 C.F.R. § 80.40 and Federal Register Volume 72 require data collection
through EDFacts from all of the States. Also cited in the report are various sources of Federal
guidance, which specify the requirements and data set definitions for the data States are required to
submit to EDFacts. Federal guidance required States to submit violent incident counts to EDFacts but
allowed States to define violent incidents. However, GaDOE reported in its 2006-2007 CSPR that it
had not yet defined violent incidents; consequently, we 1) compared CCPS documentation of incidents
to incidents that GaDOE reported to EDFacts as violent for the same school year; and 2) applied the
categories of incidents that GaDOE used in reporting violent incidents in our review of the discipline
incidents. Of the 24 incidents determined to have been reportable, 16 were consistent with incidents
reported to GaDOE and EDFacts in the same 2006-2007 school year. CCPS’ documentation of eight
additional discipline incidents supported a classification that those incidents were serious and of a
nature consistent with the Department’s examples of violent incidents. The eight incidents cited in the
report included pushing a student to the ground then hitting him; kneeing, kicking, and hitting female
students in the buttocks; slapping; punching; and choking. However, none of the eight were reported to
GaDOE and, therefore, were not reported in EDFacts.

Although GaDOE did not concur with Recommendation 1.4, GaDOE did not provide evidence that its
internal controls meet Federal regulations. Among other discrepancies in the dropout data, we found
that a significant number of dropouts (102) were not reported in EDFacts but should have been,
indicating internal control weaknesses over reported data. As such, we did not change the
recommendation.

GaDOE’s reasons for its nonconcurrence with draft Recommendation 1.7 (Recommendation 1.5 in the
final report) does not address why it should not follow-up on issues identified at CCPS. The issues
identified in this report, as it pertains to CCPS, indicate systemic problems with CCPS’ identification of
dropouts and reportable discipline incidents. We maintain that archived data should be corrected,
especially, when it could be used to make important planning, policy, and management decisions.

FINDING NO. 2 - Program Reviews Do Not Ensure the Accuracy of Reported Data
GaDOE’s use of program reviews as a monitoring tool is not sufficient to ensure the accuracy of LEA
reported data. With the exception of the Special Education Office and Title I Cross-Functional
Monitoring Reviews, program reviews may be sufficient to determine whether the LEAs were in
compliance with applicable program requirements, but they were not sufficient to ensure the accuracy
of the data reported. The program reviews for 1) GaDOE’s Special Education Office did not contain
sufficient documentation to show what was obtained and reviewed for half of the selected reviews for
both years; and 2) Title I Cross-Functional Monitoring Team did not contain sufficient documentation
to show what was obtained and reviewed for half of the selected reviews for the 2006-2007 school
year. As such, we could not determine whether the reviews for those two programs were sufficient to
determine LEAs’ program compliance. In addition, we found that none of the program reviews
selected for the 2006-2007 and 2007-2008 school years included a step to determine the accuracy of
the LEAs’ reported Title I; Homeless; Career, Technical, Agricultural Education (CTAE); English to
Speakers of Other Languages (ESOL); and Special Education data in EDFacts.

With the exception of the ESOL Office, we found that a sufficient number of program reviews were
performed for both the 2006-2007 and 2007-2008 school years for the offices reviewed. The results of
Final Report
ED-OIG/A04J0003                                                                      Page 16 of 23
our review of GaDOE’s five program offices that have data reported in the EDFacts system are
detailed in Appendix B.

Although program reviewers may test 32 the accuracy of data provided to them while on-site, such tests
do not ensure that the same data are reported to GaDOE and EDFacts because program reviews are
conducted during the school year while final reporting of data is done after the completion of the
school year. Therefore, data presented during program reviews are subject to change throughout the
school year. The program reviewers do not test or verify the accuracy of any data that have been
reported to EDFacts.

According to 34 C.F.R. § 80.40, “…Grantees are responsible for managing the day-to-day operations
of grant and subgrant supported activities. Grantees must monitor grant and subgrant supported
activities to assure compliance with applicable Federal requirements and that performance goals are
being achieved. Grantee monitoring must cover each program, function or activity.” Annual reporting
of education data to the Department is a Federal requirement for grantees. Therefore, since GaDOE, as
a grantee, submits LEA reported data to the Department (EDFacts), GaDOE is responsible for
monitoring its subgrantees’ (LEAs) reported data to ensure that they are accurate, reliable, and
complete, and to ensure that the LEAs are in compliance with applicable Federal requirements.

The Department obtained approval from OMB to require the electronic submission of data through
EDFacts. 33 The Department published Paperwork Reduction Act Notices for this data collection under
the title Annual Mandatory Collection of Elementary and Secondary Education Data for the Education
Data Exchange Network (now EDFacts). According to the Paperwork Reduction Act Submission
Supporting Statement - Annual Mandatory Collection of Elementary and Secondary Education Data
through EDFacts, data collection through EDFacts from States is mandatory starting with the 2006-
2007 school year.

According to Georgia State Board of Education Rule 160-5-1-.07(1)(c), “Georgia Statewide Student
Information System (GSSIS) – The GSSIS is the system that will be used to collect data from school
systems and to produce Full-Time Equivalent, Student Record, and Discipline reporting requirements
beginning in school year 2006-2007.” GaDOE uses data in GSSIS to report to EDFacts.

GaDOE’s weak controls in ensuring the accuracy of EDFacts reported data could be partly due to its
program reviewers not being familiar with the EDFacts system. Only three program office staff from
all five offices with program reviews evaluated stated that they had knowledge of the EDFacts system;
and neither of the three program office staff with knowledge of the system verified data in EDFacts. In
addition, four of the program office staff stated that they do not have access to EDFacts and three of
the four stated that they do not have access to the GSSIS.

Because GaDOE does not ensure the accuracy of reported EDFacts data, LEAs could be providing
inaccurate or unsupported performance data to GaDOE, for reporting to EDFacts. Data in EDFacts are
used to pre-populate some parts of the CSPR that SEAs are required to submit to the Department
annually. EDFacts also provides data for planning, policy, and management at the Federal, State, and
local levels. Because GaDOE does not have sufficient controls to ensure the accuracy of data in

32
     The results of their program reviews are presented in program review reports.
33
     See Federal Register Vol. 72.
Final Report
ED-OIG/A04J0003                                                                               Page 17 of 23
EDFacts, GaDOE could be providing inaccurate data to decision makers for use in making planning,
policy, and management decisions. An added step to program reviews to check the quality of reported
data in its monitoring visits to LEAs could significantly reduce the risk of LEAs providing inaccurate
or unsupported data to GaDOE.

Recommendations

We recommend that the Chief Financial Officer, in conjunction with the Assistant Secretary for
Elementary and Secondary Education and the Assistant Secretary for Special Education and
Rehabilitative Services, instruct GaDOE to –

2.1     Require its program reviewers to test the accuracy of EDFacts reported data; and

2.2     Require Special Education program reviewers to document what is obtained and reviewed
        during program reviews in support of its monitoring efforts at LEAs.

We recommend that the Assistant Secretary for Planning, Evaluation, and Policy Development request
Georgia’s Governor’s Office of Student Achievement 34 to –

2.3     Review reported education data for a sufficient number of LEAs each year to provide
        reasonable assurance that Georgia’s EDFacts data are accurate.


GaDOE Comments

In its comments, GaDOE stated that it does not concur with Finding 2 for the reasons summarized in
its nonconcurrence of Recommendations 2.1 and 2.2.
        2.1 GaDOE stated it does not concur with Recommendation 2.1 because program reviews are
            not required to “test the accuracy of EDFacts reported data.” GaDOE added that the
            Federal EdFacts office uses the same data verification procedures that are currently used by
            GaDOE. GaDOE maintained that its program reviews are modeled after the program
            reviews conducted by the Department and follow the Federal regulations set forth by their
            respective Program Office. In addition, GaDOE stated that its current process of having the
            LEA superintendent sign off and attest to the certification is a sufficient control.
        2.2 GaDOE stated it does not concur with Recommendation 2.2 because its Special Education
            program reviews are documented as required by Federal law.

GaDOE did not state whether or not it concurred with Recommendation 2.3.

OIG Response

GaDOE did not provide sufficient information nor did it provide additional documentation in support
of its nonconcurrence. As a result, no changes were made to the finding and related recommendations.

34
  The Official Code of Georgia Annotated (O.C.G.A.) 20-14-26 authorizes the Governor’s Office of Student Achievement
to audit and inspect the records of Pre-Kindergarten-12 grade schools, as well as colleges and universities.
Final Report
ED-OIG/A04J0003                                                                     Page 18 of 23
Per 34 C.F.R. § 80.40 and Federal Register Volume 72, annual reporting of education data to the
Department through EDFacts is required of grantees; and GaDOE, as a grantee, submits LEA reported
data to the Department (EDFacts). As such, GaDOE is responsible for monitoring its subgrantees’
(LEAs) reported data to ensure that they are accurate, reliable, and complete, and to ensure that the
LEAs are in compliance with applicable Federal requirements.

Although having the LEA superintendents sign-off and attest to the reported data is an important
control, it should not be the only control because it does not provide sufficient control over the
accuracy of the reported data, as evidenced by the findings detailed in this audit report. CCPS reported
inaccurate and incomplete dropout and discipline data to GaDOE for the 2006-2007 school year.
GaDOE should have a monitoring process for periodically verifying the accuracy of LEA reported data
in EDFacts. Absent an adequate monitoring process, GaDOE has no assurance that the
superintendents’ attestation is sufficient to ensure the accuracy and completeness of the reported data.

GaDOE provided no additional support for its assertion that the documentation of its Special Education
program reviews is in compliance with Federal law, as such, no change was made to the
recommendation.



                    OBJECTIVES, SCOPE, AND METHODOLOGY


The objectives of the audit were to 1) determine whether GaDOE and CCPS established adequate
systems of internal control to provide accurate education data to EDFacts; and 2) evaluate GaDOE’s
use of program reviews as a monitoring tool for LEAs. Our audit covered selected CCPS EDFacts
data for the 2006-2007 school year and program reviews performed by GaDOE for the 2006-2007 and
2007-2008 school years.

To determine the policies, procedures, and controls over the accuracy of education data, we obtained
and reviewed Federal and State (Georgia) laws, regulations, policies, procedures, and guides, and
CCPS guidance and regulations. We interviewed officials in the Governor’s Office of Student
Achievement, GaDOE officials in its offices of Data Collections, Assessments, Accountability, Title I,
Special Education, Safe and Drug-Free Schools, CTAE, ESOL, and Migrant Education; and CCPS
officials in its departments of Technology, Learning Support Services, Student Services, Exceptional
Students (Special Education), CTAE, ESOL, and Homeless Education.

To determine whether graduates included in the CCPS graduate count reported to EDFacts for the
2006-2007 school year met graduation requirements, we obtained from GaDOE the list of 2,025 CCPS
graduates supporting the regular diploma graduate count reported to EDFacts. We selected a random
sample of 20 graduates and requested the documentation showing that the 20 graduates met graduation
requirements.

To determine whether CCPS followed its own process for identifying and withdrawing as dropouts,
students who accumulated 10 or more consecutive days of unexcused absences, we obtained 1) a list of
all students that accumulated 10 or more consecutive days of unexcused absences during the 2006-
2007 school year, and 2) the withdrawal and dropout data for CCPS students for the 2006-2007 school
Final Report
ED-OIG/A04J0003                                                                     Page 19 of 23
year from both GaDOE and CCPS. From the universe of 309 students in grades 7 through 12 who
accumulated consecutive days of unexcused absences of 10 days or more and were not withdrawn or
reported as dropouts, we reviewed all 15 students with consecutive days of unexcused absences of 50
days or more.

To determine whether CCPS students included in the dropout count reported to EDFacts for the
2006-2007 school year met the definition of a dropout, we obtained from GaDOE the list of 177 CCPS
dropouts supporting the grades 7 through 12 dropout count reported to EDFacts. We reviewed a
random sample of nine dropouts, which is 5 percent of the population.

We compared CCPS District Office’s list of 508 students identified as dropouts and no-shows for the
2006-2007 school year to GaDOE’s list of 215 students that supports the CCPS dropout count reported
in EDFacts. For the 293 students on the list provided by the CCPS District Office that were not
provided by GaDOE, we obtained and reviewed withdrawal and enrollment information on each of the
293 students in question to determine whether they should have been reported to GaDOE and EDFacts
as dropouts.

To determine whether CCPS’ 3,422 battery, sexual battery, and fighting incidents reported to EDFacts
for the 2006-2007 school year should have been reported to GaDOE as aggravated battery or
aggravated sexual battery (offenses that count toward a school’s PDS designation), we reviewed a
random sample of discipline incidents (totaling 39) from each of the 3 categories in proportion to the
total number of incidents in each category (1 percent of population).

To determine whether CCPS discipline incidents were excluded from being reported to GaDOE and
EDFacts for the 2006-2007 school year, we obtained a list of all discipline incidents that were assigned
an Option 7/Exclude code for the 2006-2007 school year. Of the 4,134 discipline incidents assigned an
Option 7/Exclude code, we determined that 148 should have been reported to EDFacts based on the
code and incident action assigned. Of the 148 discipline incidents, we reviewed a random sample of
30 to determine whether CCPS appropriately used the Option 7/Exclude code, which would be
supported by documentation proving that the students either did not serve the assigned discipline
action or were not assigned ISS or other reportable action.

To evaluate GaDOE’s use of program reviews as a monitoring tool for LEAs we identified GaDOE’s
five program offices that have data reported in the EDFacts system. For each of the five program
offices, we determined the number of LEA program reviews performed that resulted in findings (230)
and the number that did not result in findings (149), for both the 2006-2007 and 2007-2008 school
years. For each of the five program offices, we reviewed a random sample of 15 percent of each
universe or 5, whichever was greater, for each of the 2 years. In evaluating the program reviews, we
determined 1) the sufficiency of the reviews in ascertaining LEA program compliance, 2) whether a
sufficient number of reviews were performed each year, and 3) whether testing the accuracy of
EDFacts data was included in the scope of the reviews.

The audit itself was a test of the reliability of computer-processed data in EDFacts. As discussed in
Finding No. 1, we found inaccurate GaDOE/CCPS EDFacts data specifically pertaining to dropouts
and discipline incidents.
Final Report
ED-OIG/A04J0003                                                                    Page 20 of 23
We conducted our fieldwork at GaDOE, located in Atlanta, Georgia, and at CCPS, located in
Jonesboro, Georgia. Our on-site visits at the two locations took place between January 2009 and May
2009. Our followup of the visits and analyses continued through September 2009. An exit conference
was held with selected officials from GaDOE and CCPS on October 20, 2009.

We conducted this performance audit in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform audits 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.



                               ADMINISTRATIVE MATTERS


Statements that managerial practices need improvements, as well as other conclusions and
recommendations in this report, represent the opinions of the Office of Inspector General.
Determinations of corrective action to be taken will be made by the appropriate Department of
Education officials.

If you have any additional comments or information that you believe may have a bearing on the
resolution of this audit, you should send them directly to the following Department of Education
officials, who will consider them before taking final Departmental action on this audit:
                              Carmel Martin
                              Assistant Secretary
                              Office of Planning, Evaluation, and Policy Development
                              U.S. Department of Education
                              400 Maryland Avenue, S.W.
                              Washington, DC 20202
                              Kevin Jennings
                              Assistant Deputy Secretary
                              Office of Safe and Drug-Free Schools
                              U.S. Department of Education
                              400 Maryland Avenue, S.W.
                              Washington, DC 20202

                              Thelma Melendez de Santa Ana
                              Assistant Secretary
                              Office of Elementary and Secondary Education
                              U.S. Department of Education
                              400 Maryland Avenue, S.W.
                              Washington, DC 20202
Final Report
ED-OIG/A04J0003                                                                       Page 21 of 23
                               Alexa Posny
                               Assistant Secretary
                               Office of Special Education and Rehabilitative Services
                               U.S. Department of Education
                               400 Maryland Avenue, S.W.
                               Washington, DC 20202

It is the policy of the U. S. Department of Education to expedite the resolution of audits by initiating
timely action on the findings and recommendations contained therein. Therefore, receipt of your
comments within 30 days would be appreciated.

In accordance with the Freedom of Information Act (5 U.S.C. § 552), reports issued by the Office of
Inspector General are available to members of the press and general public to the extent information
contained therein is not subject to exemptions in the Act.


                                                       Sincerely,


                                                       /s/
                                                       Denise M. Wempe
                                                       Regional Inspector General for Audit

Attachment
Final Report
ED-OIG/A04J0003                                                         Page 22 of 23
Appendix A

                     Acronyms/Abbreviations Used in this Report

AYP          Academic Yearly Progress
CCPS         Clayton County Public Schools
CSPR         Consolidated State Performance Report
CTAE         Career, Technical, and Agricultural Education
Department   U.S. Department of Education
EDEN         Education Data Exchange Network
ESOL         English to Speakers of Other Languages
ESS          Education Data Exchange Network (EDEN) Submission System
FTE          Full-time Equivalent
FY           Fiscal Year
GaDOE        Georgia Department of Education
GSSIS        Georgia Statewide Student Information System
ISS          In-school Suspension
K-12th        Kindergarten through 12th
LEA          Local Educational Agency
OMB          Office of Management and Budget
OPEPD        Office of Planning, Evaluation, and Policy Development
OSS          Out-of-school Suspension
PDS          Persistently Dangerous Schools
SEA          State Education Agency
SY           School Year
VIOWINJ      Violent Incident With Physical Injury
VIOWOINJ     Violent Incident Without Physical Injury
Final Report
ED-OIG/A04J0003                                                                                  Page 23 of 23
                                                                                                     Appendix B


                        Summary Results of Selected Program Reviews

                                                                                                          Percent of
                                                 Percent of Sampled Reviews             Sufficient(b)   Reviews that
                      Program                    Sufficient(a), in Determining          Number of        did not Test
                       Review       Sample      LEA Program Compliance, or               Reviews         Accuracy of
Program Office          Year         Size              Undeterminable                   Performed       EDFacts Data
                                                                                       Yes; 1/3 of
  Title I Cross-       2006-07         12       41.7% suffic. 58.3% undeterm.          LEAs                  100%
   Functional
  Monitoring(c)        2007-08         13                  100% suffic.                Yes; 1/3 of           100%
                                                                                       LEAs
                                                                                       Yes; all
   Title X,            2006-07          8                  100% suffic.                LEAs(d)               100%
McKinney-Vento
  Homeless             2007-08          9                  100% suffic.                Yes; all              100%
                                                                                       LEAs(d)
                                                                                       Yes; 1/5 of
                       2006-07          6                  100% suffic.                LEAs                  100%
Perkins IV CTAE
                       2007-08          6                  100% suffic.                Yes; 1/5 of           100%
                                                                                       LEAs
                                                                                            n/a
                       2006-07        n/a(e)                    n/a                                            n/a
 Title III - ESOL                                                                      Yes; 1/3 of
                       2007-08          5                  100% suffic.                                      100%
                                                                                       LEAs
                                                                                       Yes; 1/3 of
                       2006-07         11       45.5% adeq.      54.5% undeterm.       LEAs                  100%
 IDEA – Special
   Education           2007-08         10       50% adeq.        50% undeterm.         Yes; 1/3 of           100%
                                                                                       LEAs
   (a) Sufficient - the review included a sufficient number of documented review questions/answers; and if there were
       any findings, then explanations and recommendations were included.
   (b) Sufficient number of reviews - at least 1/5 of the LEAs are reviewed each year
   (c) Includes a combined review of Title I, Title II-A, Title V-Part A, and Title VI-Part B
   (d) All LEAs participating in the Homeless Program
   (e) ESOL office did not start performing program reviews until the 2007-08 school year
                                                                                          ATTACHMENT




                                                                  Office of the State Superintendent of Schools
 Kathy Cox, State Superintendent of Schools

                                                   March 22, 2010


Ms. Denise M. Wempe
U.S. Department of Education
Office of the Inspector General
61 Forsyth Street, SW
Suite 18T71
Atlanta, GA 30303
                                               Control Number: ED-OIG/A04J0003
Dear Ms. Wempe:

Enclosed you will find the Georgia Department of Education (GaDOE) response to the findings
and recommendations presented in the audit of Georgia Department of Education’s Controls
Over Performance Data Entered in EDFacts draft audit report.

GENERAL OBJECTION

GaDOE strongly objects to the two findings issued in the draft audit report and considers the
draft audit report to be grossly unfair. GaDOE’s objection is that the findings are based on
“program requirements” when no such requirements exist in federal law, regulation or guidance.
It is unclear why these additional “program requirements” apply only to GaDOE. If these
“program requirements” are truly federal requirements for which auditing is appropriate, then
these “program requirements” should be properly promulgated and applied to all SEAs not just
the ones who are audited.


FINDING NO.1 – Inadequate Systems of Internal Control Over Reported Data
GaDOE does not concur with this finding for the reasons summarized below.

Recommendation 1.4
GaDOE does not concur with this recommendation. GaDOE’s internal controls over this process
meet federal regulations and no additional corrective action is needed on our part.




    2066 Twin Towers East ● Atlanta, GA 30334 ● (404) 656-2598 ● Fax (404) 651-8737 ● www.gadoe.org

                                              An Equal Opportunity Employer
                                                                                ATTACHMENT

Ms. Denise M. Wempe
Page 2
March 22, 2010


Recommendation 1.5
GaDOE does not concur with this recommendation. In the data collection process, many business
rules are applied to the data resulting in errors or warnings that are sent back to the school
district. These errors and warnings are the result of invalid values, inaccurate reporting, and
percentage differences on year-to-year comparisons. Before data are ready for LEA
superintendent signoff, all errors must be corrected. Consequently, GaDOE believes that our
internal controls over this process are in compliance with federal regulations and no additional
corrective action is needed.

Recommendation 1.6
GaDOE does not concur with this recommendation. GaDOE followed the federal guidelines
which were in place for the 2006-07 school year. The guidelines cited in the draft audit report
were for the 2007-08 school year. GaDOE does not agree that data collected in the 2006-07
school year should follow guidance issued for the 2007-08 school year since such guidance did
not exist at the time the data was collected. Following future guidance is impossible for any
department. GaDOE can only follow the guidance in force at the time the data was collected.
GaDOE did so. GaDOE has and intends to continue to follow current federal guidance. GaDOE
does not agree that additional corrective action is needed.

Recommendation 1.7
GaDOE does not concur with this recommendation. Once GaDOE collects the data, it is archived
and used immediately for various purposes such as AYP determinations and report card
publication as well as responding to data requests. This data is provided to numerous individuals,
entitles and organizations. Once the data is archived, GaDOE does not allow any changes to the
data. Consequently, GaDOE does not agree that additional corrective action is needed.

FINDING NO.2 – Program Reviews Do Not Ensure the Accuracy of Reported Data
GaDOE does not concur with this finding for the reasons summarized below.

Recommendation 2.1
GaDOE does not concur with this recommendation. Program reviews are not required to “test the
accuracy of EDFacts reported data” and GaDOE does not know of any such federal requirement
that exists in law, regulation, or guidance. GaDOE notes that the auditors have not cited any
legal or binding authority that requires such action. The federal EdFacts office uses the same
data verification procedures that are currently used by GaDOE. This requirement was seemingly
created by your audit team without adequate knowledge regarding EDFacts or the programs
reviewed in your audit. Last, GaDOE program reviews are modeled after the program reviews
conducted by the U.S. Department of Education and follow the federal regulations set forth by
their respective federal program office.
                                                                                 ATTACHMENT

Ms. Denise M. Wempe
Page 3
March 22, 2010


GaDOE’s current process of having the LEA superintendent signoff and attest to the certification
set forth below is a sufficient control. The LEA superintendent attests as follows:
        “I certify that the information above, as well as the information on all (data collection
        activity) reports available for this school district on the FTE web site, is accurate and
        complete. I hereby acknowledge that this information will be used for funding purposes
        and in reporting to the legislature and other policy makers, to federal and state agencies,
        and in the Georgia Public Education Report Card. I have reviewed the explanatory notes
        regarding warnings of potentially discrepant data, and I am satisfied that these notes are
        accurate.”

Since the control in place is sufficient, GaDOE believes that no further corrective action is
needed.

Recommendation 2.2
GaDOE does not concur with this recommendation. Our Special Education program reviews are
documented as required by federal law. Therefore, GaDOE does not believe that further
corrective action is needed in this area.

In conclusion, GaDOE urges you to carefully consider our position on these finding and that they
be reconsidered in light of the fact that they are apparently based on unwritten, unpublished and
therefore unreasonable standards. If you have any questions or concerns about the foregoing,
please contact my Chief of Staff Stephen Pruitt at 404-657-6165.

Sincerely,



Kathy Cox

KC/gs