Other 2000 Census Data Issues
In addition to issues directly related to the panel's charge for this first interim report, the panel's work raised some related issues concerning the 2000 census. This section presents the panel's findings and analysis on those issues.
SETTING PRIORITIES FOR EVALUATION STUDIES
Many of the group evaluation plans10 described to the panel ended with data collection, providing little detail on how the data would be used.
Recommendation: The Census Bureau should develop a detailed plan for each evaluation study on how to analyze the data collected and how to use the results in decision making. The Census Bureau should then use the plan to identify the benefits and resources required for each evaluation study, set priorities among them, and allocate sufficient resources for the careful completion of all or, at least, the highest priority evaluations.
These plans should include detailed information on how the data will be analyzed, how the results obtained will inform decisions about the 2010
10 |
The panel was not provided with a list of the individual planned census 2000 evaluation studies. |
census design, and what resources, in terms of data collection costs and staff expertise, are required.
Understanding the costs and technical resources required and the potential information value of each proposed evaluation study will help in setting priorities and allocating resources to the studies. Setting priorities is critical because of the large number of evaluation studies being planned by the Census Bureau; there may be insufficient staff and expertise to adequately use the data being collected. In setting priorities, the objectives should be both to assess 2000 census operations and to provide information to support 2010 census design decisions. The results of this process will allow the Census Bureau to make informed decisions about which are the most important and the least resource-intensive evaluation activities to carry out following the 2000 census.
MEASURING THE ACTIVITIES OF LOCAL COMMUNITY GROUPS
For the 2000 census, the Census Bureau has greatly promoted participation by community organizations, with the goals of increasing the mail return rate and overall responsiveness to the 2000 census. If the goals are met, the result will be lower costs and, possibly, improved data accuracy, by reducing the frequency of proxy response and differential undercoverage.
Recommendation: The Census Bureau should examine various methods for collecting information on the activities of local organizations that are intended to produce a more complete census count.
Community organizations will vary in their effort and methods. Some groups might make greater use of “Be Counted” forms, speeches at public gatherings, or door-to-door canvassing. Such variation may complicate evaluation of the census because evaluation will depend on knowing what activities took place in each area. Therefore, to the extent possible, the Census Bureau should examine ways of collecting information about the activities of these groups. For example, they could be asked to supply a diary of their activities after the 2000 census is concluded, or the workers at each local census office could be asked to monitor the activities of local groups, or the staffs of local groups could be interviewed.
The Census Bureau is also planning on providing mailout-mailback areas with a day-to-day measurement of their return rate of the mailed
questionnaires. It would be useful to examine the effectiveness of this system for encouraging local activities to increase mail response. Therefore, this information on daily response rates should be retained, along with information about the areas, including the final census and adjusted counts. Finally, any adjustment of the census will suffer from not being able to incorporate information about local efforts. 11
11 |
Without these data, some areas with strong local efforts, and thereby possibly reduced undercoverage, will likely have estimated undercounts that are biased upward. Even if this information were collected, its incorporation in an adjustment model would be difficult to carry out. |