The approach used to compile the data catalog consisted of four steps: initial listings of data sources, screening of data sources, reviewing of the data sources, and developing the catalog profiles. Each of these steps is described in more detail in the paragraphs that follow.
Developing the initial listings of data sources. The project team (in consultation with ASPE, our project consultants, and our small AI/AN/NA workgroup) initially developed a substantial list of potential data sources drawing from the following sources:
- Web sites of all federal departments (e.g., Departments of Justice, Defense, Agriculture) with information on publicly available data sets and on administrative data sets that could be made available from each of these departments;
- Data repositories including National Library of Medicine, the Interuniversity Consortium for Political and Social Research (ICPSR) at the University of Michigan, and the University of Virginia to identify additional data sources located in these repositories that may be appropriate;
- Web site developed by the DHHS Data Council that provides a Directory of Health and Human Services Data Resources listing available DHHS data resources (http://aspe.hhs.gov/datacncl/datadir);
- Web site developed by DHHS' Office of Minority Health (OMH) and the Office of the Assistance Secretary for Planning and Evaluation (ASPE) under the auspices of the DHHS Data Council entitled Health and Human Services Statistics About Minorities (http://www.hhs-stat.net/omh/index.htm);
- Consultation with members of the DHHS Data Council's Racial and Ethnic Data Working Group who served as a workgroup for this project; and
- Consultation with project consultants and a small AI/AN/NA workgroup representing the American Indian and Native Hawaiian communities.
As the project progressed, data sources were continually added to the review list as they were discovered or recommended to the project team. In all, a total of 152 possible data sources were considered.
Screening of data sources. Data sources were initially screened by reviewers to determine their appropriateness for inclusion in the catalog. The screening protocol comprised the first few pages of the review protocol located in Appendix C to this document. This screening protocol ensured that the data sources to be fully profiled and considered for inclusion in the catalog met the technical parameters listed in Section 2 above. Specifically, the reviewers provided the following information on each data source screened:
- Name of data source being reviewed;
- Sponsoring agency;
- Policy issue relevant to data source;
- Brief description of data set;
- Whether the data source identified people who are AI/AN/NA;
- Availability of data set to researchers;
- Cost associated with use of the data set;
- Relevant contact and location information regarding this data source;
- Source of funding for data set;
- Total unweighted AI/AN/NA population in data source and if unweighted n < 100, feasibility of combining across multiple iterations of data collection to increase the size of the population;
- Geographic coverage of the data set;
- Timeframe of the data set;
- Nature of data source (i.e., survey, program reporting data, or registry) and availability of relevant documentation; and
- Qualitative or quantitative nature of data.
Based on this information, the project team and ASPE project officers determined whether the data source should be fully profiled and likely included in the catalog, included in the list of non-reviewed data sources (Section 6), or omitted entirely.
Reviewing the data sources. Data sources appropriate for profiling were then researched and the data were collected about as many of the 25 elements as possible by the initial reviewers. (See Section 2.1 above.)
For some data sources, not all of this information was available through the Internet, published materials, or direct contacts with employees at the agency or data collection facility. After a reasonable effort, reviewers were instructed to note that missing information could not be obtained or was unavailable. Where the information on any of these categories is not applicable to the data set, the category will not appear in the profile (e.g., weighting information for registries) but when information important to the assessment of the utility of the data source was not available, the category remains in the profile and is duly noted.
Initial reviews were evaluated by senior project staff and revisions were made in light of their comments and questions. These revised reviews were then converted to the catalog entries presented here. In summary, a total of 152 possible data sources were considered. Due to resource issues and some early elimination of unusable data sources from the review process, 110 of the 152 data sources were screened. From the 110 screened data sources, 68(6) data sources met the criteria for inclusion and are fully profiled in the catalog.
After all data reviews were completed, the data source profiles, where possible, were forwarded to the agency or point of contact for that data source for review. Westat received comments or approval on 76 percent of the data profiles in this catalog. After agencies and/or points of contact completed their review of the data source profiles, we revised many of the profiles in response to their comments. Westat then submitted the revised data catalog to members of the project DHHS workgroup, the project AI/AN/NA workgroup, the five Westat project consultants, and senior DHHS staff for review. Many of their comments and suggestions were incorporated into the final catalog. A list of the data sets by the different sponsoring agencies is included in Appendix D.
"report.pdf" (pdf, 1.56Mb)
"apa.pdf" (pdf, 107.76Kb)
"apb.pdf" (pdf, 88.97Kb)
"apc.pdf" (pdf, 233.07Kb)