The Great Divide is an investigative team that explores educational inequality in Boston and statewide. Sign up to receive our newsletter, and send ideas and tips to firstname.lastname@example.org.
After a Tuesday Globe story showing city auditors found Boston may have overstated its graduation rate for five of the last seven years, a newly appointed School Committee member on Wednesday called for an expanded audit of students the district claimed had left BPS but didn’t drop out.
During Wednesday’s School Committee meeting, Brandon Cardet-Hernandez asked for a new audit “so that we can make sure the data is accurate” and so “these sort of stories don’t float around and ... in a time of enrollment decline further hurt our position.”
As Boston school leaders celebrated rising graduation rates, Boston’s auditors found school officials wrongfully and repeatedly removed dozens of students from graduating classes by claiming, without sufficient proof, that they transferred to another school, moved to another country, or died. In each of those instances, auditors could not find the paperwork to support the reasons for the students’ departures, even though federal rules require such documentation to ensure accurate graduation rate counts.
Auditors in 2021, for example, found the district didn’t provide documentation for 16 out of the 40 students randomly chosen for the sample. The year before, data was missing for 15 out of 40 students sampled, according to the audits reviewed by the Globe.
Students who drop out lower a district’s graduation rate, which is used by the state to measure a school or district’s performance. By counting students who dropped out or went missing as transfers, a district can make the graduation rate appear better than it is.
A typical BPS graduation class starts out with around 4,000 students. For the graduating class of 2020, about 400 dropped out while 122 transferred out of state, according to state data. Another 387 students transferred to another school within Massachusetts and 8 students died.
“We are committed to the law and following the law and holding ourselves accountable for accurate reporting,” Superintendent Brenda Cassellius told the School Committee Wednesday night in response to the story.
She also said the expectations for documenting students’ departures were onerous. “The threshold for what is acceptable and accepted documentation is set at a very high level, which can make it practically difficult to produce,” she said.
Documentation can be as simple as writing down a conversation with a parent, according to federal guidelines.
But she later said that staffing might play a role in the district’s inability to collect sufficient documentation and proof. “Many of our schools don’t have registrars,” she said, adding that school principals often end up having to track down these documents.
Cassellius also raised questions about how much the findings matter. “The sample is not intended to be statistically valid,” Cassellius said. “It can be difficult to extrapolate how missing paperwork for a handful of students would impact a graduation rate.”
In response, Cardet-Hernandez, a former public school principal, acknowledged how difficult it can be to track down students and find sufficient documentation for their departure, but asked Cassellius if the district could perform another audit, this one of a larger sample of students.
“Even if it’s a perception and not reality ... is there any way that we could show a statistically significant sample to rebuild that trust?” he asked, referring to the audit.
Cassellius didn’t say whether she would commit to audit the graduation rate again, but said the district is working to hire its own auditor as part of a new four-person “risk management team.”
“These four individuals will work together to provide the kind of controls that we need for data integrity,” she said. They will “look at our systems and processes and to give us an accountability check on ourselves.”
Bianca Vázquez Toness can be reached at email@example.com. Follow her on Twitter at @biancavtoness.