How 38 counties that represent a truly diverse America voted

What effect did diversity have on the 2016 presidential election?

Trempealeau County in Wisconsin—a state that broke for the first time for a Republican in more than three decades—was identified in a recent Wall Street Journal article as a place where Donald Trump’s anti-immigrant rhetoric was taking hold. Trempealeau County is a mostly white area, but, like other counties in the Journal’s rundown, there was something special going on: burgeoning diversity.

Fusion identified the five most increasingly diverse counties in America, including Trempealeau County, as well as the five most diverse and least diverse counties in America, to track how those places are voting on election night.

The result is a list of 38 counties—both rural and urban, large and small, including a number in swing states—that represent the face of diversity, and lack thereof, in America today. This is the landscape amid which Donald Trump was elected president.

Below, you can see a map for election returns from all counties in America, including the counties selected for special examination by Fusion highlighted in orange.

Map data: Explore the map above to see election results for the 38 counties we identified as representative of a diverse America. Fusion’s Diversity Index scores counties based on the probability that any two residents chosen at random will belong to different racial or ethnic groups. To learn more about how we calculated Fusion’s Diversity Index, see below.

(All voting results as of 11-9-00 05:40 AM EST. Source: Associated Press)

America’s most diverse counties

In Tuesday’s election, the most diverse counties in America voted, as one might’ve expected, for Hillary Clinton by more than a two-to-one margin. And third-party candidates did surprisingly well in two Hawaii Counties! (Hawaii and Maui Counties, in Hawaii, are both in the top-five most diverse counties because of their large native Hawaiian population, as well as the large number of immigrants from nearby countries in the Pacific.)

Trump garnered about a quarter of the votes in Queens, in New York City, the most diverse place in the United States. But in Alameda County, CA, where Oakland is, he got a mere 15%.

Average percent vote for America’s most diverse counties:

Hillary Clinton: 67%
Donald Trump: 27%
Third-party candidates: 6%

Counties with a majority Hispanic population

Fusion also broke down the most diverse counties in America by ethnic and/or cultural groups. Below are the five most diverse counties with majority Hispanic populations. In these areas, Clinton fared almost as well in aggregate as she did in the top-five diverse counties, but she nearly split votes with Trump in California’s San Bernardino County.

Average percent vote for selected majority Hispanic counties:

Hillary Clinton: 65%
Donald Trump: 31%
Third-party candidates: 3%

Counties with a majority black population

In majority black counties, Clinton did very well—but not as well as Barack Obama did among the general African American population in his two runs for president.

Average percent vote for selected majority Black or African American counties:

Hillary Clinton: 80%
Donald Trump: 17%
Third-party candidates: 3%

Counties with a majority Asian population

There are no counties in America where Asian residents represent more than 50% of the population. Instead, Fusion’s analysis focused on three counties where the Asian population is greater than 30 percent. In these counties, Clinton garnered a whooping three-quarters of the votes cast.

Average percent vote for selected majority Asian counties:

Hillary Clinton: 73%
Donald Trump: 21%
Third-party candidates: 6%

Counties with a majority of American Indian and Alaska Native voters

Although American Indians live in some of the most diverse communities in the U.S., they did not turn out for Hillary Clinton the way some other diverse areas have. Trump, on the other hand, has a long history of making life difficult for Native Americans.

Average percent vote for selected majority American Indian and Alaska Native counties:

Hillary Clinton: 63%
Donald Trump: 28%
Third-party candidates: 9%

America’s least diverse counties

Fusion also took a look at some of America’s least diverse counties. How were they reacting to the election? The answer is: in a very pro-Trump way. Trump ended up getting almost 90% of the vote in one Kentucky county!

Average percent vote for America’s least diverse counties:

Hillary Clinton: 18%
Donald Trump: 79%
Third-party candidates: 4%

Average percent vote for selected majority white counties:

Hillary Clinton: 57%
Donald Trump: 38%
Third-party candidates: 4%

Greatest change in diversity

The demographics of our nation are ever changing. To get a sense of which places are experiencing the greatest change, we compared counties’ diversity across the years 2000 through 2015. The five counties that experienced the greatest increase in diversity since 2000 are highlighted below.

You might not be surprised to learn that these quickly-diversifying counties are places where Trump was able to establish leads—if not dominance—on Tuesday. While many majority white counties went for Clinton, it was less diverse counties, and rural places experiencing a wave of new faces for the first time, that bolstered Trump’s win. In other words, this election may have been less about race and more about differences in attitudes between those who are used to living in diverse communities versus those for whom living in a diverse community may feel new or threatening.

Average percent vote for selected counties with the greatest increase in diversity since 2000:

Hillary Clinton: 29%
Donald Trump: 67%
Third-party candidates: 4%

About Fusion’s Diversity Index:

To select counties that represented more or less diverse populations we created a Diversity Index. Fusion’s Diversity Index is based on population estimates from the U.S. Census Bureau. For each county, the index measures the percent chance that any two residents chosen at random will belong to different racial or ethnic groups. While there’s no perfect way to define which groups count as the same or different, we used the same seven groups that the U.S. Census Bureau uses for this purpose. For example, although it is Fusion’s preference to use Latino, Latina and Latinx rather than Hispanic, to avoid confusion, when discussing Census data we refer to racial and ethnic groups using the norms of the Census Bureau — in this case, Hispanic. Counties that are less diverse score lower on the index; counties that are more diverse score higher. The minimum score is  2% and 76% is the maximum. The average U.S. county is at about 30%.

To select counties heavily populated by single racial or ethnic groups, we used the five major racial and ethnic groups identified by the Census. We identified counties with a majority population of a single group (more than 50%) and ranked them by the size of the group’s population to choose the most representative counties. Only counties with a total population greater than that of the lower quartile for all counties (>10,986) were included.

Data Source: Population estimates: U.S. Census Bureau Population Estimates as of 2015. Voting results: Associated Press Projections  by county.

Daniel McLaughlin is a creative technologist exploring the 2016 presidential election. Before joining Fusion, Daniel worked at the Boston Globe and graduated from MIT with a BS in urban studies and planning.

Kate Stohr is a data journalist and community builder based in San Francisco, CA.