Efficacy of Federal Data: Revised Office of Management and Budget Standard for Native Hawaiian and Other Pacific Islanders Examined

Summary

This policy brief examines the status of federal data since the implementation of the 1997 Revised OMB 15 standards for the collection of race and ethnic data, identifies ongoing data limitations, and present recommendations to improve policy and interventions for Native Hawaiians and other Pacific Islanders (NHPI). While most federal agencies are taking appropriate steps to comply with the revised OMB standards, many are having less success reporting disaggregated information on NHPIs. This suggests that increased efforts to obtain robust samples of NHPIs warrants immediate attention in order for federal agencies to fully comply with the revised OMB standards.

Introduction

Research and survey findings have strong implications for policy and program development. Federal agencies rely on robust data collection of administrative records (e.g., vital registries for births and deaths) and national surveys, including the National Health Interview Survey (NHIS) and the Current Population Survey, to make decisions that ultimately influence the distribution of resources and services nationwide. Failure to produce highly reliable estimates on numerically small diverse populations at the national level compromises effective planning and interventions to address their social, economic, and health concerns (Department of Health and Human Services, 1999; Institute of Medicine [IOM], 2009; Williams, 1999). Consequently, identifiable segments of the total population risk being overlooked, understudied, and underserved (IOM, 1998; Panapasa, Weed, and Atkinson, 2009; Panapasa et al., 2010; Williams, 1999).

In 1997, the Office of Management and Budget (OMB) implemented a new racial and ethnic category that disaggregated Asian Americans and Pacific Islanders (AAPIs) into two groups: Native Hawaiians and Other Pacific Islanders (NHPIs) and Asians, and mandated that federal agencies collect and report data using the new racial and ethnic categories by the year 2003 (OMB, 1997). Several reasons led to the separation of NHPIs from Asians as a single federal racial/ethnic category. Some of the important reasons were: (1) NHPIs, compared to Asians, have higher rates of many chronic diseases and are more socio-economically and socio-culturally disadvantaged—issues masked by aggregation with Asians (Blaisdell, 1993; Braun et al., 1997; Chen and Hawks, 1995; Chen et al., 1993; Department of Health and Human Services, 1995; Hoyert and Kung, 1997; Lin-Fu, 1988); (2) because Asians were overrepresented in higher education, many NHPI college students were adversely affected by graduate schools’ admission policies to limit the enrollment of Asians and were bypassed for scholarships (Lin-Fu, 1993; U.S. Census Bureau, 1991); and (3) disproportionate allocation of federal resources and support did not match the extent of the medical, social, and economic issues faced by NHPIs given their population size relative to Asians (Fernandez, 1996; LinFu, 1993; Yu and Liu, 1992). The disaggregation of NHPIs from AAPIs represents a major step toward accurately characterizing these distinct populations across various key measures and outcomes (Bitton, Zaslavsky and Ayanian, 2010; Panapasa et al., 2010; Srinivasan and Guillermo, 2000). Additionally, the revised OMB standards reflected the proper attention to the diverse characteristics of Native Hawaiians, migrants from the U.S. Associated Pacific Islands, and immigrants from Pacific Island countries.

This brief represents a review of select federal data sets since the implementation of the revised OMB standards as means of highlighting existing data limitations. Based on this review, recommendations to improve federal data and information on NHPIs are provided.

Methodology

Data from six federal agencies—the Department of Commerce, the Department of Health and Human Services, the Department of Education, the Department of Agriculture, the Department of Housing and Urban Development, and the Department of Justice—were examined for their compliance with the revised OMB standards. Data set selection was based on (1) accessibility, (2) degree of national coverage of the U.S. population, and (3) the potential source of information for policy and intervention. The data sources identified are by no means exhaustive but do represent a useful cross-section of studies that collect and report race and ethnicity data post revised OMB standard (Department of Commerce, 2011; Department of Housing and Urban Development, 2011; National Center for Education Statistics, 2011; National Center for Health Statistics, 2011). Specifically, we examine whether the select federal agencies are in full compliance with the revised OMB 15 standards: to collect and report disaggregated AA and NHPI information.

Findings

Overall, sixteen of the sixteen administrative data and national surveys are in compliance with the revised OMB standard for collecting disaggregated NHPI information. However, of the sixteen data sets, only one is in compliance in reporting (U.S. Census) and two are partial (American Community Survey and Vital Fertility and Mortality Events). Significant problems persist in reporting disaggregated NHPI information. Table 1 presents a description of existing NHPI documentation in national data across various federal agencies and a notation of their compliance with the revised OMB standards.

Table 1.

Description of Select Federal Data Sources and Complaince with Revised OMB 15 Standard on the Collection and Reporting of Native Hawaiian and Other Pacific Islander Data

Federal Agency and Data Source	Collecting Data Using Revised OMB 15 Categories?	Race/Ethnicity Identifiers for Data Collection?	Reporting Data Using Revised OMB 15 Mandate?	Race/Ethnicity Identifiers for Data Collection?
1. Department of Commerce
US Census FY 2000, 2010	Yes	All	Yes	Detailed Native Hawaiian and Pacific Islander Race
American Community Survey	Yes	All	Partial	Native Hawaiian, and Other Asian American Pacific Islander
Current Population Survey (CPS)	Yes	All	No	Asian American Pacific Islander
2. Department of Health and Human Services
Vital Fertility and Mortality Events	Yes	All	Partial	Native Hawaiian, Samoan, Guamanian/Chamorro, and Other Asian Pacific Islander
National Longitudinal Mortality Study				Hawaiian, Guamanian, Samoan, “Other”
National Health Interview Survey (NHIS)	Yes	All	No	Not Releasable
National Health and Nutrition Examination Survey (NHANES)	Yes	All	No	Other Race
National Survey of Family Growth (NSFG)	Yes	All	No	Other Race
Behavioral Risk Factor Surveillance System (BRFSS)	Yes	All	No	Hawaiian/Pacific Islander
National Hospital Ambulatory Medical Care Survey (NHAMCS)	Yes	All	No	Hawaiian/Pacific Islander
National Hospital Discharge Survey	Yes	All	No	Hawaiian/Pacific Islander
National Survey on Drug Use and Health (NSDUH)	Yes	All	No	Hawaiian/Pacific Islander
3. Department of Education
Early Childhood Longitudinal Survey
Kindergarten Cohort (ECLS-K)	Yes	All	No	Hawaiian/Pacific Islander
National Household Education Survey (NHES)	Yes	All	No	Unknown
National Crime Victimization Survey	Yes	All	No	Hawaiian/Pacific Islander
School Survey on Crime and Safety (SSOCS)	Yes	All	No	Unknown
Medical Expenditure Panel Survey	Yes	All	No	Asian American Pacific Islander, Single Race and “Other Races/Multiple Races”
4. Department of Agriculture
Food Stamp Quality Control Database (FSPQC)	Yes	All	No	Hawaiian/Pacific Islander
5. Department of Housing and Urban Development
American Housing Survey (AHS)	Yes		No	Hawaiian/Pacific Islander
6. Department of Justice
Census of Jails	Yes	All		Hawaiian/Pacific Islander

Source: University of Michigan, Interuniversity Consortium for Political and Social Research.

Reporting issues that violate the intent of the revised OMB standard generally fall into one of two approaches. The NHIS, for example, reports race information for small case size respondents as “not able to release,” thus reflecting a general failure of most national sampling frames to capture representative samples of U.S. subpopulations. The other approach is to clump small racial groups into aggregate categories such as “AAPI” or “Other Races.” Neither approach aids in analysis but instead introduces potential biases (e.g., overestimates for some populations and underestimates for others) into any studies that use these categories due to unmeasured heterogeneity.

Recommendations

Evidence-based research is essential to develop effective policies and interventions. Despite the implementation of the revised OMB standards, federal data collection on NHPIs remains inadequate because sample sizes are too small to produce reliable estimates. New legislation and commitment of resources are needed to obtain robust data and reliable estimates. Representative samples of NHPIs need to be increased across national surveys and reporting of results need to be improved. Therefore, we propose the following three recommendations.

1. Collect and Report Reliable Disaggregated Information on NHPIs to Fully Comply with the Revised OMB 15 Standards

   Inclusion of separate “Native Hawaiian and Other Pacific Islander” identifiers and increased sample sizes are essential to achieve meaningful policies and interventions for NHPIs. Unlike other racial and ethnic groups, the relatively small number of NHPI respondents in national surveys will continue to be a problem without specific oversampling strategies. At a minimum, the collapsing of NHPI and Asian populations into an aggregate AAPI classification should be avoided. Studies have shown clear evidence that this practice introduces significant bias into measurements and indicators for both groups (Cho and Hummer, 2001; Miller et al., 2008; Srinivasan and Guillermo, 2000). The ongoing use of higher-order aggregations such as “Other Asian Pacific Islander” is also unacceptable as it merely reflects the failure of current survey designs to capture the real composition of the U.S. population. For example, the National Center for Health Statistics currently aggregates and reports vital events for “Other Asian Pacific Islanders” instead of two separate categories: “Other Asians” and “Other Pacific Islanders.”

2. Develop Appropriate Methodologies for Data Collection, Tabulation, and Reporting Strategies to Produce Reliable Estimates on the NHPI Population.

   Reliable estimates on NHPIs outcomes are essential to fully satisfy the requirements of the revised OMB standard. Increase support for investigator-initiated research projects in survey methodology for hard-to-reach NHPI subpopulations is necessary to improve the reliable reporting of data on the NHPI populations.

3. Develop NHPI Community Partnerships

   NHPI community advocates represent valuable resources to improve research and help contextualize data collection in local communities. The U.S. Census Bureau Advisory Committees have shown how this model helps capitalize on community input (U.S. Census, 2011). Active NHPI community involvement empowers advocacy for meaningful policy and interventions and aids in the implementation of evidence-based research. Community support also helps to reduce refusal rates by NHPIs in surveys and ensure the accurate collection of sensitive or difficult questions regarding health, finances, or family life.

Implementation of these three recommendations would more accurately identify and characterize the needs of the NHPI Americans—a significantly underserved population in the United States.

Biographies

Sela V. Panapasa, PhD, is Assistant Research Scientist with the Research Center for Group Dynamics, Program for Research on Black Americans at the University of Michigan’s Institute for Social Research. She is a sociologist and demographer who specialize in measuring NHPI socioeconomic status and health outcomes.

Kamana’opono M. Crabbe, PhD, is the Research Director for the Office of Hawaiian Affairs in Honolulu, Hawai’i, and a licensed clinical psychologist.

Joseph Keawe’aimoku Kaholokula, PhD, is Associate Professor and Chair of the Department of Native Hawaiian Health in the John A. Burns School of Medicine at the University of Hawai’i at Mānoa, Honolulu. He is also deputy director for the Center of Native and Pacific Health Disparities Research at the John A. Burns School of Medicine and a clinical health psychologist.