Strategy for Implementing Data Quality Objectives
Planned Project Work Activities
Data Access and Data Quality are two sides of the same coin. Without access to studies' underlying data, it is difficult to demonstrate suspected inaccuracies and methodological weaknesses. It is therefore important that there be a mechanism through which data quality problems can be brought to the attention of federal agency officials and that there be a mechanism to petition for the correction of such information dissemination by the government, or of regulations based upon such data. By advancing both elements - Data Access and Data Quality - real gains can be made in achieving transparent and rational regulations.
The FY 1999 Omnibus Appropriations Act (P.L. 105-277) and conference report addressed both Data Access and Data Quality concerns. A major advance has been made on the Data Access portion of the law, with OMB issuing a final rule to revise Circular A-110 and to make the underlying data from federally funded research available to the public through the Freedom of Information Act (FOIA). Attention now turns to the agencies' drafting of conforming regulations to implement the OMB rule.
Unfortunately, the Data Quality aspect of the law has seen little progress. The crux of the Congress's most recent Data Quality directive is to allow the public to petition for the correction of information used and disseminated by the federal government. This corrective mechanism would apply to information that is deficient in "quality," "objectivity," "utility," or "integrity." Such Data Quality efforts would promote sound agency policies and increase public confidence in the regulatory system. Therefore, OMB and the federal agencies must be encouraged to implement the Data Quality provisions accompanying P.L. 105-277.
Purposes of the Paper
This paper will:
Data Quality Work Activities
- Present a work plan for the Data Quality initiative and discuss its various elements.
- Discuss how each element ties in with overall Data Quality objectives.
- Identify sources of potential opposition to implementation of the Data Quality provisions.
I. Preparation of Background Materials on Data Quality
"Quality" is an extremely subjective term in the context of data and otherwise. Therefore, an effort to define "quality" and the other key terms in the legislation (i.e., objectivity, utility, and integrity) would aid in OMB and agency implementation. Also, since such a petition process would be somewhat novel (at least for OMB), a model petition process, based upon a review of existing agency petition processes, would be useful in implementing the Data Quality provisions. Specific related activities would include:
Analytical Paper Defining the Four Key Data Quality Terms
- Efforts will be made to define the key terms - "Quality," "Objectivity," "Utility" and "Integrity" - through analysis of relevant materials from a variety of sources including:
- Federal regulatory agencies
- State regulatory agencies
- Corporations and trade associations
- Non-profit scientific associations (e.g., AAAS, NAS)
- International organizations (e.g., WHO, OECD, IARC)
- A final report will define the key Data Quality terms based upon our synthesis of the competing models.
- The Center for Regulatory Effectiveness (CRE) will widely distribute the Data Quality Terms Paper to key decision makers and interested stakeholders at OMB, the federal agencies, Congress, businesses, and scientific organizations. CRE will also place the paper on the CRE Website (www.TheCRE.com) to solicit additional public comment.
Review of Existing Petition Processes at Key Federal Agencies
Work to Date
- CRE has begun surveying existing petition processes at three key rulemaking agencies: EPA, OSHA and HUD. Based upon this review, CRE will seek to demonstrate the following:
-- These agencies already administer virtually hundreds of existing petition processes.
-- These petition processes are a crucial for ensuring due process and regulatory fairness for the private sector.
-- The fact that so many of these processes have already been implemented without placing undue resource burden on the agencies demonstrates that OMB's present argument, i.e., that establishing a petition process to enable private parties to challenge violations of Data Access and Data Quality concerns would unduly burden the federal government, is without merit.
- To date, CRE has identified 399 existing references to petition processes in EPA's regulations. At least one of those provisions allows members of the public to obtain disclosure of data.
-- CRE expects that review of the 399 sections will yield approximately 60-80 individual petition processes or procedures. (This lower number reflects the fact that certain regulatory provisions cross-reference the same petition processes.)
-- These numbers make a strong argument for the proposition that adding one more petition process would hardly "shut down" the government.
- Similarly, CRE has preliminarily identified 57 references to petition processes among HUD regulations, as well as 39 at OSHA.
- CRE is completing its analysis of the petition processes identified above and will present its results in a chart or matrix format to demonstrate that petition processes are a resource-efficient means of providing the regulated community with the due process required by the American legal system.
- CRE will present its "underlying data" in terms of the following analytical categories:
- Petitions for disclosure of information. These are few in number. The most notable example is an EPA provision allowing petitions for disclosure to health professionals of chemical identities (otherwise nondisclosable as trade secrets under EPCRA). The fact that such provisions already exist provides strong evidence that this is a workable manner for implementing data access and quality requirements.
- Petitions to initiate rulemaking proceedings or to amend rules.
- Petitions for agency action that would constitute final agency action. Many of the existing processes result in "final agency actions" under the APA, because an agency's denial is final and thus subject to judicial review. It is imperative that any petition process established for data access and quality result in judicially reviewable "final agency actions."
- Petitions for variances or exceptions. For example, OSHA's procedure for petitioning for an exemption from the formatting specifications in OSHA's recordkeeping requirements. This could be a component of a data access and quality petition process.
- Petitions for administrative hearings, including reconsideration and appeal. This category enables private parties to appeal specific determinations affecting the parties.
- CRE will also classify the petition processes and procedures in terms of whether they are open to any concerned member of the public. A number of them are, especially where public harm is implicated. Once again, this provides strong evidence that a data petition procedure open to the entire public would not result in costs that the federal government could not handle.
- CRE's analysis will also provide precedents to enable CRE to recommend either of the following:
- A petition process whereby the agency's denial would be subject to administrative review. This would avoid the expense of judicial review in most instances.
- A process whereby denial would be immediately subject to judicial review. This would lower agency costs, but could increase costs to petitioners and the judicial system.
II. A Case Study of Federal Data Quality: Review of the EPA Global Warming Website
In order to build support for the Data Quality initiative among business leaders, congressional leaders, and other interested stakeholders, CRE identified and analyzed a concrete example of federal information dissemination lacking in quality. The case study chosen for this analysis was EPA's Global Warming Website.
Work to Date
III. Drafting of Regulations for Data Quality
- Researched, drafted, and widely disseminated paper discussing shortcomings of EPA's Global Warming Website.
- Consulted with the Environmental Council of the States, which is preparing an extensive analysis of the considerable Data Quality problems at EPA.
- Transmitted letters to key officials at EPA offering CRE support to new EPA Information Office. CRE urged the Information Office to begin drafting Data Quality regulations and to take a leadership role among the agencies.
IV. Discussions with Federal Agencies on Data Quality
- As noted earlier, the task of drafting Data Quality regulations may be difficult. It would therefore be useful assist OMB in this challenging task by drafting a Data Quality rule that OMB could adopt, with or without modification. Such an approach would offer several benefits:
- It would put our ideal Data Quality rule on the table for discussion with OMB and agency policy makers.
- The draft rule would give OMB a starting point for analysis and a conceptual framework within which to work.
- It would demonstrate to OMB that petition processes currently exist among the federal agencies and are administered at an acceptable cost. Thus, the draft Data Quality rule could be implemented in a similar manner.
- Presentation of a draft rule would hopefully accelerate the entire Data Quality initiative.
- At a later date, assistance would also be offered to individual agencies - particularly EPA, OSHA, and HUD - in drafting their own conforming regulations. Such regulations would tailor the broad OMB Data Quality rule to meet the specific needs of individual agency programs.
- Early and ongoing dialogue with federal agency officials may prove useful in facilitating the Data Quality initiative and the promulgation of implementing regulations. The following agencies would receive primary attention:
- While initial emphasis would be placed on the above agencies, subsequent efforts would involve a more global set of agency contacts.
Discussion topics would include the following:
- Data Quality generally and the requirements of the new law.
- Substance of the Data Quality terms paper.
- Substance of the draft Data Quality regulation.
- Implementation issues, including the Data Quality petition process.
- Conforming agency regulations.
- Assistance and information that could be provided to the agencies.
- Ideally, such discussions would build support for the Data Quality initiative among federal agency officials. At a minimum, such efforts would place the issue squarely on the table and stave off active opposition to the Data Quality efforts.
V. Solicitation of Comments on Data Quality
Public Comments through the CRE Website
The CRE has played a leading role in the Data Access and Data Quality debate, and these data topics are discussed extensively on the CRE Website. This role would continue as Data Quality efforts gear up. Through its Interactive Public Docket, CRE will continue to seek public comment -- including comments from agency officials -- on all relevant materials. The Data Quality initiative would evolve, in part, based upon the level and direction of public input.
Work to Date
- The CRE Website currently contains the following materials relating to Data Quality:
-- Background Information
-- CRE Comments on Data Quality
-- Correspondence on Data Quality
- New Data Quality materials would also be posted on the CRE Website, including:
- Data Quality Terms Paper
- Draft Data Quality Regulation
- Additional Relevant Data Quality Materials
Comments of Congressional Staff
Because congressional support for Data Access and Data Quality has been so important to bringing these issues to the forefront, careful attention will be paid to keeping the relevant congressional staff informed as to Data Quality developments. Particular attention will be paid to those Senators and Congressmen who took an active part in creating or defending the Data Access legislation.
Such congressional input will be sought on all major papers and other efforts related to Data Quality.
VI. Presentations to OMB and Congress
Once sufficient materials are in hand and a variety of viewpoints have been sought, formal briefings on Data Quality will be made to OMB officials and Congress. Such efforts will be designed to increase pressure on OMB to more forward with the promulgation of a regulation on Data Quality.
Unlike the Data Access rule whose burdens will fall primarily on federal grantees, the effort associated with the Data Quality rule will rest with the federal agencies themselves. Because they will likely consider such new requirements to be an unwelcome expenditure of time and resources, the agencies are likely to oppose implementation of Data Quality provisions. For this reason, it is imperative that the reasonableness and workability of the Data Quality provisions be clearly demonstrated, particularly to OMB.
The above activities are designed to facilitate the Data Quality initiative to the maximum extent possible. This will be done by demonstrating that outside mechanisms exist to gauge and enforce Data Quality and showing that similar petition processes are already functioning at various federal agencies. By drafting a proposed Data Quality regulation, the entire process is significantly advanced, and agency thinking can start with the revision (rather than creation) stage.
By working with OMB, the agencies, and congressional staff as outlined above, it may be possible to build support for early implementation of the Data Quality provisions referenced in P.L. 105-277.