WASHINGTON - Abandoning its secrecy claims, NASA promised Congress on Wednesday it will reveal results of an unprecedented federal aviation survey which found that aircraft near collisions, runway interference and other safety problems occur far more often than previously recognized.
The agency's chief also said, however, that before any release NASA would scrub the data to make sure none of the 24,000 pilots who were interviewed anonymously could be identified, taking until the end of the year to do what a survey expert told Congress could be done in a week.
Provoking broad criticism, NASA had said previously it was withholding the information because it feared it would upset air travelers and hurt airline profits. NASA cited those reasons in refusing to turn over the survey data to The Associated Press, which sought the information over 14 months under the Freedom of Information Act.
"We did say that, and that was the wrong thing to have said," NASA's administrator, Michael Griffin, testified during an oversight hearing. "I apologize. ... People make mistakes. This was a mistake."
Lawmakers from both sides were harshly critical. Rep. Bart Gordon, D-Tenn., chairman of the House Science and Technology Committee, said NASA's reasons for withholding the research were "both troubling and unconvincing."
"This appears to be a mess of NASA's own causing," said Rep. James Sensenbrenner, R-Wis., the top Republican on the committee's oversight and investigations panel. "You've dug yourself in a hole. I can't say you're not digging yourself deeper from what I've heard at this hearing."
Griffin said his agency will release the research data that does not contain what he described as confidential commercial information. He said NASA spent US$11.3 million on the research.
"The survey results we can legally release will be released, period," he said.
Griffin's end-of-the-year timetable wasn't fast enough for some lawmakers.
"Shouldn't it be a priority to your agency to scrub this and get this out to the public immediately?" asked Rep. Jerry Costello, D-Ill. He said NASA "made a huge mistake in how they responded to the AP and to the media."
Another lawmaker, Rep. Harry Mitchell, D-Ariz., asked: "Why does it take a hearing in Congress and public pressure for a hearing to get this information made public?"
In an odd twist, Griffin raised doubts about the reliability of his own agency's research by telling lawmakers that NASA does not consider the survey's methodology or data to have been sufficiently verified.
Griffin confirmed NASA's research project showed many types of safety incidents occurring more frequently than were reported by other U.S. government monitoring programs. But he cautioned that the data was never validated and warned, "There may be reason to question the validity of the methodology."
"We did not manage that project well," he told Congress. "We will fix it and we will try not to do it again."
"What I'm hearing you say is, we've just thrown US$11 million down a rathole," said Rep. Ben Chandler, D-Ky.
"I hope that is not the case, and I believe we should be able to get much that is useful from this data," Griffin said. "But there will be cause to question it from knowledgeable aviation experts."
Experts who worked on the study say it adhered to the highest survey industry standards. The research was "state of the art," said Jon Krosnick, a Stanford University professor who helped create the survey questions. Disputing Griffin, Krosnick told Congress that aviation experts from NASA and the Federal Aviation Administration, and the White House Office of Management and Budget and other groups reviewed the research plans. He said further scrutiny would not have been helpful.
"These peer review processes rarely yielded significant changes in the survey process," Krosnick said.
NASA's former head of the research project, Robert Dodd, told lawmakers the survey was based on "outstanding science," extensively tested and ready for meaningful analysis. Dodd said NASA's earlier explanations for withholding the information were "without merit."
"I don't believe that the ... data contained any information that could compare with the image of a crashed air carrier airplane or would increase passengers' fear of flying," Dodd said.
On Tuesday, Griffin bowed to a request from the lawmakers and sent copies to Capitol Hill of the raw data contained on four CDs. At the hearing, Gordon held the discs aloft and asked Griffin to identify where in NASA's data could pilots be identified based on what they told researchers anonymously.
"We couldn't find it," Gordon said, adding that NASA's lawyers also were unable to identify any such examples.
"When we looked at the data, we do not believe at this point the data has been scrubbed sufficiently," Griffin said.
Krosnick said the identities of pilots could be derived only in a "very small number of cases" and said this identifying information could easily be removed.
Officials who have worked on the survey have said it contains no pilot names or airline names. The questionnaire asked pilots to state how many times in the previous 60 days they had encountered a wide range of problems with equipment, weather, tower communication and other safety issues.
NASA's efforts to withhold the safety research earlier sparked tough talk on Capitol Hill and in the editorial pages of dozens of leading newspapers -- including USA Today, The New York Times and the Chicago Tribune -- which urged the agency to release its research. The Times described NASA's reasons for withholding the information as "lame excuses."
Griffin also has sought to assure lawmakers that NASA will not destroy the research. Earlier this month, NASA ordered the contractor that conducted the survey to return any project information, then purge all related data from its computers. Griffin said he has rescinded those instructions.
Although to most people NASA is associated with spaceflight, the agency has a long history of aviation safety research. Its experts study atmospheric science and airplane materials and design, among other areas.
The survey project, called the National Aviation Operations Monitoring Service, was launched after a White House commission in the late 1990s called for government efforts to significantly reduce fatal aircraft accidents.
