Chronicle questioned about its numbers

By: Joe Strupp

Was the San Francisco D.A.’s record misrepresented?

When the San Francisco Chronicle reviewed the record of District Attorney Terence Hallinan, who is facing re-election in November, the newspaper zeroed in on Hallinan’s conviction rate ? one of the most telling statistics of a prosecutor’s effectiveness.
On the face of it, the district attorney’s record looked poor, according to a lengthy Chronicle story by veteran reporter Bill Wallace that ran Sept. 2. Wallace wrote that Hallinan’s conviction rate was a dismal 32%, a far cry from the state average of 68.5%.
The piece also compared Hallinan’s numbers individually against other large California counties, such as Los Angeles and San Diego, which boasted rates of 69% and 59%, respectively.
“Hallinan ranks dead last among California’s 58 county prosecutors,” the article’s lead stated, “winning convictions in less than a third of the criminal cases lodged with his office last year.”
But, while the piece drew praise from some for laying out Hallinan’s record for scrutiny, it also raised complaints from others who accused the Chronicle of failing to tell the entire story.
Among the critics was the alternative San Francisco Bay Guardian, a Hallinan supporter, which published a counter-attack in its Sept. 8 issue that blasted its daily competitor for basing its story on the number of convictions vs. the number of arrests.
That ratio, the Guardian contended, does not take into account variables such as cases the district attorney chooses not to prosecute and cases that are subject to other programs, including mediation, drug courts, and community service. The alternative newspaper argued that the conviction rate should be based solely on those cases that are brought to trial, a statistic that would give Hallinan a conviction rate of 90% for 1998.
“The Chronicle story suggests that hundreds of hardened criminals are going unpunished on Hallinan’s watch,” said the Guardian story, by reporter A. Clay Thompson. “In fact, the paper has vastly inflated the number of cases that fall through the [district attorney’s] net.”
The Chronicle’s Wallace ? whose story also examined several individual cases to show how defendants avoided jail time or stiffer penalties, and stated that Hallinan has a higher conviction rate once cases get to court ? defended his approach, saying the methodology used was fair.
“No one has said the statistics are wrong, just that I misinterpreted them,” said Wallace. “It’s clear that they don’t understand the numbers.”
But Thompson said the interpretation is key, especially when the district attorney is employing other methods of dispensing with cases.
“[Wallace] didn’t go in-depth on what they [the numbers] meant,” he said.
The San Francisco situation highlights one of the trickiest areas of journalism ?the use of statistics, which both readers and reporters have long described as useful, but dangerous. Veteran journalism observers said the Chronicle story and the ensuing debate over its methods show that crunching numbers must be done carefully and can almost always be open to dispute.
“I’m reminded of the old line, ‘Figures don’t lie, but liars figure,'” said Anthony Mancini, a former New York Post reporter and professor of journalism at Brooklyn College in New York. “I am suspect of the whole process of statistic-based stories. It is dangerous for newspaper people to think of themselves as mathematicians.”
Mancini said he advises reporters to focus more on the human side of any story, even something like a district attorney’s effectiveness, because numbers don’t always take into account a city’s nuances and differences from neighboring areas. He said a story that looked at Hallinan’s overall record of convictions and diversion programs might be more appropriate.
“Otherwise, you drain a story of its most telling aspect, the human element,” he said. “The flesh and blood.”
George Harmon, news/editorial chair at the Medill School of Journalism at Northwestern University outside Chicago, also cringed at the notion of using statistics too often.
“On a long-term basis, it is probably the largest single abiding problem for journalists,” he said. “They are more often a stumbling block.”
But Harmon said that does not mean statistic-based stories are not useful. “They are a valuable asset,” he said.
Other news veterans, such as editor Doug Clifton of The Plain Dealer in Cleveland, agreed. He said the Chronicle approach was a fair method because it compared Hallinan and San Francisco to other counties using the same approach to the data.
“A statistical analysis of a prosecutor’s performance seems to be a good thing to do,” he said. “You have to make the comparisons as even-handed as possible. If you do that, it is a good weapon in the arsenal.”
Chronicle managing editor Jerry Roberts agreed that statistical stories can be misleading if they are not carefully pieced together.
“There is a danger if you do a quick-and-dirty job and don’t subject it to scrutiny,” he said. “But it remains a good tool for going beyond the anecdotal to provide evidence.”

(Editor & Publisher [Caption]
(copyright: Editor & Publisher September 18, 1999) [Caption]

Follow by Email
Visit Us

Leave a Reply

Your email address will not be published. Required fields are marked *