Blog & News
New Brief Examines Collection and Availability of Data on Race, Ethnicity, and Immigrant Groups in Federal SurveysSeptember 09, 2021:
Measurement of race, ethnicity, and immigration (REI) status is a critical component of research used to inform policy. However, the standards for measuring these concepts are often confusing for analysts. A new SHADAC brief aims to assist state and federal analysts with survey development and/or analysis of existing survey data to generate estimates of health insurance coverage and access to care across racial and ethnic groups and according to nativity and/or immigrant status.
The new brief presents the collection and classification of survey data for populations defined by race, ethnicity, and nativity/immigrant (REI) status as well as the availability of these data in public use files. We focus in particular on surveys that are conducted by federal agencies and that collect information about health insurance coverage and access to care on an annual or periodic basis for the general population of the United States. Because policy decisions and funding priorities are typically made at the state and local level, the brief emphasizes federal sources that afford state-level estimates as well.
The table below lists the surveys included in the brief, the federal agency responsible for each survey, and whether national and state estimates are available.
|American Community Survey (ACS)||Census Bureau||Yes||Yes|
|Behavioral Risk Factor Surveillance System (BRFSS)||Centers for Disease Control and Prevention (CDC)||Yes||Yes|
|Current Population Survey (CPS)||Census Bureau||Yes||Yes|
|Medical Expenditure Panel Survey – Household Component (MEPS)||Agency for Healthcare Research and Quality (AHRQ)||Yes||No|
|National Health Interview Survey (NHIS)||National Center for Health Statistics (NCHS)||Yes||No|
|National Survey of Children’s Health (NSCH)||Census Bureau||Yes||Yes|
|Survey of Income and Program Participation (SIPP)||Census Bureau||Yes||Yes|
While all these surveys must adhere to Office of Management and Budget (OMB) standards for the collection and classification of race and ethnicity in federal surveys, the organizations conducting each survey implement these guidelines differently, resulting in race and ethnicity data that are not always comparable across surveys. In addition, differing design components in these surveys such as target population, sample design and size, and year and frequency of data collection, also account for comparison difficulties.
An Illustration of Measurement Variation: Ethnicity Measures
|Ethnicity Group Categories Collected in Selected Federal Surveys|
|Not Hispanic or Latino||X||X||X||X||X||X||X|
|Hispanic or Latino||X||X||X||X||X||X||X|
|Mexican, Mexican-Am., Chicano||X||X||X||X||X|
|Central or South American||X|
|Central American, exc. Salv.||X*|
|Other Central American||X*|
|Other South American||X*|
|Other Latin American||X|
|Multiple Categories Allowed||X||X||X||X|
|Other Category Specified||X||X||X||X|
|*Asterisks denote categories that are not specified in survey questions but are reported as open-ended “other” responses.|
The differential implementation of OMB standards can be seen in the measurement of ethnicity across the seven surveys included in this brief. In federal surveys, “ethnicity” refers specifically to Hispanic ethnicity. The minimum OMB standard for collecting data on ethnicity includes the options: 1) Hispanic/Latino or 2) Not Hispanic/Latino. All seven surveys ask about Hispanic or Latino origin—and in some cases, Spanish as well—using three main formulations for the ethnicity question: whether the respondent is of Hispanic, Latino, or Spanish origin; what Hispanic/Latino group represents the respondent’s ethnic background; or whether the person considers themselves to be Hispanic or Latino. All seven surveys also include further identifying categories of Hispanic origin as a follow-up question to the primary ethnicity question.
State and Local Level Data
Also included in this new brief are sample sizes for selected REI groups by state for the five surveys that contain state identifiers in their public use files (ACS, BRFSS, CPS, NSCH, and SIPP). However, it is important to note that sample sizes for racial, ethnic, or immigrant groups vary depending on the state or the REI group of interest. Even with publicly available geographic identifiers, the sample sizes for specific REI groups at lower levels of geography may be too small to produce reliable estimates. For example, the BRFSS has a relatively large national 2019 sample of American Indian and Alaskan Native (AIAN) (n=6,639) and Asian (n=8,531) subpopulations. However, there are only 170 AIAN and 287 Asian respondents in the 2019 sample for Minnesota, precluding comprehensive research on health equity for these groups in this state that has a large Asian population and one of the larger American Indian populations.
Download a PDF of the brief for further information.
Collection of Race, Ethnicity, Language (REL) Data in Medicaid Applications
State Health and Value Strategies Brief by SHADAC, 2021