The National Center for Education Statistics has delayed the release of three completed surveys of students, teachers, and faculty members pending reviews to ensure that individual respondents cannot be identified.
The center also is inviting researchers to help it revise its access policies to conform with confidentiality rules mandated in the Hawkins-Stafford Elementary and Secondary School Improvement Act of 1988.
Under the Hawkins-Stafford act, which reauthorized the n.c.e.s., the agency is required to guarantee the confidentiality of survey respondents. Researchers who violate the confidentiality of their subjects can be fined up to $250,000 and imprisoned for up to five years.
Emerson J. Elliott, acting commissioner of the n.c.e.s., said the release of the three surveys has been delayed six months or more as researchers analyze data tapes to ensure that individuals cannot be identified, either directly or through sophisticated cross-checks of data.
Databases that could be used to identify individuals will be extracted from surveys and kept confidential.
Currently withheld from release, pending disclosure analyses, are:
The National Education Longitudinal Study of 1988, which includes baseline data on 8th graders that year.
The Schools and Staffing Survey, a new study that includes data on state samples of teachers and administrators, as well as a nationally representative sample of private-school teachers.
The National Survey of Postsecondary Faculty, also a new study.
The review of the surveys, which is slightly more than half completed, involves the time-consuming analysis of individual data tapes and “will run into several thousand dollars before it is done,” Mr. Elliott said.
“The new laws and procedures were not in place at the time these surveys were designed,” Mr. Elliott said. “The more we look at it, the more we believe our previous practices were not adequate to protect confidentiality.”
“We did take safeguards, but they were done primarily through contractors,” Mr. Elliott added.
Under the new rules, contractors with access to data will have to take oaths to protect confidentiality. The n.c.e.s. will have the ultimate responsibility of reviewing studies to ensure confidentiality is maintained.
Mr. Elliott said researchers’ reaction to the new rules has been “quite sharp,” with some suggesting that the center shut down if it can no longer provide them with access to the data they need.
Mr. Elliott recently invited researchers to comment on the new rules and to suggest ways his agency can balance its confidentiality concerns with its mandate to provide data for public use. He suggested researchers may wish to sign affidavits vowing to maintain confidentiality or allow the n.c.e.s. to perform tabulations upon request.
Gerald E. Sroufe, director of the governmental and professional liaison program for the American Educational Research Association, said the n.c.e.s.'s approach to the problem is “responsible and reasonable.”
