|
Minimal Information to Accompany Any Report of Survey or Census Data Census Bureau
Standard Authored by: Introduction On March 26, 2001, the Census Bureau issued principles for informing users of data quality and methodology. The first principle states, in part, "Statistical products will be accompanied by descriptions of the methods and procedures used [. . .]." The purpose of this standard is to define the minimal information that should accompany any public release of data by the Census Bureau. The standard applies to published reports, releases of electronic microdata, and news releases. It does not apply to product announcements. The standard is consistent with other standards published by the Office of Management and Budget1 and by professional associations such as the American Association for Public Opinion Research (AAPOR) and the Council of American Survey Research Organizations (CASRO). It very closely follows the "Standards for Minimal Disclosure" published by the American Association for Public Opinion Research in its Code of Professional Ethics and Practices (1991). Compliance with minimal reporting standards for information about surveys is uneven. Research sponsored by the Federal Committee on Statistical Methodology found that government statistical agencies frequently failed to report key information about the surveys they conducted. Currently, news releases and reports issued by the Census Bureau to the public provide little information about the survey that is the subject of the report. The purpose of this standard is to improve the quality and consistency of reporting about surveys by the Census Bureau and its sponsors. It clarifies the minimal information about a survey that should accompany any report or release of data. Scope The standard applies to all public releases of survey and census data by the Census Bureau. The Census Bureau also encourages its sponsors to adhere to this standard when they issue data collected by the Census Bureau, and encourage its other data users to adhere to this standard when they issue reports. Standard Essential information describing the methods and procedures of a survey must be provided (or a source for it made available) as part of any public release of survey or census data by the Census Bureau. At a minimum, the following items should be reported or made available: 1. The organizational sponsor(s) of a survey; 2. The organization(s) that conducted it; 3. The wording of questions asked and description of derived measures that are the subject of the report; 4. A definition of the population under study, and a description of the sampling frame used to identify this population; 5. A description of the sample design; 6. The size of sample, and disposition of sample cases (e.g., numbers of interviewed cases, ineligible cases, and nonresponding cases); 7. If applicable, information on eligibility criteria and screening procedures; 8. A discussion of the statistical precision of the results, at least for the major estimates. This could include estimates of sampling variances, standard errors, or coefficients of variation, or presentation of confidence intervals; 9. Description of estimation procedures, including weighting, editing, and imputation methods;1 10. If applicable, clear indication of which results are based on parts of the sample, rather than on the total sample; 11. Method and dates of data collection; 12. Discussion of nonsampling errors that may (or are known to) affect the data;2 and 13. Discussion of methods employed to ensure data quality. It is anticipated that information would be provided or a source specified for each of the 13 elements. The amount of information provided for each in the actual report or release would be adjusted to the length and purpose of a report or news release. As appropriate and necessary, links to a website or citations of technical documentation can substitute for detailed explanations or information (e.g., question wordings). Responsibilities Program areas will be responsible for the following: - preparing the appropriate information/materials based on these standards; - ensuring the proper use of the standards; - seeking the cooperation of survey sponsors (and users) in complying with the standard. It is desirable for Memoranda of Understanding (MOUs) with survey sponsors to include an agreement that sponsors will provide the required minimal information in their releases and reports of Census Bureau-collected data. The Census Bureau will work with survey sponsors to develop appropriate descriptions; and - ensuring that survey organizations that conduct surveys under the Census Bureau's sponsorship include minimal required information about the surveys in their releases and reports. The requirement to follow the standard should be written into the contract with the data collector. The Methodology and Standards Council will be responsible for the following: - the initiation of periodic evaluations, reviews, and updates to the standards, as necessary; and - guidance to program areas in the development and usage of the standards, and assistance in preparation of information to meet the standards. Implementation The standard can be implemented immediately for a six-month trial period. Questions or difficulties about its implementation may be directed to the Census Bureau Methodology and Standards Council. Inquiries Inquiries relating to the interpretation of this standard should be addressed to the Census Bureau Methodology and Standards Council. Supporting Document Examples of Revisions of Survey Descriptions to Meet Standard Recommended by Census Bureau Methodology and Standards Council:
Signed Robert N. Tinari, March 18, 2003 Signed Rajendra P. Singh, March 18,
2003 Signed Alan R. Tupek, March 18, 2003 Signed Howard Hogan, March 18, 2003 Signed Ruth Ann Killion, March 18, 2003 Signed Tommy Wright, March 18, 2003 Concurrence: Signed Preston Jay Waite, March 18,
2003 Signed Nancy M. Gordon, March 18,
2003 Signed Frederick T. Knickerbocker, March 18,
2003 Signed Marvin D. Raines, March 18,
2003 Signed Cynthia Z.F. Clark ,March 18,
2003 Signed Lawrence A. Neal ,March 18,
2003 1 The OMB's "Standards for the Publication of Statistics" (1978; pg 19265), which apply to all federal statistical surveys, state: "To help guard against misunderstanding and misuse of the data, full information should be available to users about sources, definitions, and methods used in collecting and compiling statistics, and their limitations [...]. A description of the survey design and methods [...] should be available [...]. The description should include what is measured, the source(s) of information, the sampling plan if sampling is used, the method of collecting the data, the extent of nonresponse and other sources of bias, and the methods used to deal with the problem [...]. Accuracy of the data should be stated as far as possible. This should include not only the sampling error (where probability sampling is used) but also the nature and extent of nonsampling errors." 2 See also Tech Paper 32, which is currently being revised and will be issued as a Census Bureau Standard for Discussion and Presentation of Errors in Data. Document Management & Control
The most current version of
this document is maintained on the Census Bureau Intranet and may Category:
Standard |