Methods for Collecting and Reporting Hispanic Subgroup Data Need Refinement
GAO-03-228, Jan 17, 2003
To help boost response rates of both the general and Hispanic populations, the U.S. Census Bureau (Bureau) redesigned the 2000 questionnaire, in part by deleting a list of examples of Hispanic subgroups from the question on Hispanic origin. While more Hispanics were counted in 2000 compared to 1990, the counts for Dominicans and other Hispanic subgroups were lower than expected. Concerned that this was caused by the deletion of Hispanic subgroup examples, congressional requesters asked us to investigate the research and management activities behind the changes.
In both the 1990 and 2000 census, Hispanics could identify themselves as Mexican, Puerto Rican, Cuban, or other Hispanic. Respondents checking off this latter category could write in a specific subgroup such as "Salvadoran." The "other" category in the 1990 Census included examples of subgroups to clarify the question. For the 2000 Census, the Bureau removed the subgroup examples as part of a broader effort to simplify the questionnaire and help improve response rates. The Bureau removed unnecessary words and added blank space to shorten the questionnaire and make it more readable. Although the Bureau conducted a number of tests on the sequencing and wording of the race and ethnicity questions, and sought input from several expert panels, no Bureau tests were designed specifically to measure the impact of the questionnaire changes on the quality of Hispanic subgroup data. According to Bureau officials, because federal laws and guidelines require data on Hispanics but not Hispanic subgroups, the Bureau targeted its resources on research aimed at improving the overall count of Hispanics. Bureau evaluations conducted after the census indicated that deleting the subgroup examples might have confused some respondents and produced less-than-accurate subgroup data. A key factor behind the Bureau's release of the questionable subgroup data was its lack of adequate guidelines governing the quality needed before making data publicly available. As part of its planning for the 2010 Census, the Bureau intends to conduct further research on the Hispanic origin question, including a field test in parts of New York City. However, until research on a new version of the question is finalized, Bureau officials said that other census surveys will continue to use the 2000 Census format of the Hispanic origin question.
- Closed - implemented
- Closed - not implemented
Recommendations for Executive Action
Recommendation: To ensure that the 2010 Census will provide data users with more accurate information on specific Hispanic subgroups, the Secretary of Commerce should ensure that the Director of the U.S. Census Bureau implements Bureau plans to research the Hispanic question, taking steps to properly test the impact of the wording, format, and sequencing on the completeness and accuracy of the data on Hispanic subgroups and Hispanics overall.
Agency Affected: Department of Commerce
Status: Closed - Implemented
Comments: The Census Bureau concurred with the recommendation. As a part of the 2003 National Census Test, the Census Bureau examined the effects of word changes on the reporting of specific Hispanic or race groups. The changes included reinstating the word "origin," removing "some other race" option, adding Hispanic subgroup and race examples, and modifying the instructions about answering the questions. Additional testing was done the following year as part of the Census Bureau's 2004 Census Test where the Bureau again tested the impact of wording changes and the use of examples in response categories. The final large test relating to this program area was the 2005 National Content Test. These major tests are to be supplemented by cognitive research and tests related to the use of a bilingual questionnaire. By early 2007, the Census Bureau expects to finalize the race and Hispanic-origin questions for the 2010 census and the American Community Survey (ACS). They are to be implemented for the first time in the 2008 ACS and will then be used in the 2008 Dress Rehearsal and the 2010 Census.
Recommendation: As GAO recommended in its companion report on the homeless and others without conventional housing, the Bureau should develop agencywide guidelines governing the level of quality needed to release data to the public, when and how to characterize any limitations, and when it is acceptable to delay or suppress data.
Agency Affected: Department of Commerce: Bureau of the Census
Status: Closed - Implemented
Comments: The Census Bureau concurred with the recommendation. In response, the Associate Director for Methodology and Standards, the Chairperson of the Census Bureau's Methodology and Standards Council, formed the Inter-Directorate Quality Framework Standards Working Group to examine this issue and develop/codify guidelines as recommended. The Working Group's guidelines, "Dissemination of Census and Survey Data Products", became effective on May 19, 2005.