AIR Race Ethnicity FAQ

Below are responses to frequently asked questions about the new IPEDS Race/Ethnicity standards. In addition, you may visit the Race/Ethnicity BLOG for current discussions on this topic.


  • Q: Where can I find definitions of the race/ethnicity categories?
    A: See definitions from IPEDS glossary (PDF)

Data Collection

  • Q: Is it mandatory that we use two questions to collect the race/ethnicity data?
    A: Yes. The first question must ask ethnicity and the second question asks for racial information.
  • Q: Is there a firm "drop dead" date by which we must use the two-question format to collect race/ethnicity data?
    A: There is no one specific date. However, institutions are required to implement the collection of race and ethnicity information using the 2-question format so that data for the 2010-11 academic year can be reported using the new categories. Since institutions are not required to resurvey continuing students or existing staff and faculty, old race/ethnicity data may be mapped to the new categories for these individuals. On the other hand, race and ethnicity data for students, staff, and faculty new to the institution for 2010-11 must be collected using the 2-question format.
  • Q: Can wording be “Choose all that apply” for the race categories?
    A: No – wording must be “Select one or more,” "Choose one or more," "Check one or more," etc.
  • Q: Can I include an explanation for why the race/ethnicity questions are being asked?
    A: Yes, you can customize language similar to the sample below to introduce the questions.
    Colleges and universities are asked by many, including the federal government, accrediting associations, college guides, newspapers, and our own college/university communities, to describe the racial/ethnic backgrounds of our students and employees. In order to respond to these requests,, we ask you to answer the following two questions:
    Do you consider yourself to be Hispanic/Latino?
    _ Yes
    _ No
    In addition, select one or more of the following racial categories to describe yourself:
    _ American Indian or Alaska Native
    _ Asian
    _ Black or African American
    _ Native Hawaiian or Pacific Islander
    _ White
  • Q: On our various applications we ask race/ethnicity under a heading called “Optional Items”. Is it still permissible to have race/ethnicity under this heading?
    A: Yes, it is permissible to label the race/ethnicity questions optional.
  • Q: Can I collect the data in the same format that I will be submitting to IPEDS?
    A: No. You must ask the two questions.
  • Q: We are a women’s college. Can we change Hispanic/Latino to Hispanic/Latina?
    A: Yes.
  • Q: Are “Unknown,” “Two or more races,” and “Nonresident alien” valid options for collecting race/ethnic data?
    A: No. These are valid reporting categories, but should not be used in data collection.
  • Q: Can sub-categories of ethnicity and race be collected?
    A: Yes, as long as they can be aggregated to the IPEDS reporting categories. Remember to account for all subcategories in order to get an accurate count (to ensure this, add an “Other” category to your list of sub-categories).
  • Q: Can an institution collect subcategories of race and/or ethnicity from students and not from employees? Or from employees and not from students?
    A: As long as the institution uses collection processes that follow final guidance, it does not have to use exactly the same collection process for everyone. For example, subcategories can be collected from students, but not from employees.
  • Q: Can you change the order of race to reflect student body?
    A: Yes – the categories of race can be reordered. The order of races presented in the NCES statistical standard example 2-question format does not dictate how institutions must order the categories on their own data collection forms; it is the order used by NCES in its own data collections from individuals. Because the final guidance does not address order, it's left up to the institution to decide. Alphabetizing is an arbitrary order that most people accept and should not imply any bias. Putting one category first (e.g., "White" because it's most prevalent) may generate questions as to why that category was given preference when there is no one to explain the rationale. The bottom line is that the order of the race categories is each institution's decision.
  • Q: Should R/E be collected for international students/employees (Nonresident aliens [NRAs])?
    A: That is not required. However, if you collect the R/E data for NRAs, you’ll have the information if the person changes residency status and/or if you want to use the data for non-IPEDS reporting purposes. Also, it’s easier to collect the R/E data for everyone than to omit a subset of the population.
  • Q: How would students/staff from Brazil or Portugal respond to these questions?
    A: Provide the definitions (be sure to use language from the IPEDS glossary) and allow for self-selection.
  • Q: If a person answers yes to the ethnicity question, is s/he supposed to also answer the race question?
    A: Yes. Both questions are to be answered.
  • Q: Can I require students/employees to complete the race/ethnicity questions?
    A: No. You may only ask.
  • Q: How do I know if a student or employee refused to answer the questions or just overlooked them?
    A: You don't.
  • Q: What is the level of effort needed to collect the new information?
    A: Presenting the data collection form to students/employees is sufficient to ensure that individuals have had an opportunity to respond. Postsecondary institutions can report unknown when the respondent doesn’t reply—there is no need to use third-party observation to supply race/ethnicity.

Mapping and Re-Surveying

  • Q: What is Mapping?
    A: Mapping is the process of converting one set of categories to a different set of categories so that all individuals can be reported using the same categories. For example, to change everyone in your system to the new r/e categories, you can map (convert) their old r/e values to the new ones.
  • Q: Why should I map?
    A: If you do not map, all continuing students and employees will be “Unknown” when you transition unless you re-survey. If you re-survey and have not mapped first, non-respondents for whom you previously had data will be "Unknown."
  • Q: How do I map from the old categories to the new ones?
    A: Use the following table. All single-category old codes will map “reasonably” well except the Hawaiians and Pacific Islanders embedded in the current "Asian or Pacific Islander" category will be incorrectly assumed to be Asian.
    Current New
    "Hispanic" "Hispanic"
    “American Indian or Alaskan Native” “American Indian or Alaska Native”
    “Black, non-Hispanic” “Black or African American”
    “White, non-Hispanic” “White”
    “Asian or Pacific Islander”* “Asian*”
    “Unknown” “Unknown”
  • Q: Does anyone have a guideline for mapping new to old? All of the templates I have seen are for old to new, so how do you take two or more races and determine which would have been the one response?
    A: Probably the best way to map new race/ethnic codes to old codes is to use the following table. Everything should map well enough except "two or more races" which has to go into Unknown.
    New Old
    “Hispanic/Latino” “Hispanic”
    “American Indian or Alaskan Native” “American Indian or Alaska Native”
    “Black or African American” “Black , non-Hispanic”
    “White” “White , non-Hispanic”
    “Asian” “Asian or Pacific Islander”
    “Native Hawaiian/Other Pacific Islander” “Asian or Pacific Islander”
    “Two or More Races” “Unknown”
    “Unknown” “Unknown”
  • Q: If I map and re-survey, which should I do first?
    A: It is suggested that you map first and then re-survey. As part of the re-survey process, you can show the students and employees what race/ethnic data you currently have for them and ask them to update. With that approach you could assume non-respondents don’t wish to change their old data. If you re-survey and then map, you could be over-writing the respondent’s most recent preferred category with an older one.
  • Q: Is re-surveying required?
    A: No. Re-surveying is encouraged but not required.
  • Q: Do we have to survey students who have left the institution?
    A: No-surveying is NOT REQUIRED for any cohort. Surveying dropouts and/or early graduates is a possibility. However, using mapped data for those who left the institution prior to the transition is sufficient for GRS cohorts, particularly since many of their addresses (e-mail or otherwise) may no longer be valid.
  • Q: Can I just re-survey the students in GRS cohorts and not all students?
    A: .“Yes. Although it is recommended by your AIR colleagues that you use mapping for the GRS cohorts. If you resurvey, you may not be able to reach the entire cohort, which means will likely be inflating graduation rates for Hawaiian/Pacific Islanders and “Two or more races” and deflating them for Asians.”
  • Q: Do I use the data collected in the re-survey process to update race/ethnicity for the GRS cohorts who entered prior to the transition?
    A: No. It is suggested that you map old categories to new ones for the cohorts that matriculated before the transition but do not over-write those data with re-surveyed data because not all students will have had an opportunity to be re-surveyed (e.g., students who dropped out or graduated before the re-survey).


  • Q: Can we store the data as reported or do we have to maintain the detail from the collection questions?
    A: You need the detail. You might need it for internal reporting purposes (e.g., providing a list of all Pacific Islanders) or for a compliance review.
  • Q: Can I consider my institution’s archived student databases as sufficient for maintaining prior year’s data?
    A: Yes.


  • Q: How do you report to IPEDS if the ethnicity question is not answered?
    A: If ethnicity is blank (missing), report any racial information that is supplied.
  • Q: How do you determine non-resident alien?
    A: Through the person’s Visa type. Do not use the response to a citizenship question.


  • Q: When do the changes become effective?
    A: Academic Year (AY) 2010-11 Reporting
    Winter 2010-11 for HR; Spring 2011 for enrollment
  • Q: What is the earliest an institution can start reporting using the new standards?
    A: 2008-09. Although implementation isn’t mandatory until 2010-11, it is optional in 2008-09 and 2009-10.