An unusual survey of pilots by NASA suggests that safety problems occur more often than previous government reports have indicated.

But the agency has kept the information bottled up.

That's a mistake. The public has a right to the information. It's also troubling that NASA ordered a contractor to purge survey-related data from its computers.

Part of NASA's mission is maintaining America's lead in aviation. In an $8.5 million project that ended in 2005, the agency surveyed 24,000 commercial and general-aviation pilots on safety concerns.

A NASA official, Thomas S. Luedtke, replied to an Associated Press Freedom of Information request by refusing to release the survey on the grounds the findings might undermine public confidence in the airlines. That's a sad and condescending excuse.

Peggy Gilligan, a Federal Aviation Administration official, said she had not seen the survey but had been briefed on it. She questioned the suggestion that safety problems occur much more often than reported by the FAA.

Raw data from the NASA survey, she said, apparently weren't detailed enough to correct for potential double-counting. For example, separate accounts by two pilots might be based on the same occurrence, she said.

Surely there's a way to sift the survey data to create a meaningful and accurate picture. If that's done and the result is significantly different from the FAA's information, then something important has been learned.

NASA Administrator Michael Griffin says he is reconsidering how much information can be made public. That's a step in the right direction.

Congress, meanwhile, should ensure that the survey data are made public in a form that paints as accurate a picture of air travel as possible.