II.9.6 Definition of "Used"

Criticism: When policy and science are often inextricably linked, which research done today will be needed for tomorrow's policy? How long will researchers have to retain their data?

Response: Circular A-110, as currently effective, requires that researchers retain their data for three years. Regardless of whether the data are more than three years old, if the data are still in the researcher's possession, the government has a right to request the data.

Because we believe that three years is too short a period for the public to be able to access data that are used as the basis of federal policy, CRE has proposed that such data be retained by the awardee for a longer period. See Section II.8.1, supra.

CRE favors a framework where government decision makers are prohibited from relying on studies for which underlying data are unavailable. To the extent that current or future government rules or policies are developed on the basis of studies for which no data are available due to passage of time, government decision makers may not be getting the full picture. CRE therefore urges OMB to discourage agencies from implementing rules based on research findings for which no underlying data are publically available.

With respect to the mixing of older and newer data, and data used for developing policy or rules versus data not so used, these issues may need to be addressed on a case by case basis. See also sections II.6.4, II.6.5, supra, for definition of "used . . . in developing policy or rules."

Criticism: Would it not be possible under the new law and the proposed definition of "used" for a member of Congress, citing a particular study, to "introduce" legislation relating to that study merely for purposes of permitting a FOIA request to be filed for the data?

Response: One would hope that this is an extreme and unlikely scenario. A member of Congress who participated in such a nefarious scheme would be obliged to answer to the voters, and we leave the results of that debate to the electorate. However, it clearly would be unwise to adopt a data access process that requires agencies to speculate on the motivations of individual members of Congress in introducing legislation. If members of Congress have sufficient confidence in a study to cite the study as the basis for legislation, then opponents of the legislation should have the right to examine the data underlying that very important study.

The criticism also touches on the definition of "published." If the researcher who performed the study in the example given is confident enough in its conclusions to allow the study to be disclosed and used in this manner, then yes, the study should be open to fair examination by the public.