Planning: variable queue and editorial TODOs — ethnicity_plan.md (Task 167).
Feelings are converging¶
On a scale of 0 to 100, how do you feel about Black people? It’s a strange question, but the ANES has asked people to answer it, every four years, since 1964. Before we see the responses, think about how you would answer. The question comes with this explanation:
Ratings between 50 and 100 degrees mean that you feel favorably ... toward the group; ratings between 0 and 50 degrees mean that you don’t feel favorably... If you don’t feel particularly warm or cold ... you would rate them at 50 degrees.
I am not sure how I would answer. On average, I think Black people are about average -- so I might say 50 degrees. But I generally have warm feelings about other people, so I might say 70. It seems like other people follow a similar thought process, because the most common answer is 50 degrees and the average is around 70 (room temperature).
The following figure shows the mean rating of Black people, as rated by non-Hispanic Whites and Blacks, and other respondents, which includes Hispanics.
Figure 1:VCF0206 — Time series by race of respondent: mean feeling thermometer rating of Black people, in degrees.
White respondents’ average rating of Blacks rose from 60 to 68 degrees, and Other respondents’ from 60 to 70. Black respondents rated Black people higher than the other groups did, but their rating fell from 89 to 82.
Here’s how people rated Whites.
Figure 2:VCF0207 — Time series by race of respondent: mean feeling thermometer rating of White people, in degrees.
Black respondents’ rating of White people rose from 67 to 73, then fell to 64. White respondents’ rating of Whites fell from 85 degrees to 70, and other respondents’ from 75 to 65.
These declines are driven primarily by generational replacement -- if we factor out the cohort components, the remaining period components are close to flat. And that suggests an interpretation, which is that these changes reflect a decreasing tendency to judge people by their racial group: among White respondents, it might reflect a decline of White Supremacy; among Black respondents, it might reflect an inclination to say that on average, White people are about average.
To help see whether this interpretation holds up, let’s look at one more figure, the ratings of Asian Americans, which the ANES added in 1992.
Figure 3:VCF0227 — Time series by race of respondent: mean feeling thermometer rating of Asian Americans, in degrees.
White respondents’ average rating of Asian Americans rose from 59 to 70 degrees, Black respondents’ from 62 to 67, and Other respondents’ from 65 to 71.
And the ratings of Hispanics, which the ANES added in 1976 (wording has varied over the years: Chicanos, Hispanics, Hispanic-Americans, Hispanics or Latinos).
Figure 4:VCF0217 — Time series by race of respondent: mean feeling thermometer rating of Hispanics, in degrees, 1976–2024.
White respondents’ average rating of Hispanics rose from 54 degrees in 1976 to 69 in 2024, and Black respondents’ from 63 to 70. Other respondents, including Hispanics, rated Hispanics higher than the other groups did, and their rating changed less, from 71 to 74. For all respondents together, the average rose from 56 to 70.
I think we can summarize all of these results in one word: convergence. In 2024, if we look at the ratings of four groups by three groups, we have twelve averages. Ten of them are between 64 and 71, and statistically all but indistinguishable. The exceptions are in-group ratings: Black people have warmer feelings toward other Black people, and Other respondents, who include Hispanics, rate Hispanics a little higher (74) -- but even the exceptions are on the path to convergence.
So the idea that we should feel especially warm or cool toward an entire racial group, as a group, seems to be disappearing.
Social distance¶
Since 1990, the GSS has asked about three groups:
What about having a close relative marry a [Black / Hispanic American / Asian American] person? Would you be very in favor of it happening, somewhat in favor, neither in favor nor opposed to it happening, somewhat opposed, or very opposed?
Everyone is asked about all three groups, so we can see how each group’s own members answer, too. The following figures show the percentage who said somewhat or very opposed, by race and ethnicity of the respondent.
Figure 5:marblk — Time series by race and ethnicity of respondent: percent somewhat or very opposed to a close relative marrying a Black person (GSS).
Among White respondents, opposition to a close relative marrying a Black person fell from 65% to 7% between 1990 and 2024; among Hispanic respondents, from 36% to 3%; among Asian respondents, from 26% in 2004 to 9%. Black respondents’ opposition was 5% in 1990 and 3% in 2024. Gallup’s question about approving of marriage between Black and White people, which goes back farther, shows the same decline: among White adults, disapproval fell from 65% to 5% between 1972 and 2021, and among Black adults from 21% to 3% in 2013. The levels differ from the GSS because the questions do: in Gallup’s data, disapproval among White adults was lower than GSS opposition in the early 1990s, and among Black adults it was higher.
Figure 6:marhisp — Time series by race and ethnicity of respondent: percent somewhat or very opposed to a close relative marrying a Hispanic American person (GSS).
Opposition to a close relative marrying a Hispanic person fell from 43% to 4% between 1990 and 2024 among White respondents, and from 27% to 3% among Black respondents; among Asian respondents, it fell from 19% in 2004 to 6%. Hispanic respondents’ opposition was 5% in 1990 and 1% in 2024.
Figure 7:marasian — Time series by race and ethnicity of respondent: percent somewhat or very opposed to a close relative marrying an Asian American person (GSS).
Opposition to a close relative marrying an Asian person fell from 45% to 4% between 1990 and 2024 among White respondents, from 26% to 5% among Black respondents, and from 31% to 3% among Hispanic respondents. Among Asian respondents, it was 7% in 2004 and 3% in 2024.
Attitudes are one thing; behavior is another. The following figure shows the percentage of newlyweds in each group who married someone of a different race or ethnicity, based on census data from IPUMS USA.
Figure 8:newlywed_intermarriage_by_group — Percent of newlyweds married to someone of a different race or ethnicity: Hispanic (any race) and non-Hispanic Asian and Pacific Islander, Black, and White; multiracial people are their own group. 2008–2024: married in the past year (American Community Survey); lines from the time model. 1900–1980 (hollow circles, joined to 2008 by dotted lines): approximated from years married (1900, 1910, 1950) or age at first marriage (1930–1980, first marriages only); the 1990 and 2000 censuses did not ask. Hispanic from 1980; Asian values resting on fewer than 100 newlyweds left out. IPUMS USA.
In 1960, 18% of Asian newlyweds married someone of a different race; by 1980 it was 31%, and it has been near 30% since. Among Hispanic newlyweds, it was 25% in 1980 and 31% in 2024. Among Black newlyweds, it was 1% in 1960, 5% in 1980, 17% in 2008, and 26% in 2024; among White newlyweds, it was less than 1% in 1960, 4% in 1980, 9% in 2008, and 16% in 2024.
These estimates are consistent with those in a 2017 Pew report, which found that 29% of Asian, 27% of Hispanic, 18% of Black, and 11% of White newlyweds in 2015 had married someone of a different race or ethnicity.
Stereotypes¶
The GSS and the ANES ask respondents to rate several groups, including White people, on seven-point scales. In the GSS, the instructions are:
Now I have some questions about different groups in our society. I’m going to show you a seven-point scale on which the characteristics of people in a group can be rated. In the first statement a score of 1 means that you think almost all of the people in that group are “rich.” A score of 7 means that you think almost everyone in the group are “poor.” A score of 4 means you think that the group is not towards one end or another, and of course you may choose any number in between that comes closest to where you think people in the group stand.
The GSS then asks whether people in each group “tend to be hard-working or if they tend to be lazy,” whether they “tend to be unintelligent or tend to be intelligent,” and, in 1990 and 2000 only, whether they “tend to be violence prone or if they tend not to be prone to violence.” The ANES uses a similar seven-point scale for hard-working or lazy, since 1992, and for peaceful or violent, since 2016.
To measure stereotypes about a group, I compare each respondent’s rating of the group with their own rating of White people on the same scale. For example, if someone rates White people 4 on the hard-working–lazy scale and Hispanic people 5 (closer to lazy), the difference is −1; if they rate both 4, the difference is 0. The figures show the average of these differences, in points on the seven-point scale. Positive values mean the group is rated more favorably than White people (harder-working, more intelligent, more peaceful); negative values mean the group is rated toward the negative end of the scale; zero means no difference.
The lines show results separately for White, Black, Hispanic, and Asian respondents. White and Black respondents are non-Hispanic. In the GSS, Hispanic and Asian respondents are identified by the country of their ancestors, which is consistent back to 1972; in the ANES, they are identified by self-reported race and Hispanic origin. Solid lines show the GSS; dashed lines show the ANES. In the figures for peaceful or violent, which come from the GSS in 1990 and 2000 and the ANES since 2016, both are solid, connected by a dotted line. The GSS had few Asian respondents before 2000, so their lines appear only where there are enough of them.
Hispanic people¶
Figure 9:intlhsp_diff — Time series by respondent group: mean difference between the rating of Hispanic people and White people on the GSS intelligent–unintelligent scale, 1990–2022. Positive = rated more intelligent.
In 1990, the average difference was −1.0 for all respondents: Hispanic people were rated less intelligent than White people. By 2022, it was −0.1. Among White respondents, it went from −1.0 to −0.2; among Black respondents, from −0.8 to 0.0; among Hispanic respondents, from −0.6 to −0.1.
Figure 10:workhsp_diff — Time series by respondent group: mean difference between the rating of Hispanic people and White people on the GSS hard-working–lazy scale, 1990–2024. Positive = rated harder-working. The ANES series (VCF9272_diff) is not shown; it is very similar.
In the GSS, the average difference went from −1.0 in 1990 to +0.8 in 2024: Hispanic people were rated lazier than White people in the 1990s and harder-working than White people in recent years. Among White respondents the difference went from −1.1 to +0.6; among Black respondents, from −0.8 to +1.5; among Hispanic respondents, from +0.1 to +1.3; and among Asian respondents, from −0.9 to +0.6. The ANES asks the same question, and its results are very similar: the average went from −0.7 in 1992 to +0.7 in 2024.
Figure 11:violhsp_diff, violent_hisp_diff — Time series by respondent group: mean difference between the rating of Hispanic people and White people on the violence scale. GSS, 1990 and 2000, and ANES, 2016–2024, with a dotted line connecting them. Positive = rated more peaceful.
In the GSS, the average difference was −0.8 in 1990 and −0.5 in 2000: Hispanic people were rated more violent than White people. Among White respondents it went from −0.8 to −0.5; among Black respondents, from −0.8 to +0.2; and among Hispanic respondents, it was −0.6 and −0.5. In the ANES, the average difference was −0.1 in 2016 and +0.1 in 2024. Among White respondents it went from −0.3 to −0.1, and among Asian respondents from −0.3 to 0.0; among Black respondents, it was +0.5 and +0.6, and among Hispanic respondents, +0.3 and +0.4.
Black people¶
Figure 12:intlblk_diff — Time series by respondent group: mean difference between the rating of Black people and White people on the GSS intelligent–unintelligent scale, 1990–2022. Positive = rated more intelligent.
In 1990, the average difference was −0.8 for all respondents; by 2022, it was −0.1. Among White respondents, it went from −0.9 to −0.1; among Hispanic respondents, from −0.8 to −0.2; among Asian respondents, from −0.6 to −0.5. Among Black respondents, it went from −0.3 to +0.2.
Figure 13:workblk_diff — Time series by respondent group: mean difference between the rating of Black people and White people on the GSS hard-working–lazy scale, 1990–2024. Positive = rated harder-working. The ANES series (VCF9271_diff) is not shown; it is very similar.
In the GSS, the average difference went from −1.2 in 1990 to −0.1 in 2024. Among White respondents it went from −1.3 to −0.2; among Hispanic respondents, from −1.0 to −0.1; among Asian respondents, from −1.7 to −0.3; and among Black respondents, from −0.2 to +0.7. The ANES asks the same question, and its results are very similar: the average went from −1.0 in 1992 to −0.2 in 2024.
Figure 14:violblk_diff, violent_black_diff — Time series by respondent group: mean difference between the rating of Black people and White people on the violence scale. GSS, 1990 and 2000, and ANES, 2016–2024, with a dotted line connecting them. Positive = rated more peaceful.
In the GSS, the average difference was −1.0 in 1990 and −0.7 in 2000. Among White respondents it went from −1.1 to −0.7, among Hispanic respondents from −0.9 to −1.0, and among Black respondents from −0.6 to 0.0. In the ANES, the average difference went from −0.8 in 2016 to −0.3 in 2024. Among White respondents it went from −1.0 to −0.5, among Hispanic respondents from −0.8 to −0.3, and among Asian respondents from −1.0 to −0.9; among Black respondents, it was +0.4 and +0.5.
Asian people¶
Figure 15:intlasn_diff — Time series by respondent group: mean difference between the rating of Asian people and White people on the GSS intelligent–unintelligent scale, 1990 and 2000. Positive = rated more intelligent.
The GSS asked about Asian people only in 1990 and 2000. The average difference was −0.4 in 1990 and −0.1 in 2000. Among White respondents it went from −0.4 to −0.1, among Black respondents from −0.4 to 0.0, and among Hispanic respondents from −0.6 to −0.2.
Figure 16:VCF9273_diff — Time series by respondent group: mean difference between the rating of Asian people and White people on the ANES hard-working–lazy scale, 1992–2024. Positive = rated harder-working. The GSS asked this question about Asian people only in 1990, 1994 and 2000 (workasn_diff); not shown.
In the ANES, the average difference went from 0.0 in 1992 to +0.7 in 2024: Asian people are now rated harder-working than White people. Among White respondents it went from −0.1 to +0.5; among Hispanic respondents, from 0.0 to +1.0; among Black respondents, from +0.1 to +1.0; and among Asian respondents, from +0.6 to +1.4. The GSS asked the same question about Asian people in 1990, 1994 and 2000; the average difference was −0.2 in 1990 and +0.2 in 2000.
Figure 17:violasn_diff, violent_asian_diff — Time series by respondent group: mean difference between the rating of Asian people and White people on the violence scale. GSS, 1990 and 2000, and ANES, 2016–2024, with a dotted line connecting them. Positive = rated more peaceful.
In the GSS, the average difference was −0.2 in 1990 and +0.2 in 2000. In the ANES, it was +0.6 in 2016 and +0.8 in 2024: Asian people are rated more peaceful than White people. In the ANES, among White respondents it went from +0.4 to +0.6; among Hispanic respondents, from +0.8 to +1.0; among Asian respondents, from +0.9 to +1.2; and among Black respondents, from +1.2 to +1.4.
Discrimination¶
Since 2012, the ANES has asked:
How much discrimination is there in the United States today against each of the following groups? [Blacks / Hispanics / Asians / Whites]
The responses are a great deal, a lot, a moderate amount, a little, or none at all. The question about Asian people was added in 2016.
Pew has asked a similar question since 2017:
Please tell me how much discrimination there is against each of these groups in our society today ... a lot, some, only a little, or none at all?
The following figure shows the percentage of ANES respondents who said a great deal or a lot (lines), and the percentage of Pew respondents who said a lot (circles). The squares show the same question in phone surveys in 2001 (Washington Post, Kaiser Family Foundation, and Harvard) and 2009 (Pew).
Figure 18:discrim_black, discrim_hisp, discrim_asian, discrim_white — Time series: percent saying there is a great deal or a lot of discrimination against each group (ANES, lines). Hollow circles: percent saying a lot, Pew American Trends Panel, online, 2017–2025 (Asian people from 2021). Hollow squares: the same question by phone, April 2001 (Washington Post / Kaiser / Harvard) and November 2009 (Pew), from Pew’s published topline; dotted lines join each group’s Pew points.
In the ANES, the percentage who said there is a great deal or a lot of discrimination against Black people rose from 38% to 51% between 2012 and 2024. For Hispanic people, it was 33% in 2012 and 35% in 2024; for Asian people, it was 11% in 2016, 19% in 2020, and 21% in 2024. For White people, it was 8% in 2012 and 9% in 2024.
In the Pew surveys, the percentage who said there is a lot of discrimination against Black people was 39% in 2017, peaked at 47% in 2021, and was 36% in 2025. For Hispanic people, it ranged from 28% to 35%; for Asian people, it was 27% in 2021 and 22% in 2025; for White people, it ranged from 10% to 14%. In the phone surveys, the percentages were lower: 25% and 18% for Black people in 2001 and 2009, 19% and 23% for Hispanic people, and 8% for Asian people in both years.
The two surveys use different scales, so their levels are not directly comparable. And in an experiment in 2017, Pew asked the question by phone and online to randomly assigned groups: online respondents were more likely to say there is a lot of discrimination against Hispanic people (34% compared to 29%).
Gallup has asked a related question since 2001:
Next, we’d like to know how you feel about the way various groups in society are treated. For each of the following groups please say whether you are very satisfied, somewhat satisfied, somewhat dissatisfied, or very dissatisfied with the way they are treated.
The following figure shows how members of each group see the treatment of their own group: the percentage of Black adults who said they were very or somewhat satisfied with the way Black people are treated, and the same for Hispanic adults and Hispanic people. Gallup asked about White people only in 2021 and 2025, so White adults’ views of their own treatment are two points.
Figure 19:gtreat_black, gtreat_hisp — Percent very or somewhat satisfied with the way their own group is treated, Gallup, 2001–2025: Black adults on Black people and Hispanic adults on Hispanic people (lines from the time model; hollow circles the published values), and White adults on White people (2021 and 2025). Gallup does not publish results for Asian adults.
Among Black adults, satisfaction with the treatment of Black people was 41% in 2001 and 22% in 2025. Among Hispanic adults, satisfaction with the treatment of Hispanic people was 49% in 2001 and 33% in 2025. Among White adults, satisfaction with the treatment of White people was 69% in 2021 and 74% in 2025.
Members of each group are less satisfied than others are with the way their group is treated: in 2025, 24% of Black adults were satisfied with the treatment of Black people, compared to 45% of White adults, and 31% of Hispanic adults were satisfied with the treatment of Hispanic people, compared to 47% of White adults.
Gallup does not report results for Asian adults, but among all adults, satisfaction with the treatment of Asian people was 69% in 2001, reached 78% in 2013, fell to 46% in 2021, and was 56% in 2025.
The drop for Asian people in 2021 coincides with reports of harassment and violence against Asian Americans during the coronavirus pandemic. In June 2020, Pew asked:
Compared to before coronavirus, how common do you think it is for people to express racist or racially insensitive views about each of the following groups in our society?
About 40% said racist views about Asian people had become more common, compared to 30% for Black people, 19% for White people, and 14% for Hispanic people. In April 2021, Pew asked whether violence against Asian Americans was increasing, decreasing, or staying the same: 57% said it was increasing. In April 2022, 42% said so.
Gallup has also asked about relations between groups:
Next, we’d like to know how you would rate relations between various groups in the United States these days. Would you say relations between [two groups] are very good, somewhat good, somewhat bad, or very bad?
The following figures show the percentage who said very or somewhat good, as rated by members of the two groups in each question. Gallup does not publish results for Asian adults, so the last figure shows White adults’ ratings of relations between White and Asian people and Black adults’ ratings of relations between Black and Asian people.
Figure 20:grel_wb_white, grel_wb_black — Percent saying relations between White and Black people are very or somewhat good, White and Black adults, Gallup, 2001–2025. Lines from the time model; hollow circles the published values.
Among White adults, the percentage who said relations between White and Black people are good was 68% in 2001 and 52% in 2025; among Black adults, it fell from 68% to 41%.
Figure 21:grel_wh_white, grel_wh_hisp — Percent saying relations between White and Hispanic people are very or somewhat good, White and Hispanic adults, Gallup, 2001–2025. Lines from the time model; hollow circles the published values.
For relations between White and Hispanic people, it was 69% in 2001 and 64% in 2025 among White adults, and 74% and 59% among Hispanic adults.
Figure 22:grel_bh_black, grel_bh_hisp — Percent saying relations between Black and Hispanic people are very or somewhat good, Black and Hispanic adults, Gallup, 2001–2025. Lines from the time model; hollow circles the published values.
For relations between Black and Hispanic people, it was 74% in 2001 and 71% in 2025 among Black adults, and 70% and 73% among Hispanic adults.
Figure 23:grel_wa_white, grel_ba_black — Percent saying relations are very or somewhat good: White adults on relations between White and Asian people (2001–2025) and Black adults on relations between Black and Asian people (2015–2025), Gallup. Lines from the time model; hollow circles the published values.
Among White adults, the percentage who said relations between White and Asian people are good was 78% in 2001 and 76% in 2025. Among Black adults, the percentage who said relations between Black and Asian people are good was 75% in 2015 and 62% in 2025.
Representation in government¶
The following figure shows the number of Hispanic members of the House and Senate in each Congress, based on records from the House Historian.
Figure 24:hispanic_members_of_congress — Number of Hispanic members of the House and Senate in each Congress, 1821–2025: the most serving at the same time, not counting non-voting Delegates and Resident Commissioners, including territorial delegates before statehood (House Historian).
The first Hispanic member of Congress was Joseph Marion Hernández, who represented the Florida Territory as a non-voting delegate starting in 1822. The House Historian lists Charles Bouligny of Louisiana, who served from 1824 to 1829, as the first Hispanic senator. Romualdo Pacheco, who had been governor of California, became the first Hispanic voting member of the House in 1877.
For most of the next century, there were never more than one or two Hispanic members in each chamber, and many of them came from New Mexico. Octaviano Larrazolo of New Mexico became the first Mexican American senator in 1928; Dennis Chávez, also of New Mexico, served in the House from 1931 and in the Senate from 1935 until his death in 1962.
The number in the House began to grow in the 1960s, with Henry B. González of Texas in 1961 and Edward Roybal of California in 1963. In 1989, Ileana Ros-Lehtinen of Florida became the first Hispanic woman in Congress. The number of Hispanic representatives went from 3 in 1965 to 11 in 1985, 18 in 1993, 27 in 2005, and 53 in 2023; in the current 119th Congress there are 50.
The Senate had no Hispanic members from 1977 until 2005, when Mel Martinez of Florida and Ken Salazar of Colorado took office. In 2017, Catherine Cortez Masto of Nevada became the first Latina senator. In the current Congress, there are 7 Hispanic senators.
The next figure shows the percentage of seats held by Hispanic Americans in the House, Senate, Supreme Court, and Cabinet since 1917. The shaded area shows the Hispanic share of the population. The Census has asked about Hispanic origin since 1970; for earlier years, the shading uses estimates from census records, based on birthplace, parents’ birthplace, and Spanish surname, from IPUMS USA.
Figure 25:hispanic_representation_in_government — Percent of seats held by Hispanic Americans in the House and Senate (House Historian), on the Supreme Court, and in the Cabinet, 1917–2025, and, shaded gray, the Hispanic share of the U.S. population (Census from 1970; before 1970, estimated from IPUMS USA census samples with the Gratton–Gutmann rules; each census holds until the next). Cabinet: the most Hispanic members serving at the same time in each administration, as a percent of cabinet and cabinet-level posts, counted as in the gender and race chapters (posts from CAWP).
The Hispanic share of House seats was below 1% until 1971, reached 2.5% in 1985 and 6.2% in 2005, and was 11.5% in 2025. Over the same period, the Hispanic share of the population grew from 1.2% in 1920 to 4.5% in 1970 and 18.7% in 2020. In the Senate, it was between 2% and 4% from 2005 to 2019, 6% in 2021 and 2023, and 7% in 2025.
Sonia Sotomayor became the first Hispanic justice of the Supreme Court in 2009.
In the executive branch, there are some notable milestones:
In 1988, Lauro Cavazos became Secretary of Education, the first Hispanic member of a presidential cabinet.
In 2005, Alberto Gonzales became the first Hispanic Attorney General.
In 2021, Alejandro Mayorkas became the first Hispanic Secretary of Homeland Security, and the first immigrant to hold the post.
In 2025, Marco Rubio became the first Hispanic Secretary of State.
No Hispanic American has been nominated by a major party for President or Vice President.
The highest percentages of cabinet and cabinet-level posts were 17% in Obama’s second term (4 of 23) and 16% under Biden (4 of 25); the lowest since 1988 were about 5%, in George W. Bush’s first term and Trump’s first.
Hispanic Americans in politics¶
There are not many long-running survey questions about Hispanic Americans in politics, but there are a few. Since 2007, Gallup has asked:
If your party nominated a generally well-qualified person for president who happened to be Hispanic, would you vote for that person?
The percentage who said yes was 87% in 2007, 91% in 2012 and 2015, and 95% in 2019.
Since 2008, the ANES has asked:
Would you say that [group] have too much influence in American politics, just about the right amount of influence in American politics, or too little influence in American politics?
The following figure shows the percentage who said too little, for Black, Hispanic, Asian, and White people.
Figure 26:VCF9275, influence_hisp, influence_asian, influence_white — Time series: percent saying the group has too little influence in American politics (ANES). Black people since 2008, Hispanic and White people since 2012, Asian people since 2016.
The percentage who said Hispanic people have too little influence rose from 37% to 52% between 2012 and 2024, almost exactly the same as for Black people, which rose from 35% in 2008 to 51%. For Asian people, it went from 46% in 2016 to 50%. For White people, it fell from 9% to 4%.
Since 2016, the ANES has also asked:
How important is it that more [group] get elected to political office?
The following figure shows the percentage who said extremely or very important.
Figure 27:elect_hisp, elect_black, elect_asian — Time series: percent saying it is extremely or very important that more of the group get elected to political office (ANES). Hispanic people since 2016, Black and Asian people since 2020.
For Hispanic people, the percentage was 21% in 2016, 32% in 2020, and 26% in 2024. For Black people, it was 38% in 2020 and 31% in 2024; for Asian people, 32% and 25%.
Hispanic Americans in the workplace¶
As we saw in the chapters on gender and race, private companies with 100 or more employees report to the Equal Employment Opportunity Commission (EEOC) how many of their employees are in each job category, by sex, race, and ethnicity.
The following figure shows the percentage of Hispanic employees at these companies, among all employees and in selected roles: professionals, mid-level managers, and senior managers. It also shows the percentage of CEOs at Fortune 500 companies who are Hispanic. The shaded area shows the Hispanic share of the population.
Figure 28:corporate_leadership_hispanic — Percent Hispanic among all employees, professionals, and managers at companies that report to the EEOC, and among the CEOs of Fortune 500 companies (purple; Hispanic CEOs serving on June 1 of each year, out of 500). Through 2013, “Officials and Managers” is one category (gray line); from 2014 it is split into mid-level and senior managers. Shaded gray: the Hispanic share of the U.S. population (Census from 1970; IPUMS USA estimate for 1960; each census holds until the next).
In 1966, 2.5% of employees at EEOC-reporting companies were Hispanic; in 2023, 18.6% were. That’s close to the Hispanic share of the population, which was 18.7% in 2020, and of all employed people, which was 20% in 2025. Among professionals, the percentage rose from 0.8% to 9.3% between 1966 and 2023.
Among “Officials and Managers,” it rose from 0.6% to 7.4% between 1966 and 2013. After that category was split, representation reached 11.6% among mid-level managers and 5.7% among senior managers in 2023.
At the top, Roberto Goizueta, who was born in Cuba, led Coca-Cola from 1981 until his death in 1997. In 2017, Geisha Williams of PG&E became the first Latina CEO of a Fortune 500 company. The number of Hispanic CEOs on the list was 2 in 1995, 10 in 2012, and 17 in 2024 and 2025; in 2026 it is 15, which is 3% of the list.
Asian Americans in politics¶
In this section and the next, “Asian Americans” includes Pacific Islanders, following the House Historian’s table and the EEOC’s historical categories.
The following figure shows the number of Asian and Pacific Islander members of the House and Senate in each Congress, based on records from the House Historian.
Figure 29:asian_members_of_congress — Number of Asian and Pacific Islander members of the House and Senate in each Congress, 1903–2025: the most serving at the same time, not counting non-voting Delegates and Resident Commissioners, including Hawaii’s territorial delegates before statehood and the Philippines’ Resident Commissioners (House Historian).
Before 1957, the only Asian and Pacific Islander members of Congress were non-voting: Jonah Kūhiō Kalanianaʻole represented the Territory of Hawaii from 1903 until his death in 1922, and Resident Commissioners represented the Philippines from 1907 to 1946.
In 1957, Dalip Singh Saund of California, who was born in India, became the first Asian American voting member of Congress. When Hawaii became a state in 1959, Hiram Fong became the first Asian American senator and Daniel Inouye joined the House; Inouye moved to the Senate in 1963 and served there until his death in 2012. In 1965, Patsy Mink of Hawaii became the first woman of color in Congress. In 1990, Daniel Akaka became the first Native Hawaiian senator.
For several decades, most Asian and Pacific Islander members came from Hawaii and California. The number in the House was 2 or 3 through the 1980s, 5 in 1993, 9 in 2009, and 18 in the current 119th Congress. In the Senate, there have been 1 to 3 at a time since 1959. In 2013, Mazie Hirono of Hawaii became the first Asian American woman in the Senate; Tammy Duckworth of Illinois and Kamala Harris of California joined her in 2017. In 2024, Andy Kim of New Jersey became the first Korean American senator.
The next figure shows the percentage of seats held by Asian and Pacific Islander Americans in the House, Senate, Supreme Court, and Cabinet since 1917. The shaded area shows the Asian and Pacific Islander share of the population.
Figure 30:asian_representation_in_government — Percent of seats held by Asian and Pacific Islander Americans in the House and Senate (House Historian), on the Supreme Court, and in the Cabinet, 1917–2025, and, shaded gray, the Asian and Pacific Islander share of the U.S. population (Census; each census holds until the next; from 2000, Asian or Pacific Islander alone or in combination with another race). Cabinet: the most Asian and Pacific Islander members serving at the same time in each administration, as a percent of cabinet and cabinet-level posts, counted as in the gender and race chapters (posts from CAWP).
The Asian and Pacific Islander share of House seats was below 1% until 1993, reached 2.1% in 2009, and was 4.1% in 2025. Over the same period, the Asian and Pacific Islander share of the population grew from 0.8% in 1970 to 7.7% in 2020. In the Senate, the share has been 1% to 3% since 1959, and 3% since 2017.
No Asian or Pacific Islander American has served on the Supreme Court.
In the executive branch, there are some notable milestones:
In 2000, Norman Mineta became Secretary of Commerce, the first Asian American member of a presidential cabinet; he was Secretary of Transportation from 2001 to 2006.
In 2001, Elaine Chao became Secretary of Labor, the first Asian American woman in the Cabinet; she was Secretary of Transportation from 2017 to 2021.
In 2021, Kamala Harris became the first Asian American Vice President, and in 2024 she was the Democratic nominee for President.
The highest percentages of cabinet and cabinet-level posts were 13% in Obama’s first term (3 of 23) and 12% under Biden (3 of 25); in Trump’s second term, it was 4% (1 of 23).
In the ANES questions in the previous section, the percentage who said Asian people have too little influence in American politics went from 46% to 50% between 2016 and 2024, and the percentage who said it is extremely or very important that more Asian people get elected was 32% in 2020 and 25% in 2024.
Asian Americans in the workplace¶
The following figure shows the percentage of Asian and Pacific Islander employees at companies that report to the EEOC, among all employees and in selected roles: professionals, mid-level managers, and senior managers. It also shows the percentage of CEOs at Fortune 500 companies who are Asian or Pacific Islander. The shaded area shows the Asian and Pacific Islander share of the population.
Figure 31:corporate_leadership_asian — Percent Asian or Pacific Islander among all employees, professionals, and managers at companies that report to the EEOC, and among the CEOs of Fortune 500 companies (purple; Asian and Pacific Islander CEOs serving on June 1 of each year, out of 500). Through 2013, “Officials and Managers” is one category (gray line); from 2014 it is split into mid-level and senior managers. Shaded gray: the Asian and Pacific Islander share of the U.S. population (Census; each census holds until the next).
In 1966, 0.5% of employees at EEOC-reporting companies were Asian or Pacific Islander; in 2023, 7.6% were. That’s close to the Asian and Pacific Islander share of the population, which was 7.7% in 2020, and of all employed people, which was 7.5% in 2025.
Among professionals, the percentage rose from 1.3% to 14.4% between 1966 and 2023, about twice the share among all employees. Among “Officials and Managers,” it rose from 0.3% to 5.5% between 1966 and 2013. After that category was split, representation reached 8.6% among mid-level managers and 7.7% among senior managers in 2023.
At the top, Andrea Jung led Avon from 1999 to 2012, and Indra Nooyi led PepsiCo from 2006 to 2018. The number of Asian and Pacific Islander CEOs on the list was 2 in 1995, 7 in 2000, and 8 in 2012; since then it has grown to 17 in 2020 and 34 in 2026, which is 6.8% of the list. Many of the recent CEOs lead technology companies, including Satya Nadella at Microsoft, Sundar Pichai at Alphabet, Jensen Huang at Nvidia, and Lisa Su at AMD.
White Americans in government and business¶
To wrap up the discussion of representation, the last two figures show the same measures for non-Hispanic White Americans. No source counts White officeholders directly, so we compute these series by subtraction: the number of seats or positions, minus the Black, Hispanic, Asian and Pacific Islander, and Native American people who held them. People who appear on more than one list, like Kamala Harris, are subtracted once. Following the Census, people of Middle Eastern and North African descent are counted as White.
Figure 32:white_representation_in_government — Percent of seats held by non-Hispanic White Americans in the House and Senate, on the Supreme Court, and in the Cabinet, 1917–2025, and, shaded gray, the non-Hispanic White share of the U.S. population (Census from 1970; before 1970, estimated from IPUMS USA census samples; each census holds until the next). House and Senate: composition at the start of each congressional term, voting seats minus the Black, Hispanic, Asian and Pacific Islander (House Historian), and Native American (Biographical Directory) members serving that day. Cabinet: the minimum White share during each administration, 100% minus the largest share of non-White members serving at the same time (posts from CAWP).
In 1917, White members held 99.3% of the seats in the House; the share was still 98.2% in 1961, then fell to 86.2% in 1993, 83.2% in 2009, and 71.0% in 2025. In the Senate, the White share was 96% or more until 2003, between 94% and 96% from 2005 to 2015, and 84% in 2025. Over the same period, the non-Hispanic White share of the population fell from 88.3% in 1920 to 83.5% in 1970 and 57.8% in 2020; the values before 1970 are estimates from census records, from IPUMS USA.
Every justice of the Supreme Court was White until Thurgood Marshall joined the Court in 1967; since 2022, six of the nine justices have been White.
For the Cabinet, the figure shows the minimum White share during each administration. It was 100% in most administrations before 1977. The lowest minimums were 44% under Biden, when 14 of 25 cabinet and cabinet-level posts were held by Black, Hispanic, Asian and Native American people at the same time, from October 2022 to March 2023, and 61% in Obama’s first term. In Trump’s second term, the minimum was 78%, from March 24 to April 20, 2026; since Tulsi Gabbard left in June 2026, it has been 87%.
The next figure shows the White share of employees, professionals, and managers at companies that report to the EEOC, and of Fortune 500 CEOs.
Figure 33:corporate_leadership_white — Percent non-Hispanic White among all employees, professionals, and managers at companies that report to the EEOC (1966–2013: total minus the African American, Hispanic, Asian and Pacific Islander, and American Indian tables, and, from 2007, minus the share reporting two or more races; 2014–2023: White, not Hispanic, from the public-use files), and among the CEOs of Fortune 500 companies (purple; 500 minus the Black, Hispanic, and Asian and Pacific Islander CEOs serving on June 1 of each year). Through 2013, “Officials and Managers” is one category (gray line); from 2014 it is split into mid-level and senior managers. Shaded gray: the non-Hispanic White share of the U.S. population (Census from 1970; IPUMS USA estimate for 1960; each census holds until the next).
In 1966, 88.6% of employees at EEOC-reporting companies were White; in 2023, 54.9% were, a little below the White share of all employed people, which was 58.1% in 2025. Among professionals, the percentage fell from 96.5% to 63.7% between 1966 and 2023, and among “Officials and Managers,” from 98.1% to 79.1% between 1966 and 2013. In 2023, 68.4% of mid-level managers and 80.4% of senior managers were White.
The White share of Fortune 500 CEOs was 99.2% in 1995, 95.2% in 2013, and approximately 88% in 2026. Because these numbers come from subtraction, any CEO missing from the other lists is counted as White, so they are upper bounds — but probably not by more than a few CEOs.
Outcomes¶
The race chapter compares the unemployment, income, and housing of White, Black, and Hispanic Americans. This section shows the same measures with Asian Americans included[1], and adds college degrees. In these figures, White, Black, and Asian are non-Hispanic, and Asian includes Pacific Islanders; Hispanic is any race. The surveys don’t identify Asian people separately until 1985 (housing) or 1988 (the others).
Figure 34:unemployment_by_group — Unemployment rate, annual average, people 16 and older, by race and Hispanic origin. White, Black, and Asian or Pacific Islander are non-Hispanic; Hispanic is any race. Computed from monthly CPS microdata (IPUMS CPS); Asian from 1989.
From 1989 to 2025, the unemployment rate of Asian Americans was within 2 percentage points of the White rate every year, and within 1 point in most years. The largest difference was in 2020, when it was 8.5% for Asian Americans and 6.6% for White Americans. In 2025, the rates were 3.5% for White, 3.7% for Asian, 5.1% for Hispanic, and 6.9% for Black Americans.
Figure 35:income_by_group — Median household income, in thousands of 2025 dollars, by race and Hispanic origin of the householder. White, Black, and Asian or Pacific Islander are non-Hispanic; Hispanic is any race. Computed from CPS Annual Social and Economic Supplement microdata (IPUMS CPS); Asian from income year 1987.
Median household income was higher for Asian households than for any other group in every year from 1987, when it was $82,000, compared with $72,000 for White households. In 2025 it was $125,600 for Asian households, $96,500 for White, $73,000 for Hispanic, and $60,000 for Black households.
Figure 36:inadequate_people_by_group — Percent of people living in units the Census Bureau classifies as moderately or severely inadequate, by race and Hispanic origin of the householder. White, Black, and Asian or Pacific Islander are non-Hispanic; Hispanic is any race. American Housing Survey, 1985–2023.
In 1985, 5.6% of Asian Americans lived in inadequate housing, about the same as White Americans (5.7%); the percentages were 18.5% for Hispanic and 23.1% for Black Americans. In 2023, they were 3.1% for Asian, 3.9% for White, 6.0% for Hispanic, and 7.6% for Black Americans.
Figure 37:crowded_people_by_group — Percent of people living in homes with more than one person per room, by race and Hispanic origin of the householder. White, Black, and Asian or Pacific Islander are non-Hispanic; Hispanic is any race. American Housing Survey, 1973–2023; Asian from 1985.
Crowding is the exception: in 1985, 19.5% of Asian Americans lived in homes with more than one person per room, compared with 23.6% of Hispanic, 14.0% of Black, and 3.3% of White Americans. By 2023, the percentages were 12.8% for Hispanic, 5.2% for Asian, 4.8% for Black, and 2.3% for White Americans.
Figure 38:bachelors_25_34_by_group — Percent of people 25 to 34 years old with a bachelor’s degree or more, by race and Hispanic origin. White, Black, and Asian or Pacific Islander are non-Hispanic; Hispanic is any race. Computed from CPS Annual Social and Economic Supplement microdata (IPUMS CPS), survey years 1971–2026; Asian from 1988. Through 1991 the CPS asked about years of school (four years of college counted here); from 1992 it asks about the highest degree.
Among people 25 to 34 years old, the percentage with a bachelor’s degree or more increased in every group. For White Americans, it went from 19.2% to 48.0% between 1971 and 2026; for Black Americans, from 6.6% to 31.1%; and for Hispanic Americans, from 4.7% to 27.8%. For Asian Americans, it went from 44.5% to 75.7% between 1988 and 2026.
Unemployment, income, and college degrees are computed from Current Population Survey microdata, distributed by IPUMS CPS; housing is from the American Housing Survey.
- Ruggles, S., Flood, S., Sobek, M., Backman, D., Cooper, G., Drew, J. A. R., Richards, S., Rodgers, R., Schroeder, J., & Williams, K. (2025). IPUMS USA: Version 16.0. Minneapolis, MN: IPUMS. 10.18128/D010.V16.0
- Flood, S., King, M., Rodgers, R., Ruggles, S., Warren, J. R., Warren, D., Breton, E., Cooper, G., Drew, J. A. R., Richards, S., Van Riper, D., & Williams, K. C. W. (2025). IPUMS, Current Population Survey: Version 13.0. Minneapolis, MN: IPUMS. 10.18128/D030.V13.0