Number 33, April 2002

Editorial

Flying Blind:
The Data Quality Act and the Atmospheric Sciences

Imagine that medical scientists developed in a laboratory a new drug that they believed held great promise of societal benefit. Could they then hang a shingle and begin offering the drug for sale to the general public? Of course not. Because of the potential for unexpected, adverse effects, all new drugs must go through some form of testing to evaluate costs and benefits before they are approved.

But in the world of policy, unlike the world of medicine, there are frequently times when dramatic interventions are introduced with no prior systematic consideration of their potential effects. The "Data Quality Act," to become law October 1, 2002, is one such intervention. According to its supporters the Act promises to revolutionize the role of science in policy making by "ensuring and maximizing the quality, objectivity, utility and integrity" of scientific information. For a summary of the Act and related references, see the WeatherZine news item on the Data Quality Act that follows.

Few would question the goals of quality, objectivity, utility and integrity of information in the Data Quality Act. One might expect that, as occurs in medicine, some period of analysis of the effects of a particular policy intervention would precede broad implementation. But with the Data Quality Act no such analysis has occurred. (While the NRC is studying the law, the law will go into effect regardless of the results of the study.)

Instead, according to a long-time congressional staffer familiar with science policy, the Act had its origins in a political dispute over air pollution. When EPA proposed to tighten regulations on air pollution, opponents of the proposal felt hamstrung by an inability to access the supporting scientific data because the research involved human subjects and for other reasons. One result was a successful effort to amend the Freedom of Information Act to apply to scientific data, passed as a "rider" to a spending bill. The Data Quality Act was passed as another "rider" to a 2001 spending bill.

The term "rider" is inside-the-Beltway-speak for a piece of (typically) unrelated legislation added to other legislation -- often funding bills (called Appropriations). Many readers will be more familiar with riders that provide a direct infusion of federal dollars into congressional districts (i.e., "pork") for capital projects like bridges and even for science (i.e. "earmarks"). An essential feature of such "riders" is that they can largely escape the normal process of review and assessment that characterizes legislation developed through congressional policy committees. If the Data Quality Act has been systematically reviewed or assessed, one won't find a record of it in congressional deliberations.

The lack of assessment and the highly politicized process that led to the Data Quality Act will not be remembered as high points in the development of United States science policies. Even so, it is important to distinguish the process that led to the Act from the content of the Act itself. And the content does have potential to help stimulate some positive changes. Consider the following:

· A considerable amount of research produced by the weather research community has potential for application but fails to make it into the hands of end users. In the area of weather prediction enhanced consideration of quality and utility of scientific information could foster improved connections of "research" and "operations." A 2000 National Research Council Report labeled the gap between research and operations a "valley of death."

· To the extent that the Data Quality Act motivates serious considerations of utility it could also help to facilitate the transfer of science and technology from the public to the private sectors. Given the vast potential for increased interactions at the interface of federal research and commercial meteorology, any motivation for closer connections would be of value.

· The Data Quality Act, again with its focus on usability, could help make climate research more practical and of immediate benefit. If so, programs such as the Regional Integrated Science and Assessments of NOAA-OGP would benefit under the Data Quality Act. (Note: here at the CIRES Center for Science and Technology Policy Research we run one such RISA, the Western Water Assessment.)

At the same time that there are potential benefits, there are also valid concerns about potential negative consequences. In each of the above examples the benefits are associated with a greater consideration of the usability of science. However, none of the cases is particularly political (at least by comparison to other atmospheric sciences issues such as those explicitly focused on regulation).

In some issue areas there are valid concerns about the limits of science in decision making. Scientific results are frequently contested, and even if not contested, uncertain to some degree. As a consequence, advocates and decision makers support particular policies based on factors other than scientific findings. Recent examples include controversy about streamflow for salmon and farmers in the Klamath River Basin and the reappointment of Robert Watson to chair the Intergovernmental panel on Climate Change.

Proponents of the Act suggest that it will improve the information base on which policies are made. James Tozzi, of the Center for Regulatory Effectiveness, stated in the New York Times, "Now in the world's most powerful government you're going to have to issue information that's accurate." But opponents worry that the Act would simply bias policy in favor of business-as-usual in the face of uncertainty. Joanne Padrón-Carney, director of the Center for Science, Technology, and Congress of the American Association for the Advancement of Science, stated in the same New York Times article, "We really would not like to have science attacked as a way of being sure that policy isn't made." It is difficult to resolve these perspectives.

To be sure, policy can benefit from improved connections of science and decision making, and to the extent that the Data Quality Act helps to address these connections it is a valuable addition to the nation's science policies. But at the same time, there is great potential for the Data Quality Act to further the politicization of science and actually impair the connections of science and policy. We won't know whether the Data Quality Act benefits or impairs national science policies until it is implemented and analysts begin evaluating its effects. Just like putting a new drug on the shelves without any testing and seeing after-the-fact whether or not it improves human health.

Roger Pielke, Jr.
Center for Science and Technology Policy Research
University of Colorado
pielke@cires.colorado.edu

For further reading:

• Herrick, C. N. and D. Jamieson, 2000. Junk Science and Environmental Policy: Obscuring Public Debate with Misleading Discourse, Philosophy and Public Policy Quarterly, Vol. 21, Spring: 11-16.
  
  
• Pielke, R. A., Jr., 2002. Policy, politics and perspective. Nature 416, 367-68.
  
  
• Sarewitz, D. 1999 (March draft). Science and Environmental Policy: An Excess of Objectivity, a revised version of this manuscript appears in: Earth Matters: The Earth Sciences, Philosophy, and the Claims of Community (Prentice Hall, 2000), edited by Robert Frodemen pp. 79-98.