The National Science Foundation has postponed the release of a major annual report on doctorate attainment as it works to reconcile its concerns over student privacy with public demands for detailed data on minority groups, NSF officials said in a statement issued today.
The NSF typically releases the full results of its annual Survey of Earned Doctorates in late November. This year, however, the agency released only a paper with selected findings from the survey, which is sponsored by several federal agencies and conducted by the National Opinion Research Center at the University of Chicago. At the time, NSF officials said the full report would not be issued until May or June 2009, but did not explain the delay.
In a statement sent today to The Chronicle, the agency’s division of science-resources statistics, which oversees the survey, blamed the delay on its efforts to revise the report in response to complaints that last year’s edition lacked sufficient data dealing with the race, ethnicity, citizenship, and gender of doctorate recipients.
The NSF statement said last year’s report, which covered data for 2006, had itself been revised to try to protect the confidentiality of survey respondents. Its concern at the time was that some segments of the population account for such a small share of enrollments in doctoral programs that it would be possible for readers to attribute certain survey findings to individual people. In deciding not to present findings based on responses from such small populations, the agency ended up producing a report with less information than those issued in the past.
The statement said the agency had subsequently “received many complaints” from people who use the report “concerning the reduced availability of information about underrepresented minorities.” The NSF ended up re-releasing sections of the report with the omitted data restored, and, over the past seven months, has been conducting a Web-based survey and holding meetings with people who use the report to try to determine what information they need.
The statement said the division of science-resources statistics would “redesign key statistical tables in a way that maximizes the reporting of information on underrepresented minorities while protecting individually identifiable information,” and then run its approach by a National Research Council panel dealing with statistics before issuing the full report on 2007 data late next spring. The next full report, based on this year’s data, should come out on schedule, in November 2009. —Peter Schmidt
