Practice 1: A Clearly Defined and Well-Accepted Mission
A CLEAR UNDERSTANDING of the mission of a statistical agency, the scope of its programs, and its authority and responsibilities is basic to planning and evaluating its programs and maintaining credibility and independence from political control (National Research Council, 1986, 1997b). Some agency missions are clearly spelled out in legislation; other agencies have only general legislative authority. On occasion, specific requirements are set by legislation or regulation (see Appendix B).
An agency’s mission should include at least three major elements: The first is responsibility for all elements of its statistical programs: sources of data, measurement methods, efficient methods of data collection and processing, and appropriate methods of analysis. The second is responsibility for evaluating, implementing, and documenting new methods and processes that better serve users’ needs for relevant, accurate, and timely statistics within budget constraints. The third is responsibility for curation to ensure the availability for future use of the agency’s data together with documentation of the methods used and the quality of the estimates.
A statistical agency’s mission should also include primary responsibility for determining the subjects and questions for which it will collect data, acknowledging that Congress and the executive branch may properly exercise authority in this area (see Practice 2). In all instances, a statistical agency’s mission must focus on information that is to be used for statistical purposes. When an agency is charged to carry out an activity that could undermine its credibility as an objective source of information (e.g., collecting data for both statistical and administrative purposes), the agency should carefully describe and structure the activity (e.g., perhaps locating
it within a clearly demarcated office). If the agency concludes that it is not possible to develop a satisfactory arrangement that is both responsive to the charge and credible, it should request that the activity be assigned elsewhere.
A statistical agency should clearly communicate its mission to others. The Internet is today the primary means for an agency, not only to disseminate its data and associated documentation, but also to publicize its mission to a broad audience and to provide related information, including enabling legislation, the scope of the agency’s statistical programs, confidentiality provisions, and data quality guidelines. Consequently, careful design of an agency’s website to maximize the ability to easily navigate the site is imperative.
A statistical agency should periodically review its mission. As part of strategic planning to carry out its mission within its budget, it should review priorities among different programs (see, e.g., National Research Council, 1976, 2000b, 2009a), the infrastructure (e.g., computing capabilities) needed to support them, and the relative importance and urgency of needed improvements, say, in timeliness versus accuracy.
Such a process will take time and may open an agency to charges that its priorities are misguided (e.g., when an agency decides to cut back on an existing program in favor of a new initiative). Nonetheless, the process should put the agency in a stronger position to justify its portfolio to stakeholders and to more efficiently allocate scarce staff time and other resources among programs and initiatives. Regular reviews may also identify authority that would be useful for an agency to have, such as authority to acquire administrative records or to be a clearinghouse for information collected in a topic area from its own programs and those of other statistical agencies.
Advice from outside groups should be sought, formally and informally, on the agency’s priorities for data collection and data products and services and on its statistical and operational methods. Such advice should be obtained from both data users and providers and from professional and technical experts in the subject-matter area, in statistical methods, and in information technology (see Practice 6). A strong research program in the agency’s subject-matter field can assist in setting priorities and identifying ways to improve an agency’s statistical programs because subject-matter researchers at an agency are well positioned to communicate data user needs to agency statistical and operational staff and vice versa (Triplett, 1991; see also Practice 10).