At the April 17 hearing, Ellerbe testified that his investigators had been able to solve 72.7 percent of arson cases — more than three times the national average — which he attributed to more experience, not a change in measurement. But his aides subsequently acknowledged that they had changed the way they classify fires as arsons in response to inquiries from The Washington Post.
Prior to the change, they explained in interviews, all “incendiary” fires — those that had been deliberately set — were generally considered cases of arson, consistent with national standards. After the change, the department began drawing a distinction between “incendiary” structure fires — those deliberately set — and “arsons” — those in which evidence existed to establish “willful, malicious intent to start a fire” that was sufficient to support an arrest.
The change resulted last year in a dramatic decrease in the number of arsons and a dramatic increase in the arson closure rate, or the percentage of cases in which arrests were made.
Quander’s remarks came in response to a harsh line of questioning by council member Tommy Wells (D-Ward 6), who said that Ellerbe, when earlier asked about a closure rate three times the national average, could have acknowledged “a big discrepancy” and agreed to review his testimony.
“But instead, he stood by information that stood out as being suspect,” Wells said.
Later this week, the committee plans to formally request written answers from the department about discrepancies arising from the new and old means for categorizing arsons and calculating closure rates, said Charles Allen, Wells’s chief of staff.
Wells previously criticized the D.C. Fire and Emergency Medical Services Department for providing inaccurate data to the committee in February on the number of firetrucks available for service.
The change in the department’s method for categorizing arsons, first reported by The Post on Sunday, resulted in a nearly 80 percent drop in arsons from 2009 to the 2012, when the department reported only 32 arsons — a figure significantly below the national average.
At the same time, the department said in performance reports submitted to the council that its arson closure rate had surged to 72.7 percent, more than three times the national average. Department officials later clarified that figure and explained that it had been based on only partial 2012 data and should have reflected the final year rate, which was 34 percent.
At the April 17 hearing, Fire Marshal Bruce Faust, the department official who oversees fire investigations, told the committee that he had compiled a “very different” closure rate. The numbers compiled by Faust, submitted to the D.C. Office of the Chief Financial Officer, used the old method in which “incendiary” fires and “arsons” were essentially one and the same and showed an arson closure rate of only 9.6 percent for 2012.
On Monday, Quander testified that Faust was the “person most responsible” for data about arsons and that Faust had “made some changes” as to how arsons were counted. Quander did not elaborate on how the department would report different figures than the analysis compiled by Faust.
Prior to Quander’s testimony, Faust issued a written statement, saying that he wanted to “clarify that I am not at odds with the Chief or the administration of the department on this topic.”
“We have been working diligently to ensure the reporting of Incendiary fires is done with accuracy and in meaningful terms according to nationally recognized practices,” he said. “The only goal is to be consistent, accurate and transparent in our reporting. The value of good data, allows benchmarking with other comparable cities, as well as performance management within our department.”
He did not respond to a request for comment after Quander’s testimony.
During the hearing, Wells said the department’s reliance on faulty data related to arsons and closure rates is problematic at a time when the council is considering the department’s plans to eliminate two full-time investigators in the proposed 2014 budget. The investigative unit is operating at only about half of its budgeted staff.
“We have a budget that says how many staff is needed to determine arson,” Wells said during the hearing. “If we’re measuring it differently but don’t know it, then we’re going to make a bad decision about our resources.”