Office of the Governor

Special Investigators

June 30, 2011

HAND DELIVERED

Governor Nathan Deal State Capitol Atlanta, GA 30335

Dear Governor Deal :

In January of this year, you continued our appointment as your special investigators to probe allegations of test tampering and related matters in the Atlanta Public School System (APS).

We have determined that cheating occurred throughout that school district. Our investigation found organized and systemic wrongdoing in APS well before the administration of the 2009 CRCT.

Our investigative report follows. The entire file is available to the appropriate authorities, as you direct. Please let us know if we may be of further service.

Very truly yours,

MICHAEEJ. BOWERS ROBERT C WILSON RICHARD L. HYDE

INVESTIGATIVE TEAM

Governor's Special Investigators

Robert E. Wilson Michael J. Bowers Richard L. Hyde

Balch and Bingham, LLP

James L. Hollis Geremy W. Gregory E. Righton Johnson

Kara M. Engelberger, Paralegal Deborah Daley, Legal Assistant Susan G. Hughes, Legal Assistant Barbara Watson, Legal Assistant

Lydia Rooks, Legal Assistant Elizabeth A. Jackson, Project Asst.

Wilson, Morton and Downs, LLC

Keri P. Ware Rosyln S. Mo watt

Linda Weaver, Paralegal Debbie Morelli, Paralegal Tracey Duren, Legal Assistant Cheryl Hicks, Legal Assistant

Georgia Bureau of Investigation

Director Vernon M. Keenan Inspector John Heinen Special Agent Heather Strickland (Case Agent)

Kelly Aldrich Renea Anderson Elizabeth Bigham

Rocky Bigham

Amy Braswell

Leigh Brooks Michael Brooks

Tonya Cales Ryan Carmichael

Derek Coffey Jerri Lynn Coody

Karen Crowe

Dan Kirk, Assistant Director Monica Ling Gregory Linton

Chad Lott Kendra Lynn ASAC Jesse Maddox

Lindsay Marchant Christopher McKeown Megan Miller Rhiannon Morgan Carlos Murray David Norman

i

Agie George

SAC Denise Norman

Lindsey Giddens

Richard Otwell

Earl Glover

Trebor Randle

Wendell Goodman

Latoria Reynolds

Brian Hargrove

Janet N. Rhodes

James Harris

Evelyn Rodgers

Mary Holder

Deborah Rollins

wesiey riorne

Amanaa Kowien

jj/iigene riowara

iveoecca onaw

Terry Howard

Jamie Skelton

Cecil Hutchins

Kristina Smalley

uayce rngaiis

r am omitn

/\nita ivy

jonatnan opuriocK

Marko Jones

Sara Thomas

Deanna Jury

Lisa Vorrasi

Lawrence Kelly

Cynthia Wahl

Klay Kilcrease

Benny Williams

Georgia Information Sharing and Analysis Center

Meredith Bailey

Emily Butler Yvonne Darrell Heather Davis

Kevin Garrett Yalanda Greene

Laurie Lane ASAC Cynthia Ledford Stephanie Lockridge

Jessica Price Mark Reinking Terri St. Romain Jan Roulain, Department of Corrections

Deanna Scott Tammy Starckey Wendi Walker Keesha Walker

ii

Office of State Inspector General

Elizabeth Pequeno Archer, Esq. Deron R. Hicks, Esq. William L. Donaldson, III, CPA, CFE Deborah Wallace, CIG

District Attorney, Atlanta Judicial Circuit

Hon. Paul Howard Judge Eleanor L. Ross (Formerly Executive Assistant District Attorney)

Solicitor-General, DeKalb County

Hon. Sherry Boston Investigator Steve Barresi

Georgia State Patrol

Col. Bill Hitchens Lt. Col. Mark McDonough Major Russell Powell Lt. Mark Hambert Sgt. Robert Moody Trooper Christopher Hinkle Trooper K. Reeder Trooper Chris Stallings Trooper Larry Miller

iii

Report Limitations

This report is an overview of the evidence and our findings. It is not

intended to include every detail or fact developed during this investigation. Nor does it include every relevant document. All notes, documents, transcripts and interview summaries related to this investigation will be available to you, and the appropriate authorities for whatever action, if any, is appropriate.

Special Thanks

The investigators wish to express their gratitude to Governors Perdue and Deal, and their staffs, for their support of our work. We also wish to extend our appreciation to Ms. Kathleen Mathers, Director of the Governor's Office of Student Achievement, for her indispensable assistance throughout this investigation.

iv

TABLE OF CONTENTS

Volume 1

Map of Schools 1

Overview 2

The CRCT 6

Interviews and Document Review 7

2009 Erasure Analysis 9

Standard Deviations Chart 10

APS Erasure Analysis 11

Verification of the Erasure Analysis 12

Use of the Erasure Analysis in This Investigation 15

School Summaries 15

Investigative Compilation 17

Parks Middle School 20

Venetian Hills Elementary School 53

Gideons Elementary School 60

Kennedy Middle School 66

F.L. Stanton Elementary School 72

Perkerson Elementary School 77

Connally Elementary School 85

Usher/Collier Heights Elementary School 91

Peyton Forest Elementary School 98

East Lake Elementary School 106

V

Cook Elementary School 117

Woodson Elementary School 125

Scott Elementary School 134

Deerwood Academy 144

Humphries Elementary School 151

Glossary 158

vi

TABLE OF CONTENTS

Volume 2

(School Summaries continued)

Dunbar Elementary School 161

D.H. Stanton Elementary School 168

Finch Elementary School 175

Coan Middle School 185

Dobbs Elementary School 191

Toomer Elementary School 204

Benteen Elementary School 207

Beecher Hills Elementary School 215

Fain Elementary School 219

Slater Elementary School 226

Thomasville Heights Elementary School 233

Fickett Elementary School 242

Hutchinson Elementary School 247

Capitol View Elementary School 251

Towns Elementary School 257

Blalock Elementary School 262

Whitefoord Elementary School 266

Boyd Elementary School 269

West Manor Elementary School 273

Turner Middle School 276

vii

White Elementary School 280

Harper Archer Middle School 282

M. Agnes Jones Elementary School 296

Parkside Elementary School 303

Bethune Elementary School 305

Miles Elementary School 312

Grove Park Elementary School 315

Jackson Elementary School 318

Cleveland Elementary School 320

Alonzo A. Crim Open Campus High School 322

Benjamin S. Carson Preparatory Academy 323

C.W. Hill Elementary School 326

Adamsville Elementary School 327

Cascade Elementary School 328

Heritage Academy Elementary School 329

University Community Academy 330

Williams Elementary School 335

Herndon Elementary School 336

Bolton Academy Elementary School 337

Morningside Elementary School 339

Morris Brandon Elementary School 341

2009 vs. 2010 342

APS Percentage of Classes with Flagged WTR's Chart 345

Glossary 347

viii

TABLE OF CONTENTS

Volume 3

Questions 350

Why Cheating Occurred 350

Targets 350

Culture of Fear 356

Dr. Jackie Boyce 357

Jimmy e Hawkins 359

Michael Milstead 361

Former High-Level Official 361

Patrick Crawford 362

Santhia Curtis 363

Teachers 365

Ethics 365

Early Warnings 366

Allegations of Cover-Up 368

Parks Middle School 368

Investigation at Parks 369

Meeting with Senior Leadership 371

Retaliation by Waller 374

Dramatic Gains at Parks 374

Deerwood Academy 378

Alteration and Destruction of Documents 383

ix

Porter and Reeves Reports 385

Media Request for Porter Report 389

APS Reaction to GOSA 390

Blue Ribbon Commission 392

Test Security 395

KPMG and APS Interviews 396

APS Response to BRC Report 398

The Business Community 401

Findings 402

Dr. Sharon Davis-Williams 406

Michael Pitts 406

Tamara Cotman 407

Veleter Mazyck 408

Millicent Few 408

Dr. Kathy Augustine 408

Dr. Beverly Hall 409

Glossary 411

X

Map of Schools

Legend

School Locations

i Elementary Middle High | SRT 1

SRT2 SRT 3 SRT 4

Jackson

Jackson Ann

Brandon Anns*

Sinill i K Ann ex

.n f^agine-Wesle?

Hui-hinsnn Lleveland Avenue

Humphries Long SButfrAllynLy

Heritage Academy

Atlanta Public Schools Facility Locations 2009-2010

1

OVERVIEW

Thousands of school children were harmed by widespread cheating in the Atlanta Public School System (APS). In 30 schools, educators confessed to cheating. We found cheating on the 2009 Criterion-Referenced Competency Test (CRCT) in 44 of the 56 schools (78.6%) we examined, and uncovered organized and systemic misconduct within the district as far back as 2001. Superintendent Beverly Hall and her senior staff knew, or should have known, that cheating and other offenses were occurring. Many of the accolades, and much of the praise, received by APS over the last decade were ill-gotten.

We identified 178 educators as being involved in cheating. Of these, 82 confessed. Thirty-eight of the 178 were principals, from two-thirds of the schools we examined. The 2009 erasure analysis suggests that there were far more educators involved in cheating, and other improper conduct, than we were able to establish sufficiently to identify by name in this report.

A culture of fear and a conspiracy of silence infected this school system, and kept many teachers from speaking freely about misconduct. From the onset of this investigation, we were confronted by a pattern of interference by top APS leadership in our attempt to gather evidence. These actions delayed the completion of this inquiry and hindered the truth-seeking process.

2

The APS General Counsel told us that one of her main duties was to provide Superintendent Hall with "deniability." Her aim was to insulate Dr. Hall from the burden of responsibility for making difficult decisions. This veil of deniability at the school level was aptly illustrated by long-time Gideons Elementary principal Armstead Salters, who told his teachers: "If anyone asks you anything about this, just tell them you don't know . . . just stick to the story and it will go away."

There was a failure of leadership throughout APS with regard to the ethical administration of the 2009 CRCT. There are two main reasons for this failure. Dr. Hall's insular style and her isolation from the rank-and-file was a major factor. In addition, Dr. Hall and her top managers refused to accept responsibility for anything other than success. As Dr. Hall's Chief of Staff, Sharron Pitts, explained to us, "nobody ever wants to take responsibility for anything" in APS.

Deputy Superintendent Kathy Augustine oversaw daily classroom instruction, and operated as the de facto second-in-command. She told us that she should not be held responsible for cheating that took place in APS classrooms under her authority.

While this may be an appropriate defense to criminal charges, it is an absurd leadership concept. Dr. Hall and her senior cabinet accepted accolades when those below them performed well, but they wanted none of the burdens of failure.

3

The first person to report cheating to us provided the same information months earlier to his superiors, only to have the wrongdoers quickly exonerated while he was reprimanded. This educator made these allegations known to the proper officials inside of APS. However, the district improperly handled this complaint in violation of its own policies. That inquiry was brought to a swift, and predictable, conclusion. The guilty went free; the whistle-blower was punished. This was not an isolated occurrence and was illustrative of the culture of fear and intimidation which promoted a code of silence.

The Office of Internal Resolution (OIR) was responsible for internal investigations, but lacked independence and gave those who wanted to report improper activity little confidence that complaints would be objectively, fairly and competently investigated.

As early as 2006, APS officials improperly manipulated and hid information relating to CRCT administration, and illegally altered documents related to that test. The school district often failed to comply with Georgia's open records laws, withheld public information and gave false data to an agency of this state.

Dr. Hall stated publicly, and several times, that she would "fully cooperate" with our efforts. However, the district was slow in producing documents and claimed legal exemptions where none existed.

4

All of this was done to keep from public view, and this inquiry, information which might raise doubts about the validity of the 2009 CRCT scores, and other indicators of success in the classroom.

Blue Ribbon Commission (BRC) expert, Dr. John Fremer, wrote an op-ed piece for The Atlanta Journal - Constitution (AJC) which said: "...[wjholesale organized cheating in some Atlanta Public Schools occurred and must be addressed." (Ex. 1). Experts who assisted us expressed similar sentiments in saying that cheating is the only plausible explanation for the abnormally high standard deviations shown in the erasure analysis.

One of the first tasks we undertook was to test the validity of the Governor's Office of Student Achievement (GOSA) erasure analysis. This was done with the assistance of our expert during a visit to the test facility of CTB McGraw-Hill. The erasure analysis is, without question, accurate and reliable.

The statistics are astounding. For example, of the approximately 1,800 non- APS elementary and middle schools in Georgia where the 2009 CRCT was given, 54 schools were flagged with more than 20% of their classes being greater than three standard deviations outside the state norm on wrong-to-right erasures. Yet in the 90 elementary and middle schools in the APS system where the 2009 CRCT was given, 52 schools were flagged with more than 20% of their classes being greater than three standard deviations outside the norm. Incredibly, almost half of

5

the schools flagged for being greater than three standard deviations outside of the norm in our state were from the Atlanta Public School System.

THE CRCT

The CRCT is a multiple choice examination given annually to all public school students in Georgia. There are five subject areas that are tested: reading; English/language arts; math; social studies and science. Students are scored as "meets standards," "exceeds standards" or "does not meet standards." The CRCT is considered an important test because its results help determine whether a school makes "Annual Yearly Progress" (AYP) as required by the federal No Child Left Behind Act. Every elementary and middle school within a school district must administer the CRCT at the same time and in the same manner, during a nine-day window. During the first five days, a different subject area is tested each day. The last four days of the window are used for make-up testing.

Georgia law requires that the test be administered under tightly-controlled conditions. The test materials are delivered to the individual schools several days before the test begins. Each school designates a certified educator to be responsible for test administration. In APS, this person is known as the testing coordinator, who must ensure that the test is administered according to the test protocols. But the principal bears ultimate responsibility for ensuring how the test is administered.

6

Teachers receive training on test administration using procedures that specifically set forth how the test must be given. Any deviation from the test protocols is prohibited.

hi first and second grade, teachers read the test questions aloud and students answer questions in the test booklet by marking the correct answer. (Ex. 2). Teachers must read each question only twice, with no voice inflection that could suggest the answer. Third through eighth graders read the test questions for themselves and answer questions on a separate Scantron® sheet by filling in the appropriate bubble by pencil. (Ex. 3). Each test section is timed and contains between forty and sixty questions. Only special education students with specified accommodations may have variances in the test administration.

INTERVIEWS AND DOCUMENT REVIEW

On August 26, 2010, Governor Sonny Perdue appointed us as his Special

Investigators to investigate alleged test tampering, and related matters, in the Atlanta Public School System. (Ex. 4). This order was augmented by oral directives that we were to:

Find the truth with regard to cheating, if any, on the 2009 CRCT within APS;

Assist state regulators in sanctioning educators who participated in cheating;

Submit information to prosecuting authorities regarding criminal conduct, if discovered.

7

Governor Perdue emphasized that our mandate was to find the truth. He also stressed that teachers who were honest in their testimony should not be criminally prosecuted. You restated these directives to us upon assuming office. (Ex. 5).

hi order to gain an understanding of the overall structure of APS, how the testing process works, the relevant players, and what documents would be needed, we first conducted benchmark interviews of top officials in the district, including Dr. Hall, Dr. Augustine, Dr. Cari Ryan, and Dr. Alexis Kirijan. Most of these officials were interviewed again toward the end of this investigation.

We interviewed the teachers and administrators at each of the flagged schools, as well as current and former executive directors of each school reform team (SRT). The SRT executive directors function as assistant superintendents, assigned to one of four geographic areas of elementary and middle schools for the district. They oversee principals at the schools within their SRT and report directly to Dr. Kathy Augustine.

hi addition to interviews of district personnel, we also spoke with scores of individuals from outside the system, who participated in the BRC investigation or served as consultants. We conducted over 2,1 00 interviews and reviewed in excess of 800,000 documents.

8

2009 ERASURE ANALYSIS

In February 2010, the Governor's Office of Student Achievement (GOSA)

produced an erasure analysis performed by CTB McGraw-Hill on the spring 2009 CRCT. The results of this analysis raised the possibility of testing irregularities. The GOSA erasure analysis, which was performed on the test answer documents for every elementary and middle school student in the State of Georgia, compared the number of wrong-to-right (WTR) erasures by grade, test subject and class to the average number of WTR erasures state-wide for the corresponding grade and test subject. The results of the erasure analysis showed that in 35 Georgia school districts, including APS, a significant number of classes had WTR erasures that were dramatically and disconcertingly higher than the state average.

Specifically, CTB McGraw-Hill determined that if a class had WTR erasures more than three standard deviations above the expected norm (i.e., the state average), it was almost statistically impossible for such a high number of WTR erasures to have occurred without some external force operating to cause it. For example, at three standard deviations there is only a one in 370 chance that the high erasures occurred by coincidence and at five standard deviations there is only a one in 1.7 million chance. By seven standard deviations, it is virtually impossible only a one in 390 billion chance that such a high number of WTR erasures occurred randomly.

9

STANDARD DEVIATIONS CHART

Standard Deviations

Chance of Occurring Randomly

3

1/370

4

1 / 15,788

5

1 / 1,774,278

6

1 / 560,800,000

7

1 7390,600,000,000

In other words, some external force operated to cause the WTR erasures. Although a WTR erasure analysis does not indicate that the external force was cheating, it does suggest that something other than normal student erasing occurred.

Thirty-five Georgia districts had schools with more than five percent of the classes flagged for standard deviations higher than three. (Ex. 6). The GOSA study grouped schools into four categories based on the percentage of flagged classrooms: "clear of concern"; "minimal concern"; "moderate concern"; and "severe concern." Eighty-percent of Georgia's elementary and middle schools fell into the "clear of concern" category, 10% fell into "minimal concern," 6% fell into "moderate concern," and 4% fell into the "severe concern" category.

10

APS ERASURE ANALYSIS

The percentage of flagged classes in APS far exceeded any other district in

Georgia. Of the middle and elementary schools 51% fell into the "severe concern" category. Of the "moderate concern," were 18%, and 8% were of "minimal concern." (Ex. 7). APS accounts for over half of the "severe" category schools in the state. Parks Middle School, with 89.5% of its classes flagged, led the state in percentage of classes flagged for WTR erasures, with Gideons Elementary and Peyton Forest Elementary not far behind at 88.4% and 86.1%, respectively.

The erasure analysis only flagged classes that departed from the norm by three or more standard deviations. But many classes in APS had standard deviations ranging from the 20's to the 50's. (Ex. 8). One classroom was at 53. It is virtually impossible for so many WTR erasures to occur without human intervention.

Amazingly, many APS teachers had high WTR erasures in all three subject areas English/language arts, reading and math. Not only did numerous teachers do something that was virtually impossible one time, but did it three times in a row. Even more amazing, several teachers in the same school did this multiple times.

Dr. Gregory Cizek, our expert, analogized the chances of this occurring to the Georgia Dome being filled to capacity, with every person in the Dome being

11

seven feet tall. Dr. John Fremer of Caveon Test Security, hired by the BRC to conduct its own statistical analysis, described this in terms of flipping two coins three times in a row, and the coins land on their edge, perfectly balanced, one on top of the other, all three times.

VERIFICATION OF THE ERASURE ANALYSIS

We verified that the results of the erasure analysis were accurate and

consistent. This study served as a guide to identify where cheating may have occurred, and it established the foundation for this investigation. We took the following steps to ensure its validity:

Retained an expert to review the GOSA erasure analysis;

Inspected the CTB McGraw-Hill facility and interviewed several members of the staff who were involved in grading the CRCT and conducting the erasure analysis;

Observed the answer document scanning process;

Compared the results of the erasure analysis to the results of a reanalysis of selected and random test documents;

Manually reviewed thousands of answer sheets and compared them to the results of the original erasure analysis; and,

Interviewed experts in the educational testing and statistics field.

Based on these efforts, we concluded that the GOSA erasure analysis is accurate, reproducible, and reliable.

12

We retained Gregory J. Cizek, Ph.D., of the University of North Carolina, who is one of the foremost experts on educational testing and statistics in the nation. Dr. Cizek is a Professor of Educational Measurement and Evaluation in the School of Education at UNC. He currently serves as the President of the National Council on Measurement in Education. (Ex. 9). After Cizek reviewed the erasure analysis, he accompanied us to the CTB McGraw-Hill facility. We toured the CTB McGraw-Hill plant, observed the answer documents being re-scanned and interviewed CTB McGraw-Hill's statistician and other personnel familiar with the scanning process.

CTB McGraw-Hill's high-optical scanner read the students' test documents and recorded answers and erasures for each section. A computer used special software to determine when an answer was changed from wrong-to-right (WTR), right-to-wrong (RTW), or wrong-to-wrong (WTW). This data reflected the total number of erasures and the total number of WTR changes for each student in each subject area in Georgia.

Next, CTB McGraw-Hill employed a statistical test to flag excessive numbers of WTR erasures in a class. (Ex. 10). The average number of WTR erasures statewide in a given grade and subject were compared to the number of WTR erasures in a specific class within the APS district. The proximity of erasures to the expected norm is expressed in terms of standard deviations. CTB

13

McGraw-Hill flagged classes that were three or more standard deviations above the state average.

GOSA used a conservative criterion of three standard deviations. This was done to insure that only the most severe and questionable erasures were identified.

We interviewed company officials and manually reviewed answer documents, counted erasures, and compared our count with the computer's analysis. This manual count of erasures revealed more changes than the computer scanning process. The computer is not as stringent as the human eye. The difference is not because the scanner missed erasures, but because it is calibrated to give the benefit of the doubt to a certain level before it considers a lighter mark.

To confirm the study results, we asked that CTB McGraw-Hill re-scan both random and selected tests. The results of the re-scanned answer documents were consistent with the results of the original erasure analysis.

We interviewed the two individuals from Caveon Test Security who used the GOSA erasure data and conducted their own analysis on behalf of the BRC. Neither disputed the results of the GOSA study. The top 12 schools flagged under their "Caveon Index" were identical to the highest flagged schools under the GOSA analysis.

14

USE OF THE ERASURE ANALYSIS IN THIS INVESTIGATION

The erasure data helped us prioritize interviews of educators at the schools

to allow us to efficiently focus our efforts. We also used this information when we questioned teachers and administrators, since they had not been provided with this data by anyone in the district.

We compared the student scores with other evidence to better understand what occurred in classrooms. The student data listed every student in APS and set forth how many total erasures, versus how many WTR erasures, appeared on that student's answer document. This information provided an additional perspective for analyzing erasures.

When student-level data revealed a large number of students within a single class with high erasures that changed from wrong to right 70%- 100% of the time, such information raised an additional suspicion that someone other than the students could be changing answers.

SCHOOL SUMMARIES

Investigative summaries of the 56 schools we examined follow this section.

We found that 178 teachers and principals were involved in cheating in 44 schools. Sixty-eight percent of the principals of the 56 schools were responsible for cheating, and six of those refused to answer all questions we asked them, including about their involvement in cheating. These six pled the Fifth Amendment, which

15

for civil law purposes, such as a Georgia Professional Standards Commission (PSC) proceeding, is an implied admission.

An investigative compilation shows a breakdown of those found cheating by each school.

16

INVESTIGATIVE COMPILATION

School

Confessions

Other

Total

Parks Middle

7

6 (hicl. Prin.).

13

Venetian Hills Elementary

6

2 (Incl. Prin.).

8

Gideons Elementary

12 (Incl. Prin.).

0

12

Kennedy Middle

1

3 (Incl. Prin.).

4

FL Stanton Elementary d

1

2 (Incl. Prin.).

3

Perkerson Elementary

0

3 (Incl. Prin.).

3

Connally Elementary

1

1 (Incl. Prin.).

2

Usher Elementary

3

2 (Incl. Prin.).

5

Peyton Forest Elementary

0

10 (Incl. Prin.).

10

East Lake Elementary _ d

0

2 (Incl. Prin.).

2

Cook Elementary

2

4 (Incl. Prin.).

6

Woodson Elementary

2

3 (Incl. Prin.).

5

Scott Elementary

0

3 (Incl. Prin.).

3

Deerwood Academy

1

3 (Incl. Prin.).

4

Humphries Elementary

2

3 (Incl. Prin.).

5

Dunbar Elementary

1

7 (Incl. Prin.).

8

DH Stanton Elementary

0

2 (Incl. Prin.).

2

Finch Elementary

3

6 (hicl. Prin.).

9

Coan Middle

1

2 (hicl. Prin.).

3

Dobbs Elementary

4

2 (Incl. Prin.).

6

Toomer Elementary

3 (Incl. Prin.).

1

4

Benteen Elementary

0

3 (Incl. Prin.).

3

Beecher Hills Elementary

3

1 (hicl. Prin.).

4

Fain Elementary

2

2 (Incl. Prin.).

4

Slater Elementary

2

3 (Incl. Prin.).

5

Thomasville Heights Elementary

2

2 (hicl. Prin.).

4

Fickett Elementary

2

2 (Incl. Prin.).

4

Hutchinson Elementary

1

1 (Incl. Prin.).

2

Capitol View Elementary

0

1 (hicl. Prin.).

I

Towns Elementary

0

1 (Incl. Prin.).

I

Blalock Elementary

0

1 (Incl. Prin.).

I

Whitefoord Elementary

0

1 (hicl. Prin.).

I

Boyd Elementary

0

1 (hicl. Prin.).

i

West Manor Elementary

0

1 (Incl. Prin.).

i

Turner Middle

0

1 (Incl. Prin.).

I

White Elementary

0

1 (hicl. Prin.).

i

Harper Archer Middle

0

0

0

MA Jones Elementary

6

1

7

Parkside Elementary

3

0

3

Bethune Elementary

2

1 (Incl. Prin.).

3

Miles Elementary

2

1

3

Grove Park Elementary

2

0

2

Jackson Elementary

2

0

2

Cleveland Elementary

1

0

1

Crim Open Campus

0

0

0

Benjamin Carson Middle

0

0

0

CW Hill Elementary

0

0

0

Adamsville Elementary

0

0

0

Cascade Elementary

0

0

0

Heritage Elementary

0

0

0

University Community Academy

2

4 (Incl. Prin.).

6

Williams Elementary

0

0

0

Herndon Elementary

0

0

0

Bolton Elementary

0

0

0

Morningside Elementary

0

0

0

Morris Brandon Elementary

0

0

0

TOTAL

82

96

178

17

For each school we have prepared analyses of relevant witness interviews, statistical data and other materials. Listed below is some of the misconduct found in the school summaries. What is revealed is outrageous:

Teachers and administrators erased students' incorrect answers after the test was given and filled in the correct answers;

The changing of answers by teachers and administrators was, in some cases, so sophisticated that plastic transparency answer sheets were created to make changing the test answer sheets easier;

Changing of answers was often done at weekend gatherings, and in at least one instance at a teacher's home in Douglas County, Georgia;

A principal forced a teacher with low CRCT scores to crawl under a table at a faculty meeting;

Teachers arranged classroom seating for tests so that lower performing children could cheat off the higher scoring students;

Children were denied special educational assistance because their falsely-reported CRCT scores were too high;

Students requested that they be assigned to a certain teacher because that educator was said to cheat;

First and second grade teachers used voice inflection while reading the test to identify the answer;

Teachers pointed to the correct answer while standing at students' desks;

Teachers gave the answers aloud to students;

Some teachers allowed students to change the previous day's incorrect responses after giving them correct answers;

Teachers looked ahead to discuss the next day's questions;

18

In one classroom a student sat under his desk and refused to take the test. This child passed.

Following the school summaries is a comparison between the 2009 and 2010 erasure analyses. There was a dramatic drop in the percentage of flagged classrooms between these years. This was only after media attention and the state sent representatives to some district schools.

19

PARKS MIDDLE SCHOOL

1090 Windsor Street SW Principal: Christopher Waller SRT-2 Executive Director: Michael Pitts

Atlanta, Georgia 30310 Testing Coordinator: Dr. Alfred Kiel

I INVESTIGATIVE SUMMARY

Cheating occurred on the CRCT at Parks Middle School in 2006, 2007, 2008, 2009, and 2010. Fifty-nine people were interviewed at this school, some more than once. Seven teachers confessed to cheating. Cheating at Parks is evidenced by a high number of flagged classrooms, confessions and witness testimony. The cheating started when Principal Christopher Waller began at Parks and recruited two teachers to change answers in 2006. As the years progressed, more teachers got involved. In all years, the cheating was organized and facilitated by Principal Waller and Success-For-All Facilitator Sandra Ward. Assistant Principal Gregory Reid also participated. The cheating was reflected in the statistically improbable testing gains and extremely high numbers of flagged classrooms in 2009 for high wrong-to-right erasures. Parks had the highest percentage of flagged classrooms in the State of Georgia. Teachers gave students the answers to the tests, organized changing "parties" where the answer sheets were changed, and illegally accessed the test booklets before testing. The cheating was conducted covertly so that Testing Coordinator Dr. Alfred Kiel would not discover it.

II. STATISTICAL DATA

2009

2010

Percentage of Classrooms Flagged for WTR Erasures

89.5

4

Number of Classrooms Flagged for WTR Erasures

51

3

Number of Teachers Flagged for WTR Standard Deviations above 3.0 (Number of Teachers Flagged in Multiple Subjects)

19(18)

3(0)

Mean WTR Standard Deviations from State Norm

11.9

3.2

High Flagged Standard Deviation

29.4

3.4

Low Flagged Standard Deviation

3.4

3.1

III. SUMMARY OF EVIDENCE

A. Narrative

Christopher Waller became the principal of Parks in the fall of 2005. Waller directed cheating the first year he presided over CRCT testing in 2006. He gave teacher Damany Lewis a key to the room where the tests were kept. Lewis removed the plastic wrap from the test booklets and photocopied the tests. Lewis gave the copies to other teachers, who used the advance copies to give students the answers. A select group of teachers that Waller organized and trusted would change wrong answers to right answers each day during the week of testing. There is also evidence that Waller directed cheating on the secured writing tests.

Each year Principal Waller and his crew brought more teachers into the cheating conspiracy. Waller, Gregory Reid, or Sandra Ward went to these teachers' classrooms and told

20

them it was "time to go." The teachers understood that "time to go" meant they were to go to the room where the tests were kept and change answers.

Dr. Alfred Kiel was the testing coordinator for this school. He would not allow cheating so Principal Waller orchestrated Kiel's absence from the school building so the cheating could take place. On one occasion in 2009, Principal Waller took Kiel out for a "retirement lunch." In another year, Principal Waller scheduled an impromptu after-school dance so that the teachers could stay late in the afternoon and cheat without raising suspicion. Kiel once noticed that things in his office had been disturbed while he was out and became angry. After that occasion, teacher Damany Lewis took pictures of Kiel's office before he altered the tests so that everything would be put back in exactly the same place so as not to raise Kiel's suspicions. No one implicated Kiel except Principal Waller.

B. APS' Knowledge of Cheating

District Leadership knew Principal Waller was cheating. See discussion of Reginal Dukes ' investigation into Parks Middle School in Volume Three of this Report. Dr. Beverly Hall, Dr. Kathy Augustine, Millicent Few, and others were aware of Dukes' investigation and findings. No action was taken against Principal Waller.

Dr. Hall also should have known Waller was cheating at Parks because once he became principal, the school immediately made dramatic gains on the CRCT and other tests. For example, between the 2004-2005 and 2005-2006 school years, eighth graders meeting or exceeding standards in reading increased by 31 percentage points, from 50% to 81%. The percentage of students meeting and exceeding standards in English/language arts increased by 27 percentage points, from 54% to 81%. In math, the percentage of eighth graders who met or exceeded the standards increased from 24% to 86%. The percentage of students exceeding expectations went from 1% to 46%, a 45 point increase. In 2006-2007, one year after Dukes' investigation into Parks, the school met 100% of targets set by APS.

There is no evidence that APS management instituted any additional investigations into Parks despite the improbable gains in scores and Dukes' conclusion that cheating occurred on the eighth grade writing test in 2006. Instead, APS publicly touted Waller and Parks Middle School for its achievements. Dr. Beverly Hall praised Principal Waller's performance, saying, "You have to find someone who is able to go in and, while not being a dictator, gets people's attention and articulates a vision and mission in a way that people want to be on board with it . . . ." A copy of Sarah Torian's Beating the Odds at Atlanta's Parks Middle School is included as Attachment A.

C. Testimony of Witnesses

1. Damany Lewis (Teacher)

Damany Lewis was the first teacher to assist Principal Waller in cheating. He admitted to cheating in 2006, 2007, 2008 and 2009. In 2006, Waller asked Lewis, "Do you think you could get into something undetected?" Lewis was not sure what Waller meant, but said yes. A few days later, Lewis was summoned to the main office where he found Principal Waller and Sandra Ward with the CRCT booklets. Principal Waller looked at Lewis and then immediately looked

21

at the test booklets. Lewis then knew what Principal Waller was asking him to "get into undetected" the test booklets. Lewis found a key in his desk drawer that opened the room where the tests were kept. Lewis used a razor blade to open the plastic wrapping around the test booklets, copied the test for each grade, and resealed the wrapping using a lighter to melt the plastic. Once Lewis copied the booklets, he placed a copy of the social studies test in Damien Northern's car and a copy of the reading and language arts test in Dorothea Wilson's car.

After the students had taken the test, Lewis changed answers. On one occasion, Crystal Draper came to the room where Lewis was erasing. Lewis assumed Waller sent her. In 2006, Lewis and Draper worked together to change answers. Each year more teachers would assist in the cheating. In 2007 or 2008, Lewis, Draper, and Damien Northern changed answers. Teachers Adrienne Powell, Kimberly Oden, and Latasha Smiley may have also assisted that year. In 2009, the group of cheating teachers grew again. The following teachers were present in the room where the tests were being erased: Crystal Draper, Damien Northern, Starlette Mitchell, Dorothea Wilson, Adrienne Powell and Kimberly Oden. Principal Waller always knew when and where Lewis and the others were altering tests.

Lewis spent one to two hours per day altering tests. At Waller's direction, Lewis cheated every year that Waller was Principal. Waller gave Lewis access to the test booklets before testing started and Lewis made copies of the tests, handed them out and changed answers. During testing week, Principal Waller, Ward, or Reid would tell Lewis to go to the main office. Principal Waller would tell Lewis, "Do what you do." Lewis would get the tests and erase answers.

The teachers only changed answers when Testing Coordinator Kiel was out of the school. In 2007, Kiel noticed things in his office were in a different place than where he had left them. After that, Lewis started taking the tests from Kiel's office into the room next door to change answers. Lewis either marked where the tests had been or took pictures of the undisturbed office, so he could put everything back without Kiel noticing.

2. Crystal Draper (Teacher)

Crystal Draper admitted cheating in 2006, 2007, 2008, and 2009. Draper first changed tests in 2006 and continued through 2009. Gregory Reid usually told her where to go and alter the answer sheets. There was one year when Principal Waller said, "The bins [containing the tests] are in such-and-such room."

In 2009, Reid sent Draper to Kiel's office. Draper erased answers in the office with Damany Lewis, Damien Northern and Dorothea Wilson. While Draper was erasing answers, Sandra Ward and Starlette Mitchell came into the room with a blue cooler, put sixth grade tests in the container, and left the room with the cooler and tests. That same day, Kimberly Simpson, who had not participated in previous years, knocked on the door to Kiel's office. Because Simpson had not helped cheat before, the teachers were not comfortable with Simpson seeing them change answers, so they did not answer the door.

Principal Waller directed Draper to cheat. Tn 2006, Principal Waller told her to go to the room where the tests were kept and change wrong answers to right answers. Principal Waller

22

would often walk by Draper and make comments such as, "I need those numbers." She said that teachers were afraid of Principal Waller because he would punish people if they did not do what he asked.

3. Damien Northern (Teacher)

Damien Northern confessed to cheating in 2008 and 2009 and possibly in 2007 as well. Members of the Parks Middle School faculty cheated the entire time Waller was principal. Waller recruited Damany Lewis and Crystal Draper and directed them to get others involved. In 2007 or 2008, Waller told Lewis to recruit Northern because Lewis needed help. The teachers already cheating included: Damany Lewis, Crystal Draper, Starlette Mitchell, and Dorothea Wilson.

In 2009, Sandra Ward sent Northern to Kiel's office where the tests were kept. When he arrived, he felt there were too many people in the room. He recalls that Starlette Mitchell, Charles Mitchell, Wilson, Draper, Adrienne Powell, and Latasha Smiley were there. Northern did not change answers that year because he did not trust everyone present.

One year Principal Waller