By Benjamin Herold and Dale Mezzacappa
for the Notebook/NewsWorks
Updated: see note below
In 2009, Strawberry Mansion High School appeared to be something of a miracle.
A neighborhood high school in a rough part of North Philadelphia, Mansion saw more than two-thirds of its students score "proficient" or above on that year's state standardized tests.
Today, though, compelling evidence indicates that Mansion's unusually high test results were driven by adult cheating.
Documents obtained by NewsWorks and the Notebook show Mansion students raised alarms about cheating as far back as 2005. Suspicious patterns of "wrong-to-right" erasures on test sheets, a tipoff for adult cheating, have also been spotted in both reading and math for the 2009 and 2010 testing cycles.
Listen to Ben Herold's NewsWorks Tonight interview
Now, Lois Powell-Mondesire, the principal who presided over the school's testing gains, is no longer there, and Mansion's test scores are back in the tank. Barely 10 percent of students scored proficient or above on last spring's exams.
But Powell-Mondesire wasn't disciplined.
She was promoted.
In 2010, Powell-Mondesire was tapped for a new central office position advising other principals on how to turn around struggling schools. While Mansion struggles to get back on its feet, Powell-Mondesire will make more than $145,000, just shy of the top possible salary for a Philadelphia principal.
A statewide probe involving 53 regular city schools, four area charters, and a number of other districts across Pennsylvania is still ongoing. It is possible for cheating at a school to occur without a principal's knowledge or involvement. No proof of cheating by specific administrators has been made public, and no Philadelphia school administrator has been charged with wrongdoing.
But in a number of cases, telltale signs of adult cheating at specific schools correlate to the tenures of individual principals. Some of those principals were rewarded by the District after overseeing eyebrow-raising test-score gains.
That practice has continued, even after District officials became aware of evidence of suspicious erasures. For example, Barbara McCreery, who oversaw dubious test score increases at her former school, Communications Technology High, was installed just this summer as the new principal at Bok Technical High.
"I think this is willful blindness," said Michael Josephson, a leading national expert on ethics and education and the president of the nonprofit Josephson Institute of Ethics.
"There's a deliberate looking the other way," he said.
Fairness or slow reaction?
Powell-Mondesire did not respond to requests for comment.
But principals' union president Robert McGrogan was adamant that it is unfair to associate individual administrators with cheating until an investigation being led by the Pennsylvania Department of Education is complete.
"I believe that the media is trying people, and I think that is inappropriate," said McGrogan, the president of Teamsters Local 502, Commonwealth Association of School Administrators.
"An investigation is under way. The truth will be made known. People need to be patient until we reach that time," McGrogan said.
Cheating on state tests can be conclusively proven only by confessions or credible firsthand accounts. Seeking that kind of hard evidence, state and District investigators have been interviewing staff at about a third of the 53 District schools where there are signs of cheating.
Pennsylvania Department of Education spokesperson Timothy Eller would not comment on the investigation.
The District is taking the lead on the cheating probes at 20 so-called Tier 2 schools. District spokesperson Fernando Gallard said the District would make no comment on the cases of individual educators until those investigations are complete, likely by the end of December.
David Adamany, the former president of Temple University who is serving as an unpaid "testing integrity adviser" to the District, called that caution appropriate.
"The idea in the District is to have this so well nailed down, you're sure to get the right people," Adamany said. "If they bring charges and charges don't stick, you do terrible damage to someone's reputation and open up the impression that the School District is unable to trace wrongdoers and get them out.
"There's a lot of caution being exercised, and it's heartbreaking we cannot move faster."
Josephson, however, calls the lack of action problematic.
"Of course, we want to be fair," he said.
Update: Michael Josephson has asked that this quote be clarified:
"But if fairness prevents you from making decisions that are obvious or important, it perpetuates [cheating] behavior." to say
"But a fair result can be achieved expeditiously and dragging out the process is often an excuse for not dealing with a problem," he said. "In this case it seems they have or can easily get the information they need to make a fair decision and process is being used in a way that is excusing the failure to make an obvious and important decision and this will perpetuate cheating behavior." End update
A trail of suspicious erasures
Edward Penn opened the new school year as one of the highest-paid principals in the city. Yet forensic and statistical evidence points to possible cheating at two schools while he was principal.
In 2008-09, Penn was the principal of Thurgood Marshall Elementary. He earned $96,634, according to information obtained by NewsWorks and the Notebook under Pennsylvania's Right to Know law.
That year, Marshall, a K-8 school in the city's Olney section, saw big gains in the percentages of its students scoring proficient or advanced on Pennsylvania System of School Assessment (PSSA) exams. (The other possible ratings are basic and below basic.)
Schoolwide, the rates of students scoring proficient or above in math and reading shot up 17 percentage points. In 4th grade, reading scores rose 37 points. In 6th grade, math scores rose 40 points.
The scores dropped slightly the following year, while he was still principal.
In summer 2010, Penn was promoted to the same new post given to Lois Powell-Mondesire: "turnaround principal."
To support the leaders of the District's six initial Promise Academies – struggling schools targeted for massive intervention in the hopes of generating quick improvement.
With the promotion, along with a shift from being a 10-month to a 12-month employee, Penn got a 42 percent raise, to $137,311 per year.
Shortly afterwards, he was installed as the principal at Clemente Middle School, a new Promise Academy off to a rough start.
In Penn's first year, Clemente saw modest school-wide gains, with proficiency rates rising 4 percentage points in reading and 10 points in math. Gains in some grade-subject combinations were more dramatic. In 6th-grade math, proficiency rates rose 30 points; in 7th-grade reading, they rose 20 points.
Then, last August, the state Department of Education commissioned a statewide analysis of PSSA answer forms from 2009, 2010, and 2011. The goal was to find schools where a high number of student response sheets had unusual numbers of incorrect answers that had been erased and changed to the correct answer – a strong signal of the worst type of adult cheating.
That analysis revealed big trouble at Thurgood Marshall Elementary during Ed Penn's time there.
According to confidential documents and information obtained by NewsWorks and the Notebook, Marshall's 2009 PSSA results were flagged for suspicious "wrong-to-right" erasures in seven out of a possible 12 grade-subject combinations: in both math and reading in grades 3, 4 and 6 and in math in grade 7.
In 2010, the school was flagged four of a possible 12 times. Proficiency rates dropped.
In 2011, after Penn left Marshall, the suspicious erasure patterns went away, and test scores kept trending downward.
That same year, similar signs of cheating appeared at Clemente, now under Penn's leadership for the first time.
After not being flagged even a single time for suspicious erasures previously, Clemente was flagged six of a possible eight times for its 2011 results.
In grades 5, 6, and 7, where the signs of cheating appeared, Clemente's test scores went up in both reading and math.
In 8th grade, which was not flagged, scores went down 14 points in both subjects.
As a so-called Tier One school, Clemente has been a target of investigators from the the state Office of the Inspector General. No findings have yet been made public.
Penn did not respond to requests for comment.
A NewsWorks/Notebook analysis showed other examples where evidence of improper erasures correlated with the tenures of individual principals.
In 2008-09 and 2009-10, Michelle Burns was the principal of Tilden Middle School in Southwest Philadelphia.
In both years, Tilden was flagged for suspicious erasures in every possible grade and subject.
In summer 2010, Burns was reassigned to be principal at Kensington Urban Education High School. She also went from being a 10-month to a 12-month employee. Her annual salary rose 25 percent, to $134,775.
In 2011, the first year after her departure, the suspicious erasure patterns at Tilden stopped. The school's PSSA scores plunged 28 percentage points in both reading and math.
Burns did not respond to requests for comment on the evidence of possible cheating at Tilden.
By 2010, Barbara McCreery's last year as principal at Communications Technology High, 11th-grade proficiency rates had soared to 75 percent in reading and 70 percent in math.
The school's results were flagged for suspicious erasures in reading and math in both 2009 and 2010.
In 2011, the first year after McCreery's departure, the signs of cheating at Comm Tech went away. That year, the school's proficiency rates plummeted, dropping 38 percentage points in reading and 45 percentage points in math.
McCreery spent the last two years filling in at other schools, mostly on special assignment. Last spring, the state decided to have the Inspector General do an on-the-ground inspection at Comm Tech.
This summer, the District named McCreery the new principal at Bok Technical High.
District officials would not comment on her new assignment.
McCreery did not respond to requests for comment.
This year, she'll make $142,724.
Signs of trouble not heeded
Long before Strawberry Mansion High's test scores reached their peak in 2009, Powell-Mondesire started tussling over testing issues with one of the city's student organizing groups, Youth United for Change.
In public, YUC members testified about excessive and inappropriate test preparation at Mansion. They were particularly displeased with the school's practice of pulling select students out of their core classes for intensive test-prep drills.
In private, according to documents, YUC students and staff alleged cheating by staff at the school. They shared with high-ranking District officials the results of an extensive but unscientific survey in which almost one-third of the 153 Mansion students who responded said they had received help on standardized tests.
The complaints led to some districtwide changes to how schools handled test preparation and administration. But the district does not appear to have taken any action regarding Powell-Mondesire or other adults working at Mansion at the time.
Three years later, a staggering number of Mansion's student PSSA test answer sheets showed extremely high numbers of suspicious "wrong-to-right" erasures, according to confidential documents and information obtained by NewsWorks and the Notebook.
State officials say the average PSSA response sheet might have one wrong answer that is changed to the correct answer.
At Mansion in 2009, 78 percent of student response sheets on the math exam had five or more such erasures. In reading, 47 percent of student response sheets had five or more suspicious erasures.
Statistically, it is all but impossible that those patterns occurred by chance.
In 2010, Powell-Mondesire's last year at the school, Mansion was again flagged for suspicious erasures in both math and reading.
The next year, the flags disappeared in reading, though not in math.
By 2012, Mansion's fall from grace was complete; math proficiency rates were down about 60 percentage points and reading proficiency rates were down more than 50 percentage points from their 2009 pinnacle.
No hard evidence of cheating at Mansion has been made public. Even if such evidence were found, it might not implicate Powell-Mondesire or other senior administrators at the school.
Scores have rippling effects
But Mansion's dubious PSSA scores from 2009 and 2010 likely shielded it from some of the harsh consequences recently experienced by other struggling neighborhood high schools.
In both 2010 and 2011, Mansion was spared while seven of Philadelphia's other neighborhood high schools with chronically poor test scores were either reconstituted or converted into charter schools as part of the District's Renaissance Schools turnaround initiative. At the time of intervention, those schools - Audenried, Germantown, Simon Gratz, Martin Luther King, Olney East and Olney West, and West Philadelphia – all had test scores similar to, or even better than, Mansion's 2012 results.
And just last spring, the School Reform Commission voted to close two of Mansion's neighboring high schools, FitzSimons and Rhodes. Poor academic performance was one consideration for the decision. Those schools' students are now being sent to Mansion.
Josephson, the ethics expert, called school systems' lack of action in the face of dubious results like Mansion's "cowardly." He said such inaction can breed cynicism and mistrust among staff, parents, and the larger community.
And experts agree that the greatest harm from cheating is done to children– a point acknowledged recently by incoming Philadelphia Superintendent William Hite.
Adult cheating "impacts our students in such a way that we have no idea whether or not they have learned anything," said Hite.
"I don't think that's fair to them, and that's certainly not how I want this School District represented across the state or across the nation."