Race and Ethnicity

Below we provide two measures to assess race/ethnicity.

Option 1: Bespoke Social Science Measure

Background

Option 1 is a bespoke Social Science Measure that was developed by BSS faculty to assess race based on reviews of different recommendations from published literature and professional and community organizations. This measure acknowledges that traditional race and ethnicity categories (e.g., White, Black, Asian, Pacific Islander, Native American) are not aligned with the ways in which many people identify. In fact, some scholars argue that, in practice, “ethnicity cannot meaningfully be disentangled from ‘race’ in societies with inequitable race relations, hence the construct race/ethnicity” (Krieger, 2001). As such, the traditional racial categories have been expanded to include response options for Central American/South American, Middle Eastern/North African, and South Asian. One can also choose to ask more information in an open text field about any identity checked (for example, asking about tribal nation or descent or asking participants to specify Asian descent/ethnicity as is done in the U.S. Census).

Since many Hispanic/Latinx people do not identify with traditional or even the expanded racial categories, Hispanic/Latinx ethnicity is asked first to provide Hispanic/Latinx people with an option to self-identify before being asked to categorize their race. This should minimize missing race data, though many Hispanic/Latinx people may still list “Another race.”

Source: Krieger, N. (2001). A glossary for social epidemiology. Journal of Epidemiology & Community Health, 55(10), 693-700.

Pros and Cons

Pros: This measure more inclusively reflects the racial identities of diverse groups and acknowledges the difficulty in distinguishing race and ethnicity. The measure is also slightly shorter than the U.S. Census measure (Option 2).

Cons: Data from this measure is not as easily categorized according to the NIH racial categories and may result in more participants being coded as an “unknown” race than the U.S. Census Measure. Also, this measure was recently adapted and reviewers may have some pushback against these more nuanced and inclusive racial categories.

Measure

Do you consider yourself Hispanic, Latina/e/o, Latinx?

1. Yes

2. No

3. Decline to answer

What is your race/descent? (Select all that apply)

1. American Indian/Native American or Alaskan Native (IF YES: What is your tribal nation or descent?)

2. Asian

3. Black, African American or of African descent

4. Central American/South American

5. Caribbean

6. Middle Eastern/North African

7. Pacific Islander

8. Native Hawaiian

9. South Asian (Indian subcontinent)

10. White or of European descent

11. Another race, please specify: ____________

12. Prefer not to answer

Analysis

a. Non-NIH Reporting

We recommend reporting all racial categories in a paper’s demographics table. Hispanic/Latinx ethnicity can either be reported as a separate variable in the table or you can report the frequency of each race by Hispanic/Latinx ethnicity. If categories need to be collapsed for modeling, we recommend listing the collapsed category and the subcategories underneath or in the table footnotes. For example, in a sample of 100 participants of which 14 identify as an Asian or South Asian and 20 people identify as Native Hawaiian and Pacific Islander, we would report the total N for Asian and total N for Native Hawaiian or Pacific Islander and the n and % for the racial subcategories that comprise the umbrella racial categories as follows:

b. NIH Reporting

Coding Race: Current NIH reporting guidelines require that we categorize people according to 5 racial categories: American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; and White. NIH also provides options for Multiracial and Unknown Race. These racial categories are then stratified by Hispanic/Latinx and non-Hispanic/Latinx. We added 4 additional racial categories: Central American/South American; Caribbean; Middle Eastern/North African; South Asian (Indian subcontinent). Thus, in order to report race/ethnicity to NIH, we need to collapse our categories. This can be done as follows:

• Combine Asian and South Asian as one category

• Combine Native Hawaiian and Pacific Islander as one category

• The responses for the “another race” write-in option should be examined and individual responses should be recoded to align with one of the 5 NIH racial categories (American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; White) first, one of the newly added categories (Central American/South American; Caribbean; Middle Eastern/North African) second, or unknown race third.

• If an individual ONLY selects one of the 5 NIH race categories (American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; White), they are coded as such.

• If an individual selects two or more of the 5 NIH race categories (American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; White), they are coded as multiracial.

• If an individual selects one of the 5 NIH race categories (American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; White) AND selects one or more of our newly added categories (Central American/South American; Caribbean; Middle Eastern/North African), they should only be coded according to 5 NIH categories, they should not be coded as multiracial.

• If a participant does not select one of the 5 NIH categories (American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; White) but selects one or more of the new categories (Central American/South American; Caribbean; Middle Eastern/North African) or they write in a race that does not fit within the 5 NIH categories, they should be coded as Unknown.

• The above categorization should result in the following categories:

1. American Indian or Alaskan Native

2. Black or African American

3. Asian (include South Asian)

4. Native Hawaiian or Pacific Islander

5. White

6. Multiracial (checked off two or more of these first 5 racial categories)

7. Unknown (did not check off any categories or only checked identities that are not represented in categories 1-5.) NIH does not have a category of “Other Racial Identity” at this time.

Example 1a & 1b: If a person selects Caribbean and White, they should be coded as White according to the NIH categories. If a participant selects Caribbean, White, and Asian, they should be coded as Multiracial according to the NIH categories.

Example 2a & 2b: If a person selects Middle Eastern and Black, they should be coded as Black according to the NIH categories. If a person selects Middle Eastern and does not select another racial category, they should be coded as “Unknown” race according to the NIH categories.

Producing Final Hispanic/Latinx Ethnicity by Race Table: After race is coded as described above, the race categories should be stratified by Hispanic/Latinx ethnicity using the second question. Please note that individuals who report a Hispanic/Latinx ethnicity but do not report one of the 5 NIH race categories (American Indian or Alaskan Native; Black or African American; Asian; Native Hawaiian or Pacific Islander; White) should be coded as “unknown” race.

Option 2: The US Census Measure

Background

The 2020 US Census improved its measurement of race and ethnicity. Specific details on the rationale for the changes can be found here:

• https://www.census.gov/programs-surveys/decennial-census/decade/2020/planning-management/release/faqs-race-ethnicity.html

• https://www.census.gov/newsroom/blogs/random-samplings/2021/08/improvements-to-2020-census-race-hispanic-origin-question-designs.html

• https://www.census.gov/library/stories/2021/08/improved-race-ethnicity-measures-reveal-united-states-population-much-more-multiracial.html

Pros and Cons

Pros: The benefit of this measure is that there are example subcategories in each racial/ethnic category to help individuals better determine where they fit within each racial category. Specific Asian categories are now broken out separately. Another benefit of this measure is that it can easily be coded and reported according to NIH categories.

Cons: It is longer than other measures and may take more time for participants to complete. It places White first. It does not provide specific categories for people who identify as North African, Middle Eastern, or Central and South American people. Further, Hispanic/Latinx participants may still not identify with any of the racial categories. The use of the term “other” may be stigmatizing and so using the term “another” may be preferred.

Measure

Are you of Hispanic, Latino, or Spanish origin?

• No, not of Hispanic, Latino, or Spanish origin

• Yes, Mexican, Mexican Am., Chicano

• Yes, Puerto Rican

• Yes, Cuban

• Yes, another Hispanic, Latino, or Spanish origin – Print, for example, Salvadoran, Dominican, Colombian, Guatemalan, Spaniard, Ecuadorian, etc.

Specify:__________________________

What is your race? Check one or more boxes AND specify your origin:

___ White – For example, German, Irish, English, Italian, Lebanese, Egyptian, etc.

Specify:_____________________________________________________

___ Black or African American – For example, African American, Jamaican, Haitian, Nigerian, Ethiopian, Somali, etc.

Specify:_____________________________________________________

___ American Indian or Alaska Native – For example, Navajo Nation, Blackfeet Tribe, Mayan, Aztec, Native Village of Barrow Inupiat Traditional Government, Nome Eskimo Community, etc.

Specify:_____________________________________________________

___ Chinese

___ Filipino

___ Asian Indian

___ Vietnamese

___ Korean

___ Japanese

___ Native Hawaiian

___ Samoan

___ Chamorro

___ Other Asian – For example, Pakistani, Cambodian, Hmong, etc.

Specify:_____________________________________________________

___ Other Pacific Islander – For example, Tongan, Fijian, Marshallese, etc.

Specify:_____________________________________________________

___ Some Other Race

Specify:_____________________________________________________

Analysis

a. Non-NIH Reporting

Similar to Option 1a analyses, all racial categories can be reported or collapsed as needed for modeling. If categories are collapsed, it is still recommended to report the subcategories in Table 1.

b. NIH Reporting

The U.S. Census recommends collapsing the categories as follows. For NIH reporting, use cross tabs to identify Hispanic/Latinx and non-Hispanic/Latinx individuals within each racial category.

Hispanic/Latinx/Spanish Ethnicity

1= Yes (include Cuban, Mexican, Mexican Am., Chicano, Puerto Rican to another Hispanic, Latino, or Spanish origin)

0 = No, not Hispanic/Latinx/Spanish ethnicity

Race

Categorize as a single race if participant only checked off a race that falls within the racial categories below, otherwise, code as more than one race.

1. White ONLY

2. American Indian or Alaskan Native ONLY

3. Black or African American ONLY

4. Asian ONLY (include those who identify as Chinese, Filipino, Asian Indian, Vietnamese, Korean, Japanese, other Asian such as Pakistani, Cambodian, Hmong)

5. Native Hawaiian or Other Pacific Islander ONLY (e.g., Samoan, Tokelauan, Tahitian, Guamanian or Chamorro, Fijian, Tongan, or Marshallese).

6. Two or more races/Multiracial (checked off two or more of the 5 racial categories above)

7. Some other race (did not check off one or more of the first 5 racial categories)