Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Nov 21.
Published in final edited form as: Int J Intercult Relat. 2007 Sep;31(5):561–573. doi: 10.1016/j.ijintrel.2007.02.001

Performance of the U.S. Office of Management and Budget’s Revised Race and Ethnicity Categories in Asian Populations*

Joan L Holup a, Nancy Press b, William M Vollmer a, Emily L Harris c, Thomas M Vogt d, Chuhe Chen a
PMCID: PMC2084211  NIHMSID: NIHMS30410  PMID: 18037976

Abstract

Objectives

The U.S. Office of Management and Budget (OMB) guidelines for collecting and reporting race and ethnicity information recently divided the “Asian or Pacific Islander” category into “Asian” and “Native Hawaiian or Other Pacific Islander”. The OMB’s decision to disaggregate the “Asian or Pacific Islander” category was the first step toward providing these communities with information to better serve their needs. However, whether individuals who formerly made up the combined group categorize themselves as the new guidelines intend is a question analyzed in this report.

Methods

A subset of adults participating in the Hemochromatosis and Iron Overload Screening Study completed both the OMB-minimum and the expanded race and ethnicity measure used in the National Health Interview Survey. We compared responses on the expanded measure contained within the OMB “Asian” definition (Filipino, Korean, Vietnamese, Japanese, Asian Indian, Chinese, and/or Other Asian) to “Asian” responses on the OMB-minimum measure.

Results

Mixed heritage Asians less often marked “Asian”. Among mixed heritage Japanese, Chinese, and Filipinos, 27%, 49%, and 52% did not mark “Asian” on the OMB measure, respectively. Eleven percent of single-heritage Filipinos did not mark “Asian.”

Conclusions

Many individuals formerly making up the combined “Asian or Pacific Islander” group do not categorize themselves as the revised OMB guidelines intend. This is particularly evident among Filipinos and among Asians of mixed heritage. This research illuminates the reliability and utility of the broad “Asian” category and points to possible consequences of collapsing groups into a single category, i.e., missed information and/or erroneous generalization.

1. INTRODUCTION

Among the revisions to the U.S. federally sponsored collection and reporting of race and ethnicity information guidelines, the Office of Management and Budget (OMB) divided the “Asian or Pacific Islander” category into two categories — “Asian” and “Native Hawaiian or Other Pacific Islander.” The revised OMB standards now have five minimum categories for race: American Indian or Alaska Native; Asian; Black or African American; Native Hawaiian or Other Pacific Islander; and White. There are also two ethnicity categories: “Hispanic or Latino” and “Not Hispanic or Latino.” These classification revisions received considerable attention (Friedman, Cohen, Averbach, & Norton, 2000; Parker & Makuc, 2002; Sondik, Lucas, Madans, & Smith, 2000; Srinivasan & Guillermo, 2000; Wallman, Evinger, & Schechter, 2000). The decision to divide “Asian or Pacific Islander” was driven by a desire to appropriately distinguish the Pacific Islander category. As Native Hawaiians made up only 3% of the broad “Asian or Pacific Islander” group (Wallman et al., 2000), the ability to describe their health, socioeconomic status, and monitor potential discrimination had been minimal (Office of Management and Budget, 1997).

The OMB’s decision to disaggregate the “Asian or Pacific Islander” category was a step toward providing these communities with information to better serve their needs (Srinivasan & Guillermo, 2000). It is not clear, however, if the individuals who formerly made up the OMB’s “Asian or Pacific Islander” combined group now categorize themselves as the current guidelines intend. While the guidelines do not tell individuals who they are or how to classify themselves, the Office of Management and Budget defines “Asian” as “[a] person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent including Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam” (Office of Management and Budget, 1997). Thus, the term “Asian” blurs national origin distinctions, combining vast cultural and geographic diversity under a single rubric. Although this pan-ethnic terminology emerged in the US, this terminology does not exist in Asian countries and does not correspond to an Asian identification in any other country (Edmonston, Goldstein, & Tamayo Lott, 1996). There is no evidence that people of Pakistani, Tamil, Japanese, Vietnamese, or Filipino descent perceive themselves as belonging to a single Asian group (Laws & Heckscher, 2002).

Using meaningful racial/ethnic classifications is increasingly important due to growing US diversity and an accompanying interest in health and socioeconomic disparities (Day, 1996; Li & DiGianni, 2003; Smedley, Stith, & Nelson, 2003). Several studies have documented accuracy problems with race/ethnicity data. For example, estimates of racial/ethnic differences in health status vary depending on data collection methods (Blustein, 1994; Boehmer et al., 2002; Hahn, Mulinare, & Teutsch, 1992; Kressin, Chang, Hendricks, & Kazis, 2003; Pan, Glynn, Mogun, Choodnovskiy, & Avorn, 1999). Therefore, researchers are urged to use techniques and tools that provide the most reliable and useful results (Stewart & Napoles-Springer, 2003).

The importance of gathering accurate data for the Asian population parallels the population’s growth. Asians are projected to be the fastest-growing major US population category over the next half century. Census projections show the Asian population could grow by 213% and that Asians’ share of the nation’s population is projected to double, and stand at 8% by 2050 (Armas, 2004; U.S. Census Bureau, 2004). In this time of increasing US heterogeneity, environments where multiple Asian subgroups are represented in high numbers, like the state of Hawaii, may provide clues to America’s increasingly diverse future. Hawaii represents a unique setting for examining differences among populations because: (i) there is no clear ethnic majority, as no group constitutes more than a third of the state’s population; (ii) 21.4% of individuals responding to the 2000 census marked two or more races, in contrast with 2.4% in the US as a whole (U.S. Census Bureau, 2000); (iii) 58% marked an Asian category, and 23% indicated a Pacific Islander category, whether alone or in combination with other categories, in contrast with 4.2% and .3% of respondents in the US as a whole (U.S. Census Bureau, 2001, 2002); (iv) of the individuals who marked only one race on the 2000 census, 41.6% marked a category that would be collapsed into “Asian”, in contrast to 3.6% in the US as a whole (U.S. Census Bureau, 2000). This research, performed in Hawaii, provides an opportunity to examine how Asian populations categorize themselves and illuminates the reliability and utility of the broad “Asian” category.

A related question is whether using closed-coded, broad race/ethnicity categories like “Asian” discourages participation. That is, if race/ethnicity measures do not specifically name communities of interest, like “Chinese,” potential Chinese participants may not believe the research is interested in their population.

This report examines Asian subgroup responses to the revised OMB-minimum measure for race/ethnicity compared to a more expanded measure. We also report common Asian mixed heritage responses, and how these mixed individuals assign their primary identity. Finally, we test whether including more detailed race/ethnicity categories increases participation from groups within these categories.

2. METHODS

This study was conducted at Kaiser Permanente Hawaii (KPHI) as an ancillary effort to the multi-site Hemochromatosis and Iron Overload Screening (HEIRS) Study (McLaren et al., 2003). KPHI members participated in the HEIRS Study as part of the Kaiser Permanente field center. The HEIRS Study is designed to study the: prevalence; penetrance; genetic and environmental determinants and modifiers; and potential clinical, personal, and societal impact of iron overload and hereditary hemochromatosis (McLaren et al., 2003). Primary care patients 25 years and older were recruited to have blood drawn and tested for iron markers and genotypes indicating susceptibility to hereditary hemochromatosis. Participants also completed an initial screening form that included items on socio-demographics, including race/ethnicity.

Recruitment at the Hawaii site used two approaches. The first involved approaching patients in KPHI clinic lobbies on the island of Oahu, describing the study, and obtaining their consent. These participants underwent a blood draw and completed an initial screening form at that visit. The second strategy used existing administrative databases to identify primary care patients at KPHI clinics on Oahu and Maui who were mailed a cover letter and brochure, with detachable postcard to indicate interest in the study, or decline participation. Interested potential participants received enrollment materials, including consent forms and a questionnaire. Their blood was drawn upon a return visit to a KPHI clinic.

The initial screening form collected race/ethnicity information with the OMB-minimum standard measure, as ethnicity (Spanish, Latino, or Hispanic vs. not Spanish, Latino, or Hispanic), and race (American Indian or Alaska Native (AI/AN); Asian; Black or African-American; Native Hawaiian or other Pacific Islander (NHOPI); White or Caucasian). After starting recruitment for the HEIRS study KPHI received approval from the HEIRS study Steering Committee and Observational Study Monitoring Board, and the KPHI Institutional Review Board to include an expanded race and ethnicity measure. This race and ethnicity measure was on a yellow half sheet of paper attached to the enrollment packet for a subgroup of participants.

Recruitment took place between February 2001 and February 2003. A total of 6,729 KPHI members participated in the HEIRS study, all of whom answered the initial screening form that included the OMB-minimum race/ethnicity question. A sub-group of 2,285 participants also completed the additional, expanded race/ethnicity measure. This sub-group comprises this report’s sample.

2.1 Measures

2.1.1 OMB-minimum race and ethnicity measure

The HEIRS study race/ethnicity measure is detailed in Table 1. This measure allowed for checking multiple categories and matches the minimum standard for maintaining, collecting, and presenting data on race and ethnicity for all federal reporting purposes according to the OMB (Office of Management and Budget, 1997). Further definition of terms, e.g., the OMB definition of “Asian”, was not provided in any study materials. The OMB clarified, in the 1997 notice announcing the revisions, that its decisions on the minimum race and ethnicity classifications “permit the collection of more detailed information on population groups provided that any additional categories can be aggregated into the minimum standard set of categories” (Office of Management and Budget, 1997).

Table 1.

Race and ethnicity measures

graphic file with name nihms30410f1.jpg

Open in a new tab

Next, if you checked more than one category, please circle above the one group you feel best represents your race.

2.1.2 NHIS race and ethnicity measure

The additional expanded race/ethnicity measure used only at the KPHI study site is also presented in Table 1. We chose a measure used in the National Health Interview Survey (NHIS) (National Center for Health Statistics, 2000) that has expanded terms for Asians and Pacific Islanders and has been recommended for its comprehensiveness by Lin and Kelsey (2000). Like the OMB-minimum measure, the NHIS measure allows for checking multiple categories. An NHIS follow-up question asks participants marking multiple categories to indicate the category they feel best represents their race. As the OMB-minimum standard measure uses the term “race,” our choice of the NHIS variable, which also includes the term “race” in its question format, allows us to easily compare the outcomes using the expanded NHIS variable to those found using the OMB-minimum measure.

2.2 Analysis

Pearson’s chi-square analyses, with Bonferroni correction, compared individuals who marked the pan-ethnic category “Asian” on the OMB measure to those who marked a response on the NHIS measure contained within the OMB definition of “Asian” (i.e., those marking Filipino, Korean, Vietnamese, Japanese, Asian Indian, Chinese, Other Asian). Comparisons were made between all “single heritage” Asians (those who marked only one NHIS Asian subcategory) and all “mixed heritage” Asians (those who marked more than one NHIS subcategory, where at least one was an Asian subgroup). The three groups receiving the largest proportion of NHIS responses, Filipino, Japanese, and/or Chinese, were similarly divided and three Pearson’s chi-square tests assessed differences between single and mixed heritage respondents from each group. To evaluate the relative functioning of the OMB-minimum “Asian” category to other OMB-minimum categories, Pearson’s chi-square tests compared those marking the NHIS category “Native Hawaiian” to those who marked the “Native Hawaiian or other Pacific Islander” (NHOPI) on the OMB measure.

Using descriptive statistics we analyzed NHIS responses, focusing on those who reported at least one Asian subgroup. A response that included two or more “Asian” groups, e.g., an individual marking Japanese and Chinese, was considered a multiple/mixed race response. This differs from some tabulation methods for a dual “within-(broad) Asian category” response (Office of Management and Budget, 2000). Individuals checking more than one category were instructed to “please circle above the one group you feel best represents your race” in order to assess respondents’ primary identification.

As a separate aim, Pearson chi-square tested if more detailed race/ethnicity categories would increase participation by Asians and Pacific Islanders, comparing participation rates for individuals who received only the OMB-minimum standard race/ethnicity measure to those who also received the NHIS measure in their packet. As a related analysis, Pearson chi-square tested the proportion of Asians and Pacific Islanders among respondents to the two types of materials. In order to reduce any potential biases associated with temporal differences in response rates, individuals recruited prior to using the NHIS measure (and hence initiation of the use of two types of study packets) were excluded from these analyses.

3. RESULTS

3.1 Marking “Asian”

Of 2,285 individuals surveyed with the NHIS measure at the KPHI site, 1,029 marked an “Asian” category according to the OMB definition (Office of Management and Budget, 1997) (i.e., those marking Filipino, Korean, Vietnamese, Japanese, Asian Indian, Chinese, Other Asian). Of these 17% did not mark “Asian” on the OMB-minimum measure. Conversely, of the respondents who marked “Asian” on the OMB-minimum measure, <1% did not mark an Asian subcategory on the NHIS measure (Table 2).

Table 2.

Comparison of responses to NHIS and OMB-minimum measures

Marked “Asian” on OMB-minimum measure Did not mark “Asian” on OMB-minimum measure
Marked ≥ 1 NHIS Asian categorya, % (n) 82.7 (851) 17.3 (178) 1,029
Did not mark an NHIS Asian categorya, % (n) .5 (6) 99.5 (1,250) 1,256
Total = 2,285
Open in a new tab
a

As defined by the OMB, i.e., Filipino, Korean, Vietnamese, Japanese, Asian Indian, Chinese, Other Asian

Table 3 provides greater detail on NHIS-measure versus OMB-minimum measure responses. We compared a combined group of “single heritage” Asians (n=694) to a combined group of “mixed heritage” Asians (n=335). In the combined group comparison (χ2=279.51, df=1, p<.001), and in each of the three ethnically-specific comparisons, mixed individuals were less likely to mark “Asian” on the OMB measure than “single heritage” individuals (Chinese χ2=89.89, df=1, p<.001; Filipino χ2=44.37, df=1, p<.001; Japanese χ2=74.96, df=1, p<.001).

Table 3.

OMB-minimum responses for subgroups from the NHIS measure

Responses to NHIS measure: Asiana on OMB measure Responsesa (#) to OMB measure among those who did not mark Asian:
Hispanic AI/AN Black NHOPI White No answer
Asianb,c N # %
 single heritage 694 669 96.4 2 0 0 9 3 13
 mixed heritage 335 182 54.3 22 3 2 110 63 1
Three largest Asian categories:
Filipinoc
 single heritage 125 111 88.8 2 0 0 6 1 7
 mixed heritage 98 47 48.0 11 3 1 40 16 0
Japanesec
 single heritage 389 380 97.7 0 0 0 3 0 6
 mixed heritage 109 79 72.5 1 0 1 17 12 0
Chinesec
 single heritage 129 129 100.0 0 0 0 0 0 0
NHOPIa on OMB measure Responsesa (#) to OMB measure among those who did not mark NHOPI:
Hispanic AI/AN Black Asian White No answer
 mixed heritage 191 97 50.8 12 0 0 75 38 1
Native Hawaiianc N # %
 single heritage 85 85 100.0 0 0 0 0 0 0
 mixed heritage 267 233 87.3 2 1 0 13 21 1
  all 352 267 90.3 2 1 0 13 21 1
Open in a new tab
a

Marked alone or along with other OMB categories. Categories within both of the OMB and NHIS measures are non-mutually exclusive, i.e., participants could mark multiple categories on each measure.

b

Includes all individuals who marked a category on the NHIS measure that would be included in the OMB definition for “Asian”

c

X2, p<.001, mixed vs. single heritage responders to OMB “Asian” category

We did not collect supplementary information that could explain the response pattern for those who marked a single Asian heritage on the NHIS measure but did not mark “Asian” on the OMB-minimum measure (14 Filipino and the nine Japanese respondents). However, for the mixed heritage Asian responders on the NHIS measure who did not mark “Asian” on the OMB-minimum measure, we examined their other response(s) on the NHIS measure: 1) Mixed heritage Filipinos who did not mark “Asian” on the OMB measure most often marked “Native Hawaiian or other Pacific Islander” (NHOPI) on the OMB-minimum measure. Of the 40 who did not mark “Asian” and did mark NHOPI on the OMB measure, 35 marked either “Native Hawaiian” or another Pacific Islander category on the NHIS measure; 2) Mixed heritage Japanese who did not mark “Asian” on the OMB measure most often marked NHOPI. Of the 17 who did not mark “Asian” and did mark NHOPI, 17 had marked “Native Hawaiian” on the NHIS measure; 3) Mixed heritage Chinese who did not mark “Asian” on the OMB measure most often marked the NHOPI category. Of the 75 Chinese who did not mark “Asian,” and did mark NHOPI, 74 had marked “Native Hawaiian” or another Pacific Islander category on the NHIS measure.

We undertook the same comparison between the NHIS category “Native Hawaiian” and the OMB-minimum measure “Native Hawaiian and Other Pacific Islander” (NHOPI) category. We were interested in whether including the more specific “Native Hawaiian” in the OMB-minimum measure would affect response patterns. Although mixed individuals were less likely to mark “Native Hawaiian or other Pacific Islander” on the OMB measure than “single heritage” individuals (χ2=11.98, df=1, p<.001), among the mixed heritage group, 87% of those who marked Native Hawaiian on the NHIS also marked the corresponding OMB-minimum NHOPI category (Table 3). This is a greater proportion than the proportions of mixed Asian subgroups who marked the “Asian” OMB category.

Table 4 details primary identity for those who reported more than one NHIS race/ethnicity category and reported at least one Asian subgroup. The three most common categories are detailed. Twenty-one percent of these mixed individuals did not select a primary identification when asked to do so.

Table 4.

Primary identification of mixed heritage individuals with ≥ 1 Asian subgroupa ascertained from NHIS responses

Primary identity
Three most common mixes: Total Native Hawaiian Japanese Filipino Chinese White Don’t Know No Answer
Chinese, Native Hawaiian, White 57 25 (43.9%) 0 (0.0%) 0 (0.0%) 5 (8.8%) 16 (28.1%) 1 (1.8%) 10 (17.5%)
Japanese, White 35 0 (0.0%) 14 (40.0%) 0 (0.0%) 0 (0.0%) 8 (22.9%) 3 (8.6%) 10 (28.6%)
Chinese, Native Hawaiian 30 19 (63.3%) 0 (0.0%) 0 (0.0%) 5 (16.7%) 0 (0.0%) 0 (0.0%) 6 (20.0%)
Total 122 44 (36.1%) 14 (11.5%) 0 (0.0%) 10 (8.2%) 24 (19.7) 4 (3.3%) 26 (21.3)
Open in a new tab
a

Using OMB definition of “Asian” i.e., those marking Filipino, Korean, Vietnamese, Japanese, Asian Indian, Chinese, Other Asian.

3.2 Response rate results

Fifty-six percent of those who received both the NHIS measure and the OMB-minimum race/ethnicity measure returned their study materials (798/1430). Of these, 87.8% completed their blood draw for the HEIRS study (701/798). Fifty-five percent of those who only received the OMB measure returned their survey (713/1293). Of these, 88.2% had their blood drawn (629/713). Of those who received both measures, Asians represented 37.7% of the sample and NHOPI represented 14.7% of the sample. Of those who received only the OMB measure, Asians represented 39.6% of the sample and NHOPI represented 16.2% of the sample. Differences between those receiving and not receiving the expanded instrument were not statistically significant.

4. DISCUSSION

Researchers should strive to improve how race/ethnicity is captured so results are more meaningful (Lin & Kelsey, 2000; Mays, Ponce, Washington, & Cochran, 2003). In the past, researchers often used the category “Asian and Pacific Islander” (Meredith & Siu, 1995). This “lumping groups together” drew criticism from various disciplines who called for greater specificity (Bhopal, 1997; Centers for Disease Control, 1993). While the OMB’s decision to separate this category is a step in the right direction, our results indicate that specification beyond, or providing definitions when using the minimum OMB categories is prudent, at least on self-administered questionnaires with Asian populations.

While it is not known why 11% of the single heritage Filipino respondents did not mark “Asian” on the OMB measure, it would seem that participants did not identify with or accept this terminology. Filipinos have been known to categorize themselves as Spanish (Mays et al., 2003), Pacific Islander, Asian American, or, if mixed, White (Yu & Liu, 1992). Filipinos in the current study followed a similar pattern, as the OMB options for race/ethnicity capture data for this group in multiple categories. The Philippines consist of over 7,000 islands set in the western Pacific Ocean, so it is not surprising that many in this population would mark a category that includes “Other Pacific Islander”, particularly if not provided with a definition stating that the category “Asian” includes individuals with ancestry in the Philippine islands.

Most mixed heritage Filipinos, Chinese, and Japanese who did not mark “Asian” on the OMB-measure marked NHOPI. Most often these respondents had marked “Native Hawaiian” or another Pacific Islander category on the NHIS measure, which partially explains the OMB-minimum response. This doesn’t explain, however, why these individuals did not also mark “Asian” on the OMB-measure, as checking multiple categories was allowed. And, as this mixed heritage group had marked multiple categories on the NHIS measure, presumably these respondents were not averse to marking more than one category. Rather, the discrepancy may result from more specific terminology (e.g., “Filipino”, “Japanese”, “Chinese”) garnering responses more readily than the term “Asian.” “Asian” is not an identity that is readily embraced. When Lien, Conway and Wong (2003) provided Asian Americans multiple self-identification options — American, Asian American, Asian, ethnic American (e.g., Chinese American), or simply ethnic origin (e.g., Chinese) — 34% identified themselves as “ethnic American” and 30% by ethnic origin alone. Interestingly, 85% of the sample did not choose the more generic “Asian American” category. A follow-up question queried whether they ever think of themselves as “Asian American.” Only about half reported such a pan-ethnic consciousness. Least popular among the choices was “Asian”, 4% on average, with no group having more than 7% of its respondents preferring this label. This “Asian” term is among the OMB-minimum categories, which many of our respondents did not mark. Our findings support other research documenting the complications when race/ethnicity is assessed among mixed populations (Hahn, Truman, & Barker, 1995). In open-ended questioning, monoethnic responses, often with clear preference patterns, are common among mixed heritage individuals (Phinney & Alipuria, 1996). Such underreporting among mixed heritage populations may be important because of the relationship of race/ethnicity to health, including biologic risk factors (Bacha, Saad, Gungor, Janosky, & Arslanian, 2003; Jones et al., 2004; Lovejoy, de la Bretonne, Klemperer, & Tulley, 1996; Osei & Schuster, 1996; Rodgers & Lewandowski, 2002), health behaviors (Chung, Tash, Raymond, Yasunobu, & Lew, 1990; Hiatt et al., 1996), health perceptions (Greenwald, 1991), and mortality (American Cancer Society, 2004).

Supporting our contention that specific terminology more readily garners response, relative to the Asian mixed heritage groups, a greater proportion of mixed Native Hawaiians (per NHIS response) marked the corresponding NHOPI OMB-minimum category. In this case, “Native Hawaiian” is the terminology of the broad OMB category.

Our hypothesis that individuals with a mixed heritage including Native Hawaiian would tend to choose Native Hawaiian as their primary identification was supported. This response pattern reflects both cultural origin and the political and social benefits tied to indigenous heritage in the state of Hawaii and demonstrates how race/ethnicity may be differentially reported when it affects how individuals are viewed and treated (Olivares Doan & Stephan, 2006; Sondik et al., 2000). Future research on how mixed individuals’ primary identity correlates with data from the group with which they primarily identify is of interest. If a mixed heritage individual was raised in one culture with which they primarily identify, they may share similar beliefs, behaviors, and opportunities as the group, such that knowing the primary identity could provide useful information applicable to outcomes such as health and socioeconomic status. This may be only partly effective, however, as approximately one-fifth of the mixed heritage responders in this report did not mark a primary identity.

Contrary to our hypothesis, using the expanded race/ethnicity measure did not result in higher participation rates. As the NHIS measure was ancillary to the existing HEIRS study, we were not able to simply substitute the NHIS measure for the OMB-minimum categories. Therefore, those who received the NHIS measure were receiving an additional half-page survey instrument. The advantages of a more sensitive demographic instrument were perhaps muted by the design disadvantage of using a longer questionnaire; short questionnaires have been correlated with higher response rates (Edwards et al., 2002). We suggest future research use a randomized design substituting one measure for another, which would eliminate confounding from the study material’s length. This may also eliminate the potential for participants to skip either the OMB-minimum or the expanded measure, as these items may seem redundant to participants. Although we did not track whether such skipping occurred in the current research, this potential problem would be eliminated through substitution with an expanded measure, which is allowed under the OMB guidelines.

For the disaggregation of the “Asian or Pacific Islander” category to represent a step in providing communities with better information (Srinivasan & Guillermo, 2000), the communities to whom these categories apply must understand them. One assumption behind any valid statistic using race/ethnicity is that the categories and designations are understood by the populations questioned (Centers for Disease Control, 1993; Lin & Kelsey, 2000). If the perceptions of those completing a questionnaire differ from those analyzing the data, incorrect conclusions are a likely consequence, in this case through missed information and erroneous generalization. The revised OMB standards specifically state that they do not tell individuals who they are or how to classify themselves. There are no “wrong answers” (Office of Management and Budget, 2000). That said, the OMB General Guidelines suggest that when the minimum categories are used, to “[p]rovide definitions to the minimum race categories when possible….especially for respondents unsure of the definitions of Asian, American Indian, Native Hawaiian, and Other Pacific Islander…” and “[f]or self-administered forms, providing the definition of the category should be considered if space and formatting limitations can be overcome” (Office of Management and Budget, 2000). Our results support providing definitions when using minimum categories. For example, the OMB-issued document “Provisional Guidance on the Implementation of the 1997 Standards for Federal Data on Race and Ethnicity” (2000) provides examples of use of the OMB-minimum measure where the categories are accompanied by definitions (Example #27 on pg. 24) (Office of Management and Budget, 2000). Alternatively, a more expanded race/ethnicity measure, similar to the NHIS measure used in this report, could used and be aggregated into the OMB-minimum standard, as allowed under federal regulations.

Acknowledgments

This study was supported by grant R03 HG2763-01A1, contracts N01-HC-05189 and N01-HC-05192, and by the Kaiser Permanente Center for Health Research. The authors are grateful for assistance from Aleli Vinoya and Lanabell Kalama in carrying out this research, and for editing and administrative support from Kevin Lutz and Jeanette Murray. We also thank HEIRS Study participants for kindly contributing to this research effort.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. American Cancer Society. Hawaii Cancer Facts & Figures 2003–2004. American Cancer Society; Honolulu: 2004. [Google Scholar]
  2. Armas GC. Asian Population Surging Across America. Falls Church, VA: Associated Press; 2004. [Google Scholar]
  3. Bacha F, Saad R, Gungor N, Janosky J, Arslanian SA. Obesity, regional fat distribution, and syndrome X in obese black versus white adolescents: race differential in diabetogenic and atherogenic risk factors. Journal of Clinical Endocrinology & Metabolism. 2003;88(6):2534–2540. doi: 10.1210/jc.2002-021267. [DOI] [PubMed] [Google Scholar]
  4. Bhopal RS. Is Research into ethnicity and health racist, unsound, or important science? British Medical Journal. 1997;314(7096):1751–1756. doi: 10.1136/bmj.314.7096.1751. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Blustein J. The Reliability of Racial Classifications in Hospital Discharge Abstract Data. American Journal of Public Health. 1994;84(6):1018–1021. doi: 10.2105/ajph.84.6.1018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boehmer U, Kressin NR, Berlowitz DR, Christiansen CL, Kazis LE, Jones JA. Self-Reported vs Administrative Race/Ethnicity Data and Study Results. American Journal of Public Health. 2002;92(9):1471–1473. doi: 10.2105/ajph.92.9.1471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Centers for Disease Control. Use of Race and Ethnicity in Public Health Surveillance (Report RR-10) Atlanta, GA: US Department of Health and Human Services; 1993. [Google Scholar]
  8. Chung CS, Tash E, Raymond J, Yasunobu C, Lew R. Health risk behaviours and ethnicity in Hawaii. International Journal of Epidemiology. 1990;19:1011–1018. doi: 10.1093/ije/19.4.1011. [DOI] [PubMed] [Google Scholar]
  9. Day JC. Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050 (Current Population Reports (P25–1130) Washington, DC: US Govt. Printing Office; 1996. [Google Scholar]
  10. Edmonston B, Goldstein J, Tamayo Lott J. Spotlight on Heterogeneity: The Federal Standards for Racial and Ethnic Classification, Summary of a Workshop. Washington D.C: National Academy Press; 1996. [Google Scholar]
  11. Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R, Kwan I. Increasing response rates to postal questionnaires: systematic review. British Medical Journal. 2002;324:1183–1192. doi: 10.1136/bmj.324.7347.1183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Friedman DJ, Cohen BB, Averbach AR, Norton JM. Race/Ethnicity and OMB Directive 15: Implications for State Public Health Practice. American Journal of Public Health. 2000;90(11):1714–1719. doi: 10.2105/ajph.90.11.1714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Greenwald HP. Interethnic differences in pain perception. Pain. 1991;44:157–163. doi: 10.1016/0304-3959(91)90130-P. [DOI] [PubMed] [Google Scholar]
  14. Hahn RA, Mulinare J, Teutsch SM. Inconsistencies in Coding of Race and Ethnicity Between Birth and Death in US Infants. Journal of the American Medical Association. 1992;267(2):259–263. [PubMed] [Google Scholar]
  15. Hahn RA, Truman BI, Barker ND. Identifying Ancestry: The Reliability of Ancestral Identification in the United States by Self, Proxy, Interviewer, and Funeral Director. Epidemiology. 1995;7(1):75–80. doi: 10.1097/00001648-199601000-00013. [DOI] [PubMed] [Google Scholar]
  16. Hiatt RA, Pasick RJ, Perez-Stable EJ, McPhee SJ, Engelstad L, Lee M, Sabogal F, D’Onofrio CN, Stewart S. Pathways to early cancer detection in the multiethnic population of the San Francisco Bay Area. Health Education Quarterly. 1996;23:S10–S27. [Google Scholar]
  17. Jones BA, Kasl SV, Howe CL, Lachman M, Dubrow R, McCrea Curnen M, Soler-Vila H, Beeghly A, Duan F, Owens P. African American/White differences in breast carcinoma: p53 Alterations and other tumor characteristics. Cancer. 2004;101 doi: 10.1002/cncr.20500. August 9 (online) [DOI] [PubMed] [Google Scholar]
  18. Kressin NR, Chang BH, Hendricks A, Kazis LE. Agreement between administrative data and patients’ self-reports of race/ethnicity. American Journal of Public Health. 2003;93(10):1734–1739. doi: 10.2105/ajph.93.10.1734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Laws MB, Heckscher RA. Racial and Ethnic Identification Practices in Public Health Data Systems in New England. Public Health Reports. 2002;117:50–61. doi: 10.1016/S0033-3549(04)50108-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Li FP, DiGianni LM. Reducing the unequal burden of cancer. Cancer Epidemiology, Biomarkers and Prevention. 2003;12(Suppl):230s–231s. [PubMed] [Google Scholar]
  21. Lien P-t, Conway MM, Wong J. The Contours and Sources of Ethnic Identity Choices Among Asian Americans. Social Science Quarterly. 2003;84(2):461–481. [Google Scholar]
  22. Lin SS, Kelsey JL. Use of Race and Ethnicity in Epidemiologic Research: Concepts, Methodological Issues, and Suggestions for Research. Epidemiologic Reviews. 2000;22(2):187–202. doi: 10.1093/oxfordjournals.epirev.a018032. [DOI] [PubMed] [Google Scholar]
  23. Lovejoy JC, de la Bretonne JA, Klemperer M, Tulley R. Abdominal fat distribution and metabolic risk factors: effects of race. Metabolism: Clinical & Experimental. 1996;45(9):1119–1124. doi: 10.1016/s0026-0495(96)90011-6. [DOI] [PubMed] [Google Scholar]
  24. Mays VM, Ponce NA, Washington DL, Cochran SD. Classification of Race and Ethnicity: Implications for Public Health. Annual Review of Public Health. 2003;24:83–110. doi: 10.1146/annurev.publhealth.24.100901.140927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. McLaren CE, Barton JC, Adams PC, Harris EL, Acton RT, Press N, Reboussin DM, McLaren GD, Sholinsky P, Walker AP, Gordeuk VR, Leiendecker-Foster C, Dawkins FW, Eckfeldt JH, Mellen BG, Speechley M, Thomson E, Hemochromatosis and Iron, Overload Study Research Investigators Hemochromatosis and Iron Overload Screening (HEIRS) Study Design for an Evaluation of 100,000 Primary Care-Based Adults. American Journal of Medical Science. 2003;325(2):53–62. doi: 10.1097/00000441-200302000-00001. [DOI] [PubMed] [Google Scholar]
  26. Meredith LS, Siu AL. Variation and Quality of Self-Report Health Data: Asians and Pacific Islanders Compared with Other Ethnic Groups. Medical Care. 1995;33(11):1120–1131. doi: 10.1097/00005650-199511000-00005. [DOI] [PubMed] [Google Scholar]
  27. National Center for Health Statistics. Household Composition Instrument. National Health Interview Survey: U.S. Dept. of Health and Human Services, Centers for Disease Control and Prevention; 2000. [Google Scholar]
  28. Office of Management and Budget. Revisions to the Standards for the Classifications of Federal Data on Race and Ethnicity. Washington, DC: Office of Management and Budget; 1997. [Google Scholar]
  29. Office of Management and Budget. Provisional Guidance on the Implementation of the 1997 Standards for Federal Data on Race and Ethnicity. Washington, DC: Office of Management and Budget; 2000. [Google Scholar]
  30. Olivares Doan G, Stephan CW. The functions of ethnic identity: A New Mexico Hispanic example. International Journal of Intercultural Relations. 2006;30(2):229–241. [Google Scholar]
  31. Osei K, Schuster DP. Effects of race and ethnicity on insulin sensitivity, blood pressure, and heart rate in three ethnic populations: comparative studies in African-Americans, African immigrants (Ghanaians), and white Americans using ambulatory blood pressure monitoring. American Journal of Hypertension. 1996;9(12 Pt 1):1157–1164. doi: 10.1016/S0895-7061(96)00248-8. [DOI] [PubMed] [Google Scholar]
  32. Pan CX, Glynn RJ, Mogun H, Choodnovskiy I, Avorn J. Definition of Race and Ethnicity in Older People in Medicare and Medicaid. Journal of the American Geriatrics Society. 1999;47(6):730–733. doi: 10.1111/j.1532-5415.1999.tb01599.x. [DOI] [PubMed] [Google Scholar]
  33. Parker JD, Makuc DM. Methodologic Implications of Allocating Multiple-Race Data to Single-Race Categories. Health Services Research. 2002;37(1):203–215. [PubMed] [Google Scholar]
  34. Phinney JS, Alipuria LL. At the Interface of Cultures: Multiethnic/Multiracial High School and College Students. The Journal of Social Psychology. 1996;136(2):139–158. doi: 10.1080/00224545.1996.9713988. [DOI] [PubMed] [Google Scholar]
  35. Rodgers AL, Lewandowski S. Effects of 5 Different Diets On Urinary Risk Factors For Calcium Oxalate Kidney Stone Formation: Evidence of Different Renal Handling Mechanisms in Different Race Groups. Clinical Urology. 2002;168(3):931–936. doi: 10.1016/S0022-5347(05)64545-4. [DOI] [PubMed] [Google Scholar]
  36. Smedley BD, Stith AY, Nelson AR. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine, The National Academies Press; 2003. [PubMed] [Google Scholar]
  37. Sondik EJ, Lucas JW, Madans JH, Smith SS. Race/Ethnicity and the 2000 Census: Implications for Public Health. American Journal of Public Health. 2000;90(11):1709–1713. doi: 10.2105/ajph.90.11.1709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Srinivasan S, Guillermo T. Toward improved health: Disaggregating Asian American and Native Hawaiian/Pacific Islander data. American Journal of Public Health. 2000;90(11):1731–1734. doi: 10.2105/ajph.90.11.1731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Stewart AL, Napoles-Springer AM. Advancing Health Disparities Research: Can We Afford to Ignore Measurement Issues? Medical Care. 2003;41(11):1207–1220. doi: 10.1097/01.MLR.0000093420.27745.48. [DOI] [PubMed] [Google Scholar]
  40. U.S. Census Bureau. Profile of General Demographic Characteristics 2000, Summary File 1 (SF 1), Geographical Area: Hawaii. Honolulu: U.S. Census Bureau; 2000. [Google Scholar]
  41. U.S. Census Bureau. The Native Hawaiian and Other Pacific Islander Population: 2000 (Census Brief, C2KBR/01-14) U.S. Census Bureau; 2001. [Google Scholar]
  42. U.S. Census Bureau. The Asian Population: 2000 (Census Brief, C2KBR/01-16) U.S. Census Bureau, U.S. Department of Commerce; 2002. [Google Scholar]
  43. U.S. Census Bureau. Census Bureau Projects Tripling of Hispanic and Asian Populations in 50 Years: Non-Hispanic Whites May Drop to Half of Total Population (CB04-44) U.S. Census Bureau, Public Information Office; 2004. [Google Scholar]
  44. Wallman KK, Evinger S, Schechter S. Measuring our nation’s diversity: developing a common language for data on race/ethnicity. American Journal of Public Health. 2000;90(11):1704–1708. doi: 10.2105/ajph.90.11.1704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Yu ESH, Liu WT. US National Health Data on Asian Americans and Pacific Islanders: A Research Agenda for the 1990s. American Journal of Public Health. 1992;82(12):1645–1652. doi: 10.2105/ajph.82.12.1645. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES