Last spring, Newark officials decided their projection for that year’s graduation rate was too good to keep to themselves.

So in June, months before submitting final data to the state, Newark Public Schools officials announced that the district was “trending toward a record graduation rate” in 2018. Privately, they expected the rate to inch up from 78.5 percent in 2017 to as high as 80 percent in 2018, marking the fifth straight year of gains.

Instead, when the state publicly released the official numbers last week, the district’s rate had actually slipped nearly three percentage points, to 75.7 percent.

What happened?

No one who might know has given a public explanation. Most of the district’s top officials who oversaw the class of 2018 were fired or left the district last year after Superintendent Roger León took over in July. León has not yet publicly discussed the graduation numbers, and his spokeswoman did not respond to multiple requests for comment.

But former officials have a theory about what may have caused the decline, based on what they say was behind the graduation rate increases in recent years: summer school and “data cleaning,” or purging the rolls of students who previously left the district. The former officials, who declined to go on the record, suggested that students who graduated over the summer last year may not have been added to the official data, while former students may not have been removed — either of which could have dragged down the final rate.

While unproven, the theory sheds light on the intense work that happens behind the scenes in schools and districts to make sure data reflects reality — efforts that can improve graduation rates, even if they don’t directly affect student learning.

“This kind of thing has little direct connection to the quality of schooling provided to children,” said Ryan Pfleger, a former district researcher who was among those forced out by the León administration, in an email, “but could easily drive Newark’s rate up or down 3 points in a year.”

Each year, hundreds of Newark high schoolers attend summer school, where they can earn credits for classes they failed during the school year. Those condensed makeup courses, which can be completed online or in a classroom with an instructor, are known as “credit recovery.” Last summer, 2,045 high school students in grades 10 through 12 were enrolled in credit-recovery courses.

If a relatively small number of seniors earn the credits over the summer needed for a diploma, that can move the district’s graduation rate. In 2017, an additional 48 graduates would have raised the districtwide rate by 1 percentage point, state data show.

“If schools leverage summer school for credit recovery and have students enrolled in credit recovery courses throughout the year,” said a 2017 district memo to school leaders, “we can expect an increase in our overall graduation rate.”

In fact, the final 2017 graduation rate, which included the summer graduates, was three percentage points higher than the rate that June, according to district documents obtained by Chalkbeat. Former officials attributed the gains largely to seniors who recouped missing credits during the summer.

The district officials were anticipating a similar crop of summer graduates last year when they publicly predicted that the final 2018 graduation rate would increase, continuing the upward trend and mirroring a statewide uptick in 2018. So why did the rate fall back in Newark?

Although experts caution against reading too much into a one-year slip, it’s possible that fewer seniors than expected earned diplomas between June and August. The district has not released that data.

The former officials have a different explanation: reporting lapses.

School administrators are responsible for entering summer graduates into a central database before the district submits final graduation data to the state by Aug. 31. In previous years, district officials held presentations in August for principals reminding them to update this data.

It appears that the new administration, which took the reins of the district only in July and was under pressure to make sure schools opened smoothly in September, may not have given a similar presentation last year. Officials did still send out reminders, according to one high school principal.

“I don’t remember a formal presentation about it,” the principal said. “But I do know there was correspondence sent about updating your cohort with the final data.”

PHOTO: Patrick Wall/Chalkbeat
A. Robert Gregory, the district’s former interim superintendent, said in May 2018 that “cleaning” the data to remove students who left the district had helped raise the graduation rate.

Schools must also remove any students from their rosters who left the district and enrolled elsewhere. To do that, they must prove that the students entered other schools — otherwise, those students are treated like dropouts and count against the school’s graduation rate.

That verification process stems from a federal law that went into effect in 2011. The rules have helped make New Jersey’s graduation data more reliable, experts said. But they have also placed a heavy burden on schools to track down former students and confirm that they have enrolled in new, often faraway schools.

“When students go to other states or other countries, it sometimes becomes hard to document that,” said Stanley Karp, a former high school teacher who directs the Secondary Education Reform Project at Newark’s Education Law Center. “You have to keep a kid on the roll until that documentation is completed.”

“It’s a factor every year” in district graduation rates, he said, though he added that it is unclear how much rates are affected by students who leave the district but remain on the roster. He also noted that districts can appeal their graduation rates in the fall if they can show that certain students should not have been included in the district’s count.

While mostly invisible to the public, officials say that schools’ efforts to clean up their rosters and the accompanying data can have a significant impact on their graduation rates — especially in urban districts such as Newark, where some students move and change schools frequently.

At a conference last May at Rutgers University-Newark, the district’s interim superintendent at the time, A. Robert Gregory, said the main driver of the district’s graduation rate gains in recent years was better student tracking. Schools were asked to periodically check students’ progress toward meeting graduation requirements, Gregory said. But, crucially, they were also regularly reminded to remove any students from their graduating class who had enrolled in other districts.

“The most important step we took was getting school leaders a list of every name in their graduation cohort,” Gregory told the conference. “That allowed them to clean their data up.” (Gregory left the district last fall.)

It’s unclear whether the district’s annual push to clean up high school rosters in August was disrupted by the change in administrations. Experts cautioned that, for now, there is no evidence that shows reporting lapses contributed to last year’s graduation rate decline.

Whatever the cause, the decline follows years of steady increases in Newark and other high-poverty districts across the state — an upward trajectory that could still continue even after last year’s dip.

“In general, I would be reluctant to take one year and say that a trend has shifted,” said Mark Weber, a lecturer at Rutgers University and an education policy analyst, who has studied the district’s graduation rate. “Let’s see what happens in five years.”