RESPONSE AND COVERAGE
Much of the Census Bureau's research on response and coverage for the year 2000 census addresses two main criticisms raised about the 1990 census: the high unit cost of the census and the increase in the differential undercount by race. One of the most important findings of the 1990 census—the increasing diversity of the United States—also identifies one reason for its difficulties. While many people continue to be persuaded by appeals based on citizenship and a common national duty, and many have the skills and motivation to respond to the census mail questionnaire, a sizable and perhaps increasing proportion appear to be motivated primarily by local interests and appeals, demand control over portions of the census process or outcomes, or require specialized help or media in order to participate. If the trends uncovered by the 1990 census continue, as expected, the 2000 census will face even greater obstacles. A recognition that the United States is simply and fundamentally becoming ever more difficult to count must be incorporated into planning for 2000 in order to develop viable strategies and the organizational and political consensus to implement them.
Controlling costs and reducing the differential undercount in 2000 may require a massive reorientation—both conceptually and organizationally. The 1990 census attempted to universally apply a standard, and standardized, approach, but the outcome was a differential undercount. Furthermore, standardization necessarily broke down in many areas, particularly those in which response to the mail questionnaire was low, as enumerators with only brief training were sent out and ultimately empowered to make last-ditch enumerations, using procedures known as last resort and closeout that permitted contacting persons who were not residents of the household. The 1990 experience with coverage improvement programs—such as the housing coverage check and the second phase of the parolee-probationer check—suggests that the real alternative to planning different approaches for different populations may be using different approaches in a haphazard manner. Instead, we believe that a system could be designed that is flexible enough to help reduce the differential undercount yet maintain important aspects of standardization (such as definitions of household membership).
Some components of the research on the topics discussed in this chapter seem designed mainly to control costs by increasing response to the initial mail questionnaire and are fairly traditional. But other portions of the research program, particularly those aimed at "targeting barriers to enumeration," have the potential to help produce more innovative approaches and possibly reduce the differential
The most developed component of the Census Bureau's research on response issues addresses the problem of reducing the unit cost of each response by increasing the number of people who return the mail questionnaire. Although increasing the initial response rate may leave the differential undercount unchanged—or even exacerbate it somewhat—a higher initial response rate means better quality data, fewer nonrespondents to follow up, and better control over cost. For these reasons, the experiments demonstrating that better design of the questionnaires and sending advance letters, reminder cards, and replacement questionnaires can improve response rates in a census have had an important role to play. And these experiments have also given the Census Bureau needed experience with new operational procedures based on implementation techniques that survey researchers have developed over the last 20 years.
The research program focused on the differential undercount draws on the work of the Census Bureau's Barriers to Enumeration Working Group and has become better developed since the late fall of 1992. What we refer to here as a "program" is really a diverse set of relevant topics and projects that have been collected as the Special Methods Research and Development Program; these projects have been compared to each other and ranked by census data users during a series of Census Bureau consultations with the Commerce Department task force and other census advisory committees. The range of the projects under consideration is important, because some projects address the components of the undercount due to missing households and other projects address components due to missing people within households. In many cases, however, what has been formulated is more a topic than a research project; what is needed is development with the goal of evaluation—experimental evaluation where possible—and cost estimation.
Our review of the research projects addressing response and related coverage issues suggests that the response rate improvement research program has yielded strategies that promise to help control costs in 2000, but that research should now focus on techniques that have potential to reduce the differential undercount. We do not recommend abandoning research on improving initial mail response altogether, given its critical role in controlling costs and in the overall success of the census. But efforts to improve initial mail response should now focus on planning for key instrument and implementation experiments in the 1995 census test. The plans should draw on both research already conducted at the Census Bureau and other findings reported in the literature—for example, on the use of telephone for follow-up—to test the "best" feasible implementation methods. It is not necessary to have a large-scale experiment for every innovation in design or mail-out procedure before the 1995 test.
The remainder of this chapter discusses and makes specific recommendations on a range of topics related to response and coverage issues. We review the extensive response rate improvement research program conducted by the Census Bureau. We also focus attention on proposed research, and four major themes emerge from our discussion.
Focus on reducing the differential undercount. We recommend planning and limited experimentation now so that the most promising methods for reducing the differential undercount can, where feasible, be given operational trials and experimental and cost evaluation in the 1995 census test. Narrowing the research program to methods that may reduce the differential undercount is important given the limitations of resources and time, and it can be justified by the more advanced state of research on improving mail response rates. If operational variables associated with response, such as time of receipt and medium used, have been tracked, then much can be learned from analyses of data from the 1980 and 1990 censuses. We urge the Census Bureau to devote resources to activities of this type whenever potentially valuable information can be obtained through alternatives to expensive, large-scale experimentation.
Go local. The targeting model and tool kit proposed in the Census Bureau's research program provide for localized, decentralized outreach and enumeration activities. But more may be needed. Because undercounted groups are clustered and because their reasons for not participating may vary greatly by locality, reducing the differential undercount will probably require a major reorientation in the Census Bureau's practice. Outreach and enumeration activities may need to be more decentralized than current field offices, which, even when they know where to find undercounted groups, may not have the credibility needed to motivate them to participate. Learning how to localize outreach, promotion, and enumeration should be central to planning for the 1995 census test, as should the development of designs to evaluate the success and to estimate costs of such efforts. Good procedures for eliminating duplicate records (see Chapter 1) are also needed if the Census Bureau is to be able to manage more flexible enumeration procedures that respond to local needs (e.g., filling out a form in the shopping mall) without losing accuracy.
Develop a plan for national and locally based outreach and promotion. Messages are needed that will motivate members of groups that are hard to enumerate. Promotional efforts should include gaining access to local media markets. Such strategies may both enhance the response to the initial mailout among undercounted populations and support more targeted enumeration efforts.
Develop and test the application of residence rules for both mail and telephone implementation. Coverage errors within households are response errors. Clearer definitions and less ambiguous questions are needed to help respondents provide more accurate answers. Because the 2000 census is likely to include more use of telephone—either as a primary response mode or to follow up nonrespondents—it is also important that a telephone version of these new questions be developed and mode differences examined.
RESPONSE IMPROVEMENT RESEARCH
Taken as a whole, the Census Bureau's response improvement program attempts a coordinated approach to a wide range of issues, and the program clearly
has the potential to mitigate the decline in mail return rates or even to increase the rates. The use of a "respondent friendly" questionnaire, an advance notification letter, a reminder postcard, and a replacement questionnaire are all methods that have been widely adopted in surveys, and their extension to the census is overdue and promising.
Response Rate Improvement
The Census Bureau conducted the Simplified Questionnaire Test (SQT) in April 1992. Its primary purpose was to compare the final mail completion rates for four alternative short-form questionnaire designs (booklet, micro, micro with Social Security number, and roster) to the 1990 census short-form questionnaire in a survey environment (Bryant, 1992). The test yielded an increase of 3.4 percentage points in response rate using the respondent-friendly version of the 1990 short form.
The Implementation Test (IT) was designed to test the benefits of three components in the mail implementation plan for the short-form sample—a prenotice letter, a stamped return envelope, and a reminder postcard (Bureau of the Census, 1992d). The IT was carried out by the Census Bureau in October 1992 and found that use of a prenotice letter and reminder card together produced a gain of 12.7 percentage points in the response rate. On the basis of the SQT and IT results, Census Bureau statisticians estimate that the use of a replacement questionnaire will increase response by approximately 10 percentage points. Although it is less clear whether any of the techniques investigated in this program will reduce the differential undercount, any increase in routine mail responses improves data quality and helps to control costs.
Because these treatments were not all tested in combination with one another, we cannot confidently assess the degree to which they are additive. Also, the magnitude of these treatment effects in the environment of an actual census year is unknown. Nevertheless, the strength of the experimental findings suggests that these features are likely to have a significant positive effect on response rates in the next decennial census.
Mail and Telephone Mode Test (MTMT). The MTMT was designed to examine whether offering respondents the option of calling in a response or receiving a replacement questionnaire increases the response rate. The treatments vary in when the offer is made, and one treatment (the "preference" treatment) allows respondents to call when they first receive the form. The experiment was conducted in April 1992, and initial results suggest that offering people the opportunity to call in for telephone enumeration does not increase the total response rate. If the complete MTMT results bear out this preliminary finding, offering people more "high-tech" response modes—such as the fax—that require them to take the initiative in responding are similarly unlikely to increase the total response rate.
We have two concerns about the MTMT. One is that it may discourage
research into ways that the telephone is more likely to be used successfully. As it has in sample surveys, the telephone will probably prove itself a useful follow-up tool in the 2000 census—for reminding people to return their questionnaires and for completing cases without sending an enumerator to the field.
A second concern about experiments using the telephone is that because residence rules are complex, answers to the self-administered mail questionnaire (which displays the residence rules) may differ from answers to the version of the questionnaire used in computer-assisted telephone interviewing (CATI) (in which the rules may or may not be read to the respondent). Furthermore, the mode effect may vary depending on whether a respondent interviewed by CATI is looking at the questionnaire or not. During the MTMT, people were encouraged to have the census form in front of them when they call to be interviewed, but this result, obviously, cannot be guaranteed. In the MTMT a self-selected group was enumerated by telephone; therefore, the MTMT cannot be used to estimate the possible mode effects. And estimating the effects of mode on within-household coverage is essential if people are to be allowed to respond to the census using different modes. In addition, procedures to eliminate duplicate cases are critical to maintaining accuracy when households are allowed to respond by modes other than mail.
Appeals and Long Form Experiment (ALFE). This experiment was conducted in July 1993. The first component of this experiment tests the effects of two different motivational appeals, one emphasizing the mandatory nature of the census, the other stressing the benefits of census participation. The second component tests several respondent-friendly versions of the 1990 long form. The key substantive issue explored by this experiment is how to better motivate participation. This is a question that also affects other components of the census (see discussion below of outreach and promotion). Although we agree that research on how to motivate participation is important, we believe that it should be integrated with research on how to motivate participation by members of undercounted groups. In addition, the 1995 census test should build on the results of ALFE and further test the use of motivational appeals.
At a more specific level, the description of ALFE notes that using a statement that response is mandatory may irritate or threaten some respondents. If this research is pursued further (for example, in the 1995 census test), we suggest a treatment in which the statement that participation is mandatory is added during follow-up (e.g., with the reminder card) and not included in the initial mailing.
Multiple Response Modes. The original goal of "expanding response modes" now appears to be limited to telephone, which is the most "high-tech" mode listed on page 1 of the "Summary of Research Projects" (Bureau of the Census, 1993f). Some of the results from the MTMT, particularly the use of the telephone in the "preference" treatment, in which the initial mailing invites response by telephone, can be plausibly generalized to more specialized modes, such as fax. Although we see important possibilities for using the telephone, particularly in following up those who do not respond by mail (see below), accepting telephone responses requires research
on methods to eliminate duplication of records. Research to eliminate duplicate records is probably also essential for some components of the tool kit (described in the next section). In addition, as noted above, accepting telephone responses when the respondent does not have the census form available requires development of a telephone instrument.
A complicating feature of these experiments is that crucial features of the 2000 census—such as what information is needed to eliminate duplicate records, the method for capturing data, and the final operational form of residence rules—must necessarily remain in doubt, even while research about improving the mail response rate proceeds. We are concerned, therefore, that some of these features may interact with the experimental manipulations. Such interactions could compromise projections of response rates for the 2000 census. The feature of most concern in this respect is the operational form of the residence rules, which can affect the entire design of the questionnaire (see discussion below).
Recommendation 3.1: At this time, the Census Bureau should not initiate any further large-scale experiments designed to improve the initial mail response rate. Instead, response improvement research should now consolidate findings from research conducted to date in order to design experiments for the 1995 census test. The primary objective of these experiments should be to identify optimal field procedures that combine features such as advance notification, replacement questionnaires, and telephone follow-up.
Although the response improvement research has focused on the initial mail response, planning and operational tests for 1995 should also include an evaluation of telephone follow-up to improve mail response. Telephone follow-up is likely to be more productive than the use of the telephone tested in MTMT; outgoing telephone calls can be used both to remind people to return their forms and to complete the enumeration by telephone (for example, one week after a reminder call) using a CATI system. Current plans call for the telephone number of the household to be added to the master address file (MAF) for the 2000 census (Bureau of the Census, 1992e). This suggests that some of the planning needed to include telephone follow-up in the 1995 census test is already under way. The idea of conducting telephone follow-up has surfaced periodically and was even tested successfully in the 1980 census (Ferrari and Bailey, 1983). Preliminary results indicated that the use of telephone provided several advantages over follow-up by enumerators: lower costs per completed interview, lower item nonresponse rates for many items, and fewer duplicate questionnaires. Using telephone follow-up when possible would also reduce the number of enumerators required and make it possible to improve their quality by being more selective and by providing more extensive training. Furthermore, enumerators at telephone facilities would be centrally managed, scheduled, and
One disadvantage of telephone follow-up in 1980 was that directory listings were missing or outdated (Ferrari and Bailey, 1983). But telephone numbers could be added to the MAF using commercial firms that match telephone numbers with addresses, and this may improve the quality of telephone numbers. The 1995 census test can also build on other results from the 1980 experiment—for example, the finding that more questionnaires than expected were returned after cases were selected for telephone follow-up.
Recommendation 3.2: The prospect of having telephone numbers for a large percentage of households in the 2000 census is a potentially important development that should be explored in the Census Bureau's 1995 test—for example, by using the telephone for reminder calls and nonresponse follow-up.
In addition to using telephones for reminder calls and other nonresponse follow-up, the role of the telephone in designs such as SuperCensus or CensusPlus must also be examined. And if the telephone is to be used more frequently, the possibility of mode effects on enumeration (that is, on coverage) must be assessed (see discussion below). This is particularly important because telephone enumeration will be more likely among some groups than others. The differential biases associated with different modes can be accounted for in the models used to estimate the population unless mode effects can be estimated and controlled.
TARGETING MODEL AND TOOL KIT
The heart of the projects proposed by the Census Bureau's Special Methods Working Group is the development of the targeting model and tool kit. The targeting model is conceived as a predictive model that would identify geographic areas where enumeration barriers are likely to be present in 2000 and where deployment of special enumeration methods might be particularly useful. The model would also identify the types of enumeration problems that are anticipated, so that suitable methods would be drawn from the tool kit.
"Tool kit" refers to the collection of special methods—for example, team enumeration, "blitz" tactics, and bilingual enumerators—available for deployment in hard-to-enumerate areas. Use of the tool kit could also involve specialized outreach procedures and decisions at the headquarters level on mailout-mailback procedures, allocation of staff and resources, and differential pay rates and incentives for census workers.
Both the targeting model and the tool kit represent large development efforts, and the tool kit encompasses many potential research and evaluation projects. Presumably, the model and tool kit will be developed together, because one need only target barriers for which one has a potential solution in the tool kit. The development of the tool kit itself requires that the Census Bureau develop systematic distinctions
between promotion and outreach activities—between building awareness and developing motivation—and also identify which promotion and outreach activities should be conducted at the local and national levels.
Development of the targeting model should use the findings of the ethnographic research reports (e.g., de la Puente, 1993) that identify many structural conditions that make coverage of households and of people within households more difficult. For example, geographic areas with high-cost housing and low-income residents probably create conditions in which unrelated people crowd into small or irregular dwellings, and enumeration difficulties result. The Census Bureau is currently engaged in research that attempts to identify which demographic and housing characteristics—such as racial composition, size of household, and tenure—might be valuable in predicting which areas are particularly at risk for low mail response rates, high undercount rates, or high coverage error rates.
The targeting model and tool kit will require flexible implementation because the targeting model will not be based on current data, and the tool kit should be responsive to actual field conditions. The efforts and resources required to develop the targeting model, however, appear to be so massive that there is some risk that the tool kit, and the coordination between the targeting model and the tool kit may be underdeveloped and underevaluated. The basis for our concern is the same as that for our recommendation that the Census Bureau plan to "go local" in outreach, promotion, and (to some extent) enumeration by developing locally targeted outreach and promotion efforts: reducing the differential undercount requires decentralization and flexibility. Evaluation of the targeting model should consider who within the field management structure is given access to the model and its results and what tool kit resources are made available to field staff. (Our concerns about these issues are discussed more fully in the next sections.)
The targeting model and tool kit constitute a strategy for taking the census, and further research and development must proceed from this perspective. Currently, the targeting model and tool kit are presented more as field management tools rather than as full components of a census design. But the tool kit will ultimately include a set of enumeration, communication, outreach, and data collection methods, each of which requires development and experimental evaluation. We suggest that the various constituent components of the targeting model and the tool kit be specified more completely to help staff in identifying the kind of development and evaluation that each requires.
We believe it is particularly important that plans to evaluate the effectiveness of the targeting model and of components of the tool kit should be incorporated into the 1995 census test. Ideally, the 1995 census test would incorporate a trial of the targeting model and an evaluation—for example, by examining response rates by area. Even if the targeting model is not fully developed by 1995, techniques that are planned for the tool kit can be evaluated experimentally before, during, and after the census test. Posttest evaluation surveys can be used to determine whether people were aware of outreach and promotion efforts and can help in evaluating the effectiveness of such efforts. Components of the tool kit should be tested in
experiments such as those used to test ways of improving mail response rates. It may be possible to incorporate some tests into other field projects (e.g., experiments examining ways to improve coverage). It seems likely that the development and testing of the targeting model and tool kit will continue after the 1995 census test, and even after the 2000 census, and planning should take this schedule into account.
OUTREACH AND PROMOTION
The outreach and promotion program for the 1990 census was the most intensive to date for a decennial census. For the first time, a multifaceted mass media campaign addressed several traditionally undercounted groups as well as the general public. The community-based Census Awareness and Products Program (CAPP) was enhanced and began operations farther in advance of census day than in 1980. Programs were conducted that worked through national civic and religious organizations, schools, Head Start agencies, governmental units, and business organizations.
Based on an analysis of the data collected in the 1990 Outreach and Evaluation Survey, Bates and Whitford (1991) concluded that the Census Bureau's 1990 outreach and promotion program achieved many of its goals. The Advertising Council's mass media campaign received wide exposure, and it achieved the third highest media presence around census day in the six media markets where the coverage was monitored. By census day, over 90 percent of the population had recently heard or read something about the census, although the campaign was less effective in reaching blacks than whites and Hispanics.
Outreach and publicity may also help the census mail response rate; data from the Outreach and Evaluation Survey indicate that in 1990 the mail return rates of respondents with high awareness of census operations and knowledge of census uses were 15 to 20 percentage points higher than respondents with low awareness and knowledge (Bates and Whitford, 1991). (We cite these results with the standard caution that the question of whether these associations reflect causal relationships cannot be answered without considering potential interaction effects and controlling for other variables in a designed experiment.) Nevertheless, the mail response rate in 1990 was 10 percentage points below the 1980 census, and the differential undercount between blacks and nonblacks was the highest since the Census Bureau began estimating coverage in 1940 (U.S. General Accounting Office, 1992). Although there are undoubtedly many social factors that contributed to the response rate decline, it is also possible that the Census Bureau's 1990 outreach and promotion campaign did a better job of announcing the census than it did of persuading people to participate. Regardless, it seems clear that an even more intensive—and effective—outreach and promotion program will need to be designed, tested, and successfully implemented in the 2000 census if the Census Bureau is to prevent further erosion of census participation rates that began declining in 1970.
Components of Decennial Census Outreach and Promotion
The various components of the Census Bureau's decennial census outreach and promotion program include the establishment of a temporary Census Promotion Office and the conduct of a media campaign, national and local outreach efforts, and research and evaluation projects. In the last two censuses, Census Promotion Office has opened 2 years prior to the decennial census and closed after census operations were completed. It appears that the primary function of this office is to coordinate the media campaign with the Advertising Council and to oversee various national outreach programs conducted to promote the upcoming census. (The Census Bureau has a permanent Public Information Office, but it is not directly involved in decennial census promotion.)
In every census since 1950, the Census Bureau has relied on the Advertising Council to design and conduct the media campaign. (Working through advertising agencies that volunteer their time, the Advertising Council regularly conducts mass media advertising campaigns on a pro bono basis for government and nonprofit agencies.) In 1990 the Advertising Council for the first time designed multiple campaigns to target selected minorities (blacks, Hispanics, and Asian/Pacific Islanders) as well as the general public.
The principal national outreach program to undercounted groups is the National Services Program (NSP). Conducted by the Data User Services Division, the NSP is a continuing outreach and data dissemination program aimed at national organizations that represent undercounted minorities. The goal of the NSP is to secure the support of the national organizations and their local and regional chapters in efforts to encourage participation by the minority communities they represent. The Census Bureau also conducts a variety of national outreach activities close to census day: these include programs that target a variety of organizations, schools, governmental units and agencies, and private-sector corporations. The principal program for local outreach is the Census Awareness and Products Program (CAPP). Administered by the Census Bureau's Field Division, this program is activated a year or two prior to census day and deactivated shortly afterwards. In 1990 there were about 280 CAPP staff across the country, working out of the regional field offices.
The Census Bureau's decennial census outreach and promotion program does not include a structured plan for conducting research and development work. The absence of a permanent office responsible for decennial census outreach and promotion militates against sustaining an ongoing research program during the decade between censuses. Also, the Advertising Council does not permit its clients to undertake or commission media research on their own. The Census Bureau does, however, undertake retrospective research to evaluate decennial census outreach and promotion. Examples from the 1990 census include the Outreach Evaluation Survey, the Telephone Survey of Census Participation, the Survey of 1990 Census Participation, and the National Service Program Structured Debriefings.
The Special Methods Working Group of the Year 2000 Research and Development staff has expressed interest in outreach research for the 2000 census, but
this type of research does not appear to have a very high priority within the Census Bureau. We note, for example, that only 3 projects related to outreach and promotion are included in the list of 17 proposed special methods projects (Bureau of the Census, 1993g). One project is a study of methods to enumerate American Indians and Alaska native populations, which may include an outreach and promotion program. Another is a comparative study of paid versus pro bono advertising, and the third is a cost-benefit analysis to determine if the Census Bureau should retain an outside public relations firm. The Commerce Department Task Force committees and other advisory groups have assigned these projects priority rankings of 7, 15, and 16 respectively (out of 17), and we understand that it is doubtful whether the 2000 census research budget can support them.
We are concerned that the responsibility for outreach and promotion is split among several different units within the Census Bureau. We believe that the effectiveness and efficiency of the Census Bureau's decennial census outreach and promotion program could be improved if a permanent office were established and staffed with advertising and public relations professionals. This office would be responsible for planning, researching, and developing all outreach and promotion activities and for overseeing the implementation of the decennial program. It should be a permanent office to provide continuity between censuses and to monitor an ongoing national and local outreach and promotion program during the decade. The office should also be nonpartisan and insulated from political influence. Because the Census Bureau already has a permanent Public Information Office, we suggest that the Census Bureau consider expanding the mission of this office to include responsibility for the decennial census outreach and promotion program.
Although the establishment of a permanent decennial census outreach and promotion office would centralize the overall responsibility for outreach and promotion activities, it does not follow that such activities would become more focused on the national rather than the local level. Nor should they, as we argue below; a principal objective of the new central office should be to enhance outreach efforts at the local level. Just as the centralized Field Division is responsible for all data collection operations conducted by the regional offices, so a centralized decennial census outreach and promotion office would design, monitor, and support all field outreach activities conducted through those regional offices.
The one outreach and promotion activity that should not be consolidated under the purview of the new office is evaluation research. To preserve independence, this type of research should continue to be undertaken by other units within the Census Bureau, such as the Center for Survey Methods Research.
Recommendation 3.3: The Census Bureau should assign overall responsibility for decennial census outreach and promotion to a centralized, permanent, and nonpartisan office. The Census Bureau should consider expanding the mission of the extant Public Information Office to include this charge. Evaluation of outreach and promotion programs should be conducted by an independent unit within the Census Bureau.
Customized Local Outreach Programs and CAPP
One of the principal objectives of the 2000 census is to reduce the differential undercount by improving the participation rate of blacks, Hispanics, American Indians, Alaska natives, and other traditionally undercounted groups. It follows, then, that a key goal for the 2000 census outreach and promotion program is to identify ways of reaching the undercounted groups with a campaign that will not just announce the census, but also motivate them to participate.
It seems clear that the only way to reach the undercounted minorities effectively is to develop customized programs that use local channels for outreach and promotion. This message was repeated over and over again by the speakers at the recent Research Conference on Undercounted Ethnic Populations sponsored by the Census Bureau. It is also found in the Census Bureau's own research on the effectiveness of outreach efforts with ethnic minorities (see, e.g., Newhouse, 1992) and in our analysis of the Census Bureau's 1990 ethnographic research (see section below, ''Hard-to-Enumerate Populations''). The local programs should be developed in close consultation with the minority groups to which they are directed. The programs should also be designed to maximize the participation of minority organizations and representative in their implementation.
As many of the ethnic group representatives at the Research Conference on Undercounted Ethnic Populations pointed out, it is important to maintain some level of outreach with the undercounted ethnic minorities on an ongoing basis, in order to establish the trust necessary to ensure their participation at census time. Outreach and promotion on an ongoing basis would include not only efforts in the latter part of the decade to support decennial counting, but also postcensal feedback on census results and assistance with using census data. The CAPP program appears to be a good vehicle through which to establish and maintain relationships with the undercounted minorities at the local level. Therefore, we encourage the Census Bureau to expand this program further and support it on a continuing basis.
Overall responsibility for this program should be vested in a central office responsible for all decennial census outreach and promotion (see Recommendation 3.3 above); however, the program should continue to be implemented through the regional field offices. The CAPP staff assigned to the regional offices would be responsible for implementing customized local outreach programs. Statistics Canada coordinates all outreach programs through a central office that maintains links to provinces and local communities through regional offices. The Census Bureau might consider a similar approach.
We do not think that an ongoing CAPP program need be prohibitively expensive. On the contrary, most minority groups have a vested interest in ensuring that all of their constituents are counted at census time, and, as pointed out by several presenters at the Research Conference on Undercounted Ethnic Populations, local volunteer help is often available from the undercounted minorities themselves. Staffing may not need to be uniform over time. One or two core CAPP staff per regional office might be supplemented with temporary help during the period around
the decennial year. Given the potential of customized local outreach programs to reduce the differential undercount, we believe that the relatively modest cost of expanding the CAPP program would prove to be a wise investment.
Recommendation 3.4: The Census Bureau should commit the resources necessary to develop and implement customized, local outreach programs to target the traditionally undercounted ethnic minorities. The Census Awareness and Products Program (CAPP) should be expanded and sustained on an ongoing basis, so that it can serve as the primary vehicle for the design and implementation of these outreach programs.
Research and Development Program and the 1995 Census Test
Outreach and promotion research is not a principal component of the Census Bureau's 2000 census research and development program. We believe that the low priority assigned to this aspect of decennial operations is unfortunate in view of its importance to the success of the next census.
In our opinion, the Census Bureau should establish an ongoing research and development program for outreach and promotion. The program should be designed and conducted by a new decennial census outreach and promotion office. Two areas that should be explored are media advertising and local outreach. The advertising research program would focus on such issues as developing effective advertising messages and strategies for the undercounted minority groups, making increased use of local and regional campaigns, and using paid advertising to supplement the public service announcement campaign. The local outreach component of the research and development program would focus on developing and evaluating the customized local outreach programs described above.
The 1995 census test offers an excellent opportunity to evaluate alternative outreach and promotion strategies. If possible, the Census Bureau should include in the 1995 census test some experimental designs to evaluate alternative media campaigns and local outreach programs. At a minimum, the Census Bureau should implement the most promising of the campaigns and programs and then conduct an evaluation of their effectiveness. One way to do this would be to conduct an Outreach Evaluation Survey in 1995 patterned after the one conducted in connection with the 1990 census.
Recommendation 3.5: The Census Bureau should establish an ongoing research and development program for decennial census outreach and promotion. The 1995 census test provides an excellent opportunity to conduct and evaluate promising media campaigns and local outreach programs.
National Media Campaign
An important issue confronting the Census Bureau is whether to continue to rely on the Advertising Council to design and implement the national media campaign. As noted above, the Advertising Council's rules prohibit clients from commissioning any media research on their own. The Council also does not permit clients to supplement the pro bono campaign with any paid advertising. Thus, the Census Bureau is entirely dependent on the Advertising Council's pro bono campaign for its media research and advertising.
Because the Advertising Council depends on volunteer labor, we are concerned that its pro bono campaign may not include the same level of media research that large commercial advertisers have found beneficial. For example, the leading advertising agencies evaluate commercials by using sophisticated technology that continuously measures a focus group's response to a proposed commercial message. However, this type of research is expensive and may not be included within the scope of a pro bono campaign.
We also understand that there is a growing feeling among certain charitable organizations that national public service announcement campaigns are losing their effectiveness. One theory of this change is that U.S. society has become so diverse and complex that people are increasingly narrowing their focus to the community in which they live. As a consequence, there is a movement among pro bono advertisers away from the national campaigns toward increased use of local campaigns.
Certainly the Census Bureau must continue to rely on public service announcements run on a pro bono basis; it would be prohibitively expensive to launch an equivalent campaign on a strictly fee-for-service basis. However, the Census Bureau could discontinue use of the Advertising Council and instead work directly with local and regional agencies. This approach would allow the Census Bureau to undertake a paid media research program to identify the most effective advertising messages. It would also mean that the national campaign would include a collection of local and regional campaigns that might be more effective in reaching the communities to which they are directed.
Finally, this approach would allow the Census Bureau to supplement pro bono advertising with paid advertising—especially in hard-to-enumerate communities where a pro bono campaign may not provide sufficient or appropriate media exposure (Committee on National Statistics, 1978). We believe that the additional cost involved may be more than offset by the increase in response among the traditionally undercounted groups, with a corresponding reduction in the differential undercount. We note in this regard that Statistics Canada is a strong proponent of paid advertising: it was discontinued in 1986 as a cost-saving measure, but then promptly reinstated after a decline in participation.
Recommendation 3.6: The Census Bureau should evaluate the use of the Advertising Council to conduct the census media campaign. The Census Bureau should consider the alternatives of working directly with local and
regional agencies, undertaking paid media research, and supplementing pro bono advertising with paid advertising in hard-to-enumerate localities.
Starr (1992) analyzed why people called the Census Bureau's toll-free 800 number for help with the 1990 census. We commend this type of research; what better way to improve outreach in the next census than to analyze the problems people reported having in the last one. One of Starr's recommendations is that the Census Bureau consider a menu-driven touchtone call routing system that gives callers the opportunity to select options that address their concerns. We concur that this should be an important area for investigation and research.
Although some users may express frustration with menu-driven systems, these systems have grown rapidly in popularity, presumably because they are efficient and effective ways of serving customer needs. By the year 2000, menu-driven call routing is likely to be standard practice among large, customer-oriented businesses. We encourage the Census Bureau to undertake a literature review, develop a system for use in the 1995 census test or later dress rehearsals, and conduct an evaluation that includes the use of focus groups to assess people's reactions.
Recommendation 3.7: The Census Bureau should investigate developing a menu-driven touchtone call routing system for the 2000 census that gives callers to the Census Bureau's toll-free help line quicker access to the specific assistance they want.
We understand that the Census Bureau's Computer-Assisted Survey Information Collection (CASIC) staff has already begun looking at touchtone technology for various data collection applications. We suggest that the Year 2000 staff contact the CASIC staff for guidance in this area.
One of the most vexing and costly problems attending the enumeration process is the tendency to undercount minority and immigrant populations. The expense involved in attempting to reduce the differential undercount is considerable. Political pressure to redress the problem is also significant, because the nation's large cities want to ensure that undercounted individuals—often in need of city services—are properly represented in the count. Minority-based community organizations, especially (but not exclusively) those representing Native Americans, are particularly eager to see undercounting of their constituencies minimized because critical support funds from the federal government are often calculated on the basis of census information.
Overall, it is clear that one of the major tasks facing the 2000 census is to minimize the differential undercount through the development of more effective and specialized outreach, promotion, and enumeration methods. The 1995 census test must include some experiments with techniques aimed at the differential undercount in order to make effective cost-benefit analysis possible before the decennial census. One important consideration in assessing costs and benefits is the expense of following up nonrespondents, particularly in areas with low primary response rates. Some special methods, though intensive and costly, may have the potential to reduce the differential undercount and may do so partly by improving primary response rates in historically low response areas, thus reducing the workload and costs associated with nonresponse follow-up operations. Only when such special methods have been properly evaluated will it be possible to tell whether they reduce the differential undercount and help control the costs involved in the follow-up efforts that are implemented when people do not respond to the initial mail questionnaire.
Recognizing the urgency of these problems, the Census Bureau took constructive steps in 1990 to identify the reasons that immigrants and minorities are so hard to enumerate by conducting an alternative enumeration in 29 sample areas of the United States and Puerto Rico. The areas were chosen for their high concentration of particular minority subpopulations, urban and rural, as well as the presence of a large number of undocumented immigrants who are known to fear participating in the census. Experienced ethnographers familiar with the target populations conducted the alternative enumeration. They were able to improve the count by using their existing relationships with members of these communities, familiarity with the native languages of the immigrants, and their knowledge of the special housing practices of the poor—both native and immigrant. Results of these special census projects were compared with the enumerations of the same areas in the 1990 census.
From this comparison the Census Bureau learned a great deal about the sources of undercount and overcount. The summary by Brownrigg and de la Puente (1993) indicates that substantial numbers of erroneous enumerations and omissions occurred at the sites, and the net undercount rates varied considerably. The two extreme cases reported a net undercount of 47 percent and a net overcount of 53 percent; the median net coverage was a small undercount of approximately one percent. Clearly, the sites present very difficult counting conditions (discussed below) that create both types of census error. These errors occasionally cancel out within a geographic area, but they are more likely to result in surpluses of omissions or erroneous enumerations that produce the large variations in net undercount that were observed in 1990. For the affected communities, undercounting is a source of great policy concern with regard to issues of political representation and revenue sharing. The primary task now should be to capitalize on this research and to implement the valuable policy suggestions that emerged from this study in designing the 1995 census test and developing the tool kit of methods for the 2000 census field offices.
Manuel de la Puente's excellent summary of the 29 "ethnographic coverage reports" (de la Puente, 1993) identifies five sources of undercount or overcount: (1) irregular and complex living arrangements; (2) irregular housing; (3) residential
mobility; (4) distrust of government; and, (5) limited English proficiency. Almost all of these point to the difference between the underlying norms expressed in the census (nuclear families, discrete households, long-term or permanent residences) and the living arrangements of the nation's minorities, whether foreign born or native. Some of these problems are related to the difficulty of enumerating the population in question: gaining trust, counting all the people in a household, and the like. Others are related to locating dwellings in poverty areas of the nation's cities and rural communities and are of great importance in updating the master address file. The Census Bureau will need to take seriously the policy options proposed by the ethnographic studies for both of these domains.
Irregular and Complex Household Arrangements
The first source of enumeration difficulties involves the irregular and complex household arrangements that typify poor minority and immigrant communities. Households defined as irregular or complex contain unrelated individuals, people who are mobile or present for no other reason than to share the burden of the rent, and multiple nuclear families. Households structured in this fashion become extremely difficult to enumerate accurately. Census rules of residence, which ask the respondent to identify members of their household in relation to "person 1," cannot easily accommodate the composition of these living groups and often contradict the respondent's definitions of family or household. In such cases, residence rules are hard to apply; the individuals who "should" be counted may be absent and thus excluded. As an ethnographer working in rural Marion County, Oregon noted (cited in de la Puente, 1993:4):
If all members are not present, . . . obtaining the data pertaining to persons outside, asleep, at work, or temporarily absent is virtually impossible. It is as if those persons do not exist.
These households may also contain unrelated males who have assembled solely to cut the cost of the rent, who work long hours, and share living space in very dense fashion—with beds lining the rooms and individuals sleeping in shifts.1 Nothing, other than expediency and common ethnicity, binds these individuals into anything they would define as a "household" of people related to person 1, with emphasis on the term "related." As Rodriguez and Hagan put the matter (cited in de la Puente
For recent immigrants from Central America and Mexico, household and family are viewed as the same. Boarders and unrelated individuals are not part of the family and thus not part of the household.
At some sites ethnographers discovered that up to half of the individuals living in complex households of this kind were left out of the census (see, e.g., Romero, 1992). The same pattern of undercount was likely to develop in Haitian households, which were composed of a "core" family and a series of peripheral individuals who are "just passing through," but sometimes stay for years. Chinese families in New York may be equally perplexed by the census definitions of household and may augment those present in a household with relatives who are family (e.g., adult children) but no longer living in the house. Again, the residence rules on the census forms confuse even the willing participant—resulting in undercounts (the Haitian example) and overcounts (the Chinese example).
Irregular and complex household structures are generally to be found wherever there are high housing prices and a concentrated poor population (native or immigrant). This is an important finding in and of itself, for researchers (perhaps including the ethnographers working on this project) are prone to believe that these patterns are expressions of cultural differences. The regularity of the ethnographic findings across ethnic subpopulations and in disparate parts of the U.S. suggest there may be little that is culturally specific about formation of irregular households. Rather, irregular households form in response to structural conditions that bring the poor into areas of expensive housing, resulting in these fairly uniform "irregular" practices. If this is true, then it should be possible for the Census Bureau to identify many of these areas in advance and target special resources for enumeration within them.
Irregular housing is the second major source of the undercount, one that presents difficulties in locating housing units, rather than in enumerating the individuals living in them. Brownrigg (1991) estimates that across the 29 sample areas subjected to ethnographic evaluation, perhaps as many as 40 percent of the people who should have been included in the count, but were missed, were missed because the housing unit itself was overlooked or misidentified. Commonly, these dwellings were hidden from public view (in backyards or down rural roads) or were illegally built (and often concealed in single-family homes or garages). Irregular housing also causes overcounts because buildings have multiple addresses, multiple entrances, or are temporary dwellings that move around but have already been counted elsewhere—e.g., trailers in the neighborhood just for the weekend (see de la Puente, 1993:12). Irregular housing goes hand-in-hand with irregular and complex
households, thus compounding the enumeration problems discussed above. Families are likely to double up in areas where affordable housing is in short supply, stimulating the construction of illegally converted housing.2
The success of the ethnographic teams in locating irregular housing was impressive, particularly when compared with the 1990 census. One case study reported a census undercount of nearly 50 percent, much of which could be attributed to missed housing units. The alternative enumeration was often able to identify these irregular dwellings, drawing upon the detailed knowledge of locally-based ethnographers and/or their skill in cultivating "informants" who could guide them into the nether world of warehouses, back alleys, unscrupulous landlords, crack dens, and other unlikely places for private housing.
Residential mobility was the third reason given for enumeration difficulties in the 29 sample areas. Some populations that the census generally undercounts tend to move often, to reside in one place for a shorter period of time than what middle-class citizens think of as the norm, and so to be difficult to enumerate. High residential mobility go hand in hand with seasonal and low-wage labor markets, patterns of return migration, and the pressures of accommodating large numbers of "peripheral" members in a household (who are likely to move when the strain becomes too great). Difficulties with landlords give rise to eviction, resulting in greater mobility for poor people than others. Mahler (1993:8, quoted in de la Puente 1993:22) notes that Salvadoran immigrants move as often as three times a year in search of cheaper housing or jobs or because they are reunited with family members.
These circumstances not only make it difficult to count individuals, it exacerbates the difficulties that many of the native-born and immigrant poor face in applying the residence rules. Are temporary household residents "members"? They may be deemed such if they are also relatives, but otherwise they may be excluded because they are both highly mobile and unrelated. Experimentation with questions that ask about "who stayed here last night" may indicate ways in which these undercount problems can be attacked, since this formulation captures the mobile and the unrelated (an undercount problem) and may also cut down on the overcount of those who are kin but not living in that household on census day.
Distrust of Government
Distrust of, or ambivalence toward, the government is another cause of undercounts. In the 29 communities surveyed, many members of the target populations believe they have little to gain in cooperating with the census and fear the possible consequences of yielding information—particularly if they are among the large number of undocumented immigrants or participants in the underground economy. Contending with this problem proved difficult for the ethnographers who worked on this special project, for they were rarely equipped to survey drug enclaves, shanty towns, and the like. They had to contend with the dangers of the street and, perhaps more important, the reticence of ordinary (law-abiding) individuals living under these conditions to open their doors to strangers.
It is clear that many people in immigrant and native-born minority groups do not believe that the census is confidential. Those who harbor an undocumented individual in the household are especially fearful of discovery through participation in the census; others who are receiving public assistance are concerned about the government finding undeclared partners. It also seems likely that some immigrants would be wary of government initiatives to count and document individuals and skeptical about confidentiality claims because of their negative experiences with government activities in their native countries.
Limited English Proficiency
Finally, the ethnographic studies found that lack of English proficiency was a major source of undercounts in some communities. The studies made use of native-speaker enumerators who could conduct the census in Haitian creole, Spanish, Chinese, and a variety of other languages. Multilingualism proved to be enormously important, both enabling effective communication and inducing the respondent to trust the enumerator. In addition, immigrant populations are often characterized by a high incidence of illiteracy in the native tongue. Ethnographers sometimes found that members of their target population were unable to read or write in any language. Illiteracy necessitates even more intervention or interpretation in the enumeration process and undoubtedly introduces standardization problems. Coverage was undoubtedly improved in these case studies; consistency may be another matter.
It is important to remember that the Ethnographic Evaluation Project was conducted in particularly hard-to-count neighborhoods. Nonetheless, the project sheds light on the causes of undercounts and overcounts and suggests valuable policy initiatives. These initiatives need to be built into the research and implementation process for the 2000 census if the coverage issue is to be actively addressed. The
intensive nature of alternative enumeration methods will make them costly, but if they are effective in addressing differential undercounts they may end up saving the Census Bureau funds that might otherwise have to be used for follow-up.
The ethnographic research makes it abundantly clear that individuals familiar with the native languages, customs, and the physical layout of the communities in question were far more successful than traditional enumerators in locating and surveying hard-to-count populations. The project directors concluded, and this panel concurs, that the Census Bureau should create a more effective partnership with locally based, grass-roots organizations in order to address the coverage problem. Some of these organizations have regional and even national representation (e.g., the Urban League, Mexican American Legal Defense Fund); others are entirely local in orientation. These two kinds of ethnic organizations appeal to different groups and would require different partnership strategies with the Census Bureau, but both must be approached and integrated into the 2000 census.
Effective partnerships between the Census Bureau and local leadership will have to be inaugurated well before census day. There is reason to believe that a continuous outreach, promotion, and enumeration operation in targeted areas will end up being cost-effective in comparison with the start-and-stop approach that has been used. Developing local ties that work—that is, that help to reduce the differential undercount by creating a participatory spirit or simply by using the superior knowledge of local "ethnographers"—is not an overnight operation. If such efforts prove effective—as they have, to some extent, for Statistics Canada with regard to the Canadian aboriginal population—they should pay off in terms of reduced undercount and more effective targeting of Census Bureau resources.
Implementing an in-depth, localized, network approach will demand an enormous change in the culture of the Census Bureau where, for perfectly understandable reasons, centralized control and standardized methods have been paramount. Involvement of community organizations at the level suggested by the ethnographic studies will mean yielding a degree of control, and fostering a sense of trust, that would probably be unparalleled in the agency's history. Community leaders speak of the importance of defining the census as "theirs," the critical need to "own" the census, rather than approaching the census as an instrument that belongs to the government or to the nation as a whole. It is important to recognize how revolutionary ''going local" on this scale would be.
At the same time, the survival of many community organizations depends on the census, because their funding is linked directly to the size of the populations they serve. If for no other reason than self-interest, many Native American, Haitian, African-American, Hispanic, and Pacific Islander social service agencies are eager to help. Grass-roots groups have infrastructures that can be of enormous assistance in the enumeration process; they know where to find their people, they have lists of their members, and they are trusted in the community. These are assets that cannot be acquired by outsiders, and they are directly related to a successful enumeration. One concern with grass-roots efforts, however, is the potential for increases in erroneous enumerations and consequent overcounting when the involved organizations may be
interested more in the size rather than the accuracy of the count. Procedures will require adequate safeguards to prevent problems of this nature.
Recommendation 3.8: The Census Bureau should consider developing an extensive network of relations between field offices and local community resources. This infrastructure would be maintained in continuous operation between and during census years. The Census Bureau should develop and implement pilot programs in conjunction with the 1995 census test in order to gather information about the potential costs and benefits of a large-scale local outreach program.
There is a tendency to assume that irregular households are the product of culturally specific notions of family, but the evidence from the case studies suggests a surprising degree of uniformity in the residence practices of the nation's poor, whether immigrant or native-born. Further research is needed. Results must be incorporated into the targeting model and into any component of the census design that depends on comprehensive coverage (e.g., CensusPlus). A theoretical understanding of the distribution of the hard-to-enumerate population is needed, one that looks across ethnic and racial groups to the common characteristics of the poor; that theoretical knowledge is of considerable practical value in directing additional resources to field offices where differential undercounts are likely. The same modeling may help in determining the most effective placement of the standing infrastructure discussed in Recommendation 3.8 (above).
Recommendation 3.9: The Census Bureau should conduct further comparative studies of hard-to-enumerate areas, focusing on those parts of the country where three phenomena coincide: a shortage of affordable housing, a high proportion of undocumented immigrants, and the presence of low-income neighborhoods.
Ethnographers believe that enumerators who reflect the racial, ethnic, and cultural composition of the target population and who are residents of the community should be hired to conduct the census collection itself. Some of the advantages are trust, communication skills (especially foreign language skills), knowledge of the housing stock, and creation of the appearance (and reality) of ''local ownership" of the census. Communities that are suspicious of the federal government or that see no reward from participation are more likely to be persuaded of the importance of the census if the words come from people they know and have reason to admire or trust. One potential disadvantage with community enumerators is that respondents may be reluctant to give accurate information about more sensitive items, such as income, to a friend or neighbor.
Local enumerators are more likely to know the ins and outs of the neighborhoods that feature irregular housing, but further training and support for locating hidden housing units will be necessary. The ethnographic studies recommend
making more systematic use of mail carriers, not to conduct the count, but as sources for identifying housing units. Rental offices and landlords turn out to be poor sources of information, but mail carriers are disinterested parties whose knowledge base is currently underused.
Clearly a large-scale program of "ethnic enumerators" would require a significant change in traditional training methods. More resources would have to be available to train enumerators, and the process would take longer. It would be foolhardy to hire people and provide them with insufficient training. But the investment would be well worth it if the program contributed to reducing the differential undercount—particularly if it was tied to an ongoing organizational structure (which would not have to be reborn with every decennial census).
Ethnographers working in Spanish-speaking areas noted the importance of developing Spanish-language census forms based on a conceptual rather than a literal translation and distributing these forms more effectively than in 1990, when people were required to call a toll-free number to request a Spanish-language form. They reported frustration on the part of Spanish speakers who were unable to participate even though they are literate in their native tongue. This would seem a relatively simple and cost-effective recommendation. Every form that can be completed by mail saves on the cost of follow-up. Although this was the only non-English-language form suggestion, further research on the efficacy of translating the form into other languages might be useful. Further research is needed to assess whether the use of foreign-language tapes or interactive video might be effective ways of reaching individuals who are not literate in their native language.
Immigrant families working shifts and seasonal jobs are often difficult to enumerate because their schedules are irregular. Ethnographers recommend that enumerators conduct visits in the evening and on weekends, rather than in the middle of workdays in order to find more people at home. They also noted the importance of moving the census day so that it falls in the middle of the month; occupants of low-cost housing are more likely to move at the beginning and the end of a month. Hence, a change in the timing of the census itself will reduce that part of the undercount that is attributable to residential mobility. (Statistics Canada has already changed census day for essentially the same reasons. Officials concluded that part of the undercount problem could be solved simply by moving the census date from the first day of the month. The Census Bureau should consider following suit.)
Recommendation 3.10: In the 1995 census test, the Census Bureau should evaluate specific measures and procedures that might improve the enumeration of historically undercounted populations. Candidates for study in 1995 should include a larger repertoire of foreign-language materials (both written and audio), more aggressive hiring of community-based enumerators, and greater flexibility in the timing of enumeration (i.e., contact during evenings and weekends). In particular, the Census Bureau should examine the efficacy of moving census day to the middle of the month.
Ethnographers noted the increasingly outdated conceptualization of race and ethnic identity embodied in the census. America is becoming a multicultural society, and personal identities are changing alongside this trend. The selection of single-race identifiers is proving particularly problematic under these circumstances. Are immigrants supposed to define themselves by nationality, ethnicity, language group, or physical appearance? Are individuals supposed to select a category that expresses how they view themselves or how others define them? This is a source of great confusion. Laotian nationals of Chinese descent and Hmong origin do not know which way to turn. Spanish-speaking Filipinos are unsure whether they should check "Pacific Islander" for race and then "Hispanic" for origin.
The single identification items are based on a long-standing conceptualization of race (and ethnicity) as unambiguous, whereas contemporary identification is more fluid and situational. The United States is starting to look more like Brazil or Tahiti, among the many places in the world where fluid racial and ethnic identity is the norm and where single-category identifications are not easily reconciled with personal conceptions of race. As immigration patterns bring more people from these societies (e.g., the Dominican Republic) to the United States, patterns of identification are shifting. The Census Bureau needs to invest time and support in research on these questions, and policy makers need to clarify the purpose of the identification questions themselves. The Panel on Census Requirements in the Year 2000 and Beyond is currently investigating the subject of racial and ethnic classification with regard to content needs in the decennial census and other demographic programs.
These problems are not contributing to the undercount, in the sense of "missing persons," but they are at the heart of another kind of problem: ethnic and racial misclassification. The Census Bureau will continue to feel pressure to solve this problem as long as resource flows to ethnic community organizations depend on the accurate classification and counting of their members. We recognize that these are extremely difficult problems to solve, and we recommend, among other things, that the Census Bureau experiment with multiple check-off classification schemes in the 1995 census test. We further suggest that the Census Bureau sponsor additional research on the question of ethnic identity, utilizing the intellectual resources of cognitive anthropology (where much relevant research has already been done) and sociology.
Recommendation 3.11: The Census Bureau should consider a major program of research in cognitive anthropology, sociology, and psychology that will comprehensively examine the issue of racial and ethnic identity. This research would contribute to the development of more acceptable racial and ethnic identification questions. In particular, the Census Bureau should consider experimenting with allowing people to select more than one race category in the 1995 test.
The policy recommendations that emerged from the ethnographic studies are of enormous value for the 2000 census. What remains to be seen is how they can be
incorporated into the tool kit: it is unclear that there is any meaningful link between the two enterprises. It certainly is insufficient to list "ethnography" as part of the tool kit; the word has no content for local census officials. What must be spelled out are the strategies of outreach and enumeration that an ethnographically inspired approach would entail. For the most part, this would seem to involve localized, customized methods of recruiting enumerators drawn from minority populations, providing ongoing outreach support to their organizations, and using their greater trust and recognition within hard-to-enumerate populations as a basis for a partnership. Instructions need to go to field offices about how to find these people, particularly if they are disconnected from large and nationally known organizations (e.g., the Urban League). Training programs need to be developed, and they will take time.
It should be noted that what ethnography provides is not a set of methods for collecting the census, but a knowledge base for mapping a community and an approach for recruiting culturally knowledegable enumerators who can do an effective job in working with hard-to-enumerate populations.
Promotion, outreach, and greater flexibility in methods of enumeration will probably have their greatest effect on household coverage, not on coverage of people within households. We emphasize that improving coverage of persons within households—and reducing the contribution of within-household coverage to the differential undercount—requires reducing response error, because it is household respondents who implement the residence rules as they fill out the census form or talk with a Census Bureau enumerator.
Areas with large minority and renter populations, homeless persons, and children in shared custody arrangements are examples of cases that present special problems in accurately applying the Census Bureau's residence rules. The Living Situation Survey (LSS), a national household sample weighted toward minority populations, is an attempt to better understand the living situations in these groups. This survey is expected to help the Census Bureau determine the effects of complex living arrangements on coverage and residence rules. The results from the LSS and the ethnographic evaluations conducted for the 1990 census can contribute to improving the conceptualizations underlying residence rules as well as their application. Results from the LSS will probably not be available in time to directly influence current tests of new residence rules and their application, but the Census Bureau's research staff is aware of the potential contribution of the LSS, and we expect the knowledge gained from developing the LSS to have an indirect effect on the design of residence rules for the decennial census.
We understand that current plans for research on improving within-household coverage include testing a version of the residence rules in which respondents begin by listing people who have stayed in the household and then go through the steps of adding people who meet and subtracting people who do not meet residence criteria.
If the 2000 census is to use de jure residence rules, we think that beginning with a kind of de facto listing and then applying the rules systematically promises to increase the accuracy with which respondents apply the Census Bureau's rules. We also think that such an approach is more likely to minimize mode effects than the approach used in 1990.
Once again, we recognize that not all issues can be taken into account in designing such a complex project. But we recommend that two census forms—one to be self-administered and one to be administered by an interviewer over the telephone—be developed simultaneously. It is important to have this dual goal in mind while the instrument is being designed, because some features can be moved more easily from a telephone to a self-administered presentation than vice versa. The goal of simultaneous development is a self-administered questionnaire that can be given a comparable implementation for telephone. This comparability is particularly important for census residence rules. The research program would have to determine at the outset whether a census telephone interviewer would use computer-assisted telephone interviewing or a paper instrument because these could require very different designs.
Particularly if the telephone will be used for follow-up in 2000, the comparability of self-administered and telephone administered instruments must be evaluated. Such an evaluation could use experimental treatments, such as the following: (1) respondent completes a self-administered instrument, (2) respondent who is looking at a self-administered form is interviewed by telephone, and (3) respondent who is not looking at a self-administered form is interviewed by telephone. This evaluation would not necessarily require a study of the magnitude of the MTMT or the other response improvement studies. Information on the presence and size of mode effects on within-household coverage could also be important in correctly specifying the models used in mixed-mode coverage measurement methods, such as SuperCensus and CensusPlus.
The experiments to improve within-household coverage with new forms of residence rules discussed above offer an opportunity to test a form for the 1995 census test, and the Census Bureau should take full advantage of this opportunity. If possible, cognitive laboratory testing should be conducted using final versions of the instrument. It appears that the questionnaire designed in these experiments will build on the organizational experience gained in the Simplified Questionnaire Test (SQT). Research on residence rules in the LSS is also likely to extend well beyond the 1995 test. This project obviously has implications for almost any of the census designs and for other surveys.
Recommendation 3.12: When developing and applying residence rules, the Census Bureau should consider both the need to accurately enumerate diverse household structures and the potential for mode effects when an instrument is implemented in both self-administered and interviewer-administered forms. In particular, the Census Bureau should simultaneously develop enumeration forms designed for self-administration and telephone administration for use in