Rom, Louis, The Quill
When Philip Meyer published his book, "Precision Journalism," in 1973, only a few newspapers had used computers to assist in their reporting.
Meyer's book introduced serious journalists to the idea of using social science methodology to uncover stories that public information officers, police chiefs and mayors would not willingly offer. Using tools such as scientifically valid surveys - just like today's modern-day election polls - he encouraged journalists to tell more accurate stories and depend less on unreliable "he said, she said" hyperbole.
While covering the Detroit riots in 1967 for the Detroit FreePress, Meyer was among the first to use computer-assisted data analysis. He and two colleagues conducted a scientifically valid survey - one with a margin of error of plus or minus 4 percent of African-Americans by crunching numbers into an IBM 360, a mainframe computer about the size of a banquet table. The story, which earned Meyer a Pulitzer Prize, revealed that people who had attended college were just as likely to take part in a riot as high school dropouts.
Over the next several years, a handful of major newspapers used computers to analyze data for stories. The Miami Herald was the first newspaper to use computers to analyze government records. The Philadelphia Inquirer analyzed sentencing trends for a report called "Unequal Justice." The New York Times used data analysis to show that black people were eight times more likely to be murdered than white people.
It was, Meyer thought, the dawn of a new day in journalism. Soon, he predicted in his book, reporters would regularly cull data from thousands of electronic records and identify patterns and nuances that otherwise would not see the light of day.
And, in fact, much of what he wrote about has become reality. Today's newsrooms are blanketed with high-speed, incredibly powerful computers and software programs that make data analysis more feasible. The number of journalists who have the technical know-how to take advantage of these resources is growing, and the 2000 census led to an even larger push from newsrooms to train their staff on the benefits of CAR.
"I think a lot of what (Meyer) thought about has come to fruition," said Ron Nixon, training manager for Investigative Reporters and Editors (IRE) and the National Institute for Computer-Assisted Reporting (NICAR). "If you look at the recent census stuff, you have people using the Gini coefficient of income equality and dissimilarity indexes, and that's something sociologists have used for years, but we've never looked at."
But Meyer, now the Knight Chair of Journalism at the University of North Carolina, Chapel Hill, is less impressed.
"My original intent was not to push computers - that's a nobrainer - but to push the idea of journalism as social science ... And I thought it would diffuse much more rapidly. The use of computers diffused rapidly once PCs became available, but the higher level applications of statistical analysis and interpretation of scientific method haven't diffused very much," said Meyer.
Meyer said if the newsroom is a different place today than 30 years ago, it's not because it's full of computer geeks. He says most undergraduate students are focused on being great writers but have few statistical and computer skills "beyond finding and downloading a spreadsheet."
"Today's college students are very much into the storytelling side," he said. "But we've had a hard time convincing them to invest time in learning social science skills."
CAR GAINS MOMENTUM
It's been four years since an academician or journalist conducted a scientifically valid survey of CAR use in the newsroom. The University of Miami's Bruce Garrison conducted such a survey six times between 1994 and 1999. "We stopped when we didn't see much change," Garrison said. His last survey found that CAR was still predominantly a big paper phenomenon, though it was trickling down to smaller papers. …