Ronald Bailey (Hit & Run) writes:

What Is Diversity Anyway?

"The white populations of the District, Arlington and Alexandria have grown this decade even as the region's outer counties have grown more diverse, according to new census estimates," according to a story in yesterday's Washington Post. As a part-time resident of DC, I was curious about the Post's take on the idea of what constitutes increasing or decreasing diversity in any community. The Post noted that the percentage of whites living in town rose from 28.2 percent in 2000 to 30.3 percent in 2004. My puzzlement is whether this represents an increase in "diversity" or not? Or as the Post story seems to imply, is "diversity" maximized when no white people live in a community at all? Just wondering.

SDM (mail):
While measures of racial diversity are numerous, urban economists traditionally adopt the Herfindahl Index, an index which weights deviations from perfect equality by the square of the distance. Applied in this context, if for example four ethnic groups are assumed (Caucasian, African-American, Hispanic, Asian) then deviations are off of a 25% share of the population. This methodology is used by, I believe, US News to discuss the most racially diverse colleges. The increase in the white population share in particular DC regions would not necessarily imply a reduction in diversity, as the shares of other racial groups are necessary.

As an extension, this method may be used to define 'diversity' as devations from any other defensible share apportionment - say the national racial shares from the 2000 Census, or an ideological share. The Washington Post's statement on diminished diversity may be interpreted in this second way - the paper argues the appropriate share for caucasians is zero, and aggregates all other races into an "other" category, producing exactly their statement that diversity has decreased.
8.12.2005 7:39pm
steve bartin (mail) (www):
Diversity is whatever those in power claim it is.You'll notice that some groups don't make the standard diversity count.
8.12.2005 9:31pm
Cynicus Prime (mail) (www):
The issue of diversity is going to get very interesting as the white population loses its majority over the next 50 years. We honkey crackers in Texas officially became a minority this year. I'm about ready to start petitioning for affirmative action admission to grad school.
But more seriously, it seems natural that maximum diversity is achieved when every available ethnic group makes up an equal portion of the total population. Simplified, if out of 10 people, there is 1 white, 1 black, 1 asian, 1 arab, 1 south asian, 1 native american, 1 european, 1 latino, 1 homosexual, and 1 transgender, then maximum diversity has been achieved. Introduce more ethnic groups into the equation, and the other groups' share must necessarily decrease to maintain diversity equilibrium. Otherwise we've got an upstart hegemony on our hands. And that just won't do in our multiculti society, now will it?
8.12.2005 10:48pm
Mark (mail):
"Simplified, if out of 10 people, there is 1 white, 1 black, 1 asian, 1 arab, 1 south asian, 1 native american, 1 european, 1 latino, 1 homosexual, and 1 transgender"

But that only comprises 6 people, Cynicus Prime. You forgot to mention that in your list there is one transgendered homosexual white european latino.

Ethnicity, national origin, race, sexual preference, and gender aren't mutually exclusive categories. Using SDM's measure, you can calculate racial diversity or ethnic diversity, but you won't necessarily get the same answer for both. A population can be ethnically very diverse, and racially much less so, for example.
8.12.2005 11:15pm
Duncan Frissell (mail):
Which reminds me of yesterday's Texas Now a Majority-Minority State which made Drudge, Paul Harvey, etc.

It was full of statements like: Texas has become the fourth state to have a non-white majority population, when the author meant to say non-hispanic white population (since 85% of hispanics are white. Paul Harvey even implied a Hispanic majority even though what is being discussed is a non-hispanic white plurality (almost a majority).

Those at my lunch table had a hard time understanding the difference between the phrases non-white and non-hispanic white and the difference between a majority and a plurality.

Several years ago, the Census Bureau reported that the incomes of blacks and whites in the Borough of Queens (Queens County) New York City were now the same. Unmentioned was the contributions of white Hispanics to the equation.
8.12.2005 11:18pm
Eugene Volokh (www):
Whether there's a difference between "white" and "non-Hispanic white" depends on the definition of "white." The census and (I believe) many demographers treats white as a race and Hispanic as an ethnicity, and thus distinguishes "white" and "non-Hispanic white."

But in ordinary language, "white," "Hispanic," "black," "Asian," and "American Indian" have often been used as mutually exclusive categories (setting aside those who identify as multiracial). That includes the ordinary language of people from white racists, to people who support race preferences for blacks and Hispanics, to ordinary people who just haven't followed the demographic definitions. In my experience, when all these groups say "white" they generally mean "non-Hispanic white." That's just the dominant colloquial, as opposed to the technical, meaning of the term (though it may be changing over time).
8.12.2005 11:25pm
I am white. I am also English, Irish, German, Mediterranean, and American Indian. Who brings more diversity to an employer, school, or neighborhood; me, or a black who can trace his pure, undiluted ancestry back 500 years?
8.13.2005 12:03am
Robert Schwartz (mail):
If you have to ask it proves you are a troublemaker and a racist:-)
8.13.2005 2:05am
Anthony Argyriou (mail) (www):
Quite a lot of "White" Hispanics are actually mixed-race White and Native American. Many "Non-White" Hispanics are pretty close to full-blooded Native American. However, since "Native American" in the US context implies the aboriginal people of the area north of the Rio Grande, I'm not sure how someone who is nearly purely Mayan or Tolima or Quechua would be counted in the standard racial classification system, if "Hispanic" is not counted as a race.
8.13.2005 3:30pm
Bruce Hayden (mail) (www):
But then, if you counted Hispanic as a race, how could you ever define it. It ranges from pure Caucasian through pure American Indian (to get around the Native-American problem).

Of course, the insanity is that many Hispanics have a lot more Indian blood than many legitimate Native-Americans. But that is because the former is typically defined based on culture, whereas the later based on proven descent from a certified Indian tribe in the U.S.

So, if you could detect race through genetics, which is not that easy, you would find a lot of Hispanics being predominently Indian, while a lot of Native Americans being predominently Caucasian.

Of course, you have this problem overall. A friend has a daughter who legitimately qualified as African-American for college admission, despite being no more than 30% Black, and at least 70% White (and grew up in an almost exclusively white community). Indeed, racially, a lot of the "African-Americans" today have less than 50% Black ancestry.

Things are only getting worse through intermarriage. I know a lot more kids who are half Japanese or Chinese than full blooded. The problem they face is that if their father was Asian, they are automatically considered such by their last name, but if it was their mother, they are automatically considered White.
8.14.2005 12:44am
gopher (mail):
SDM: While measures of racial diversity are numerous, urban economists traditionally adopt the Herfindahl Index, an index which weights deviations from perfect equality by the square of the distance. Applied in this context, if for example four ethnic groups are assumed (Caucasian, African-American, Hispanic, Asian) then deviations are off of a 25% share of the population.

Thank you, SDM, that's interesting, and, if accurate, it clarifies to me, a mathematician, that "urban economists" are quite completely insane. To assume a priori equipartition leads to weird and self-refuting results, such as "Planet Earth is not racially diverse" since Planet Earth does not contain precisely 20.000% each of {Australoid, Capoid, Caucasoid, Mongoloid, Negroid} (or whatever).
8.14.2005 1:11am
I seriously doubt white Americans will actually let America become minority nonwhite. The next 5-10 years, with the immigration issue at a boiling point and evolutionary science advancing, will be interesting.
8.14.2005 2:43am
tioedong (mail) (www):
Diversity as 35 percent? OK.

However, who "decides" who is what? I mean, my daughter, who lived in Arlington, is 1/2 Filippina with some Spanish mixture from her father and 1/2 white from German/Swedish mother married a WASP...what does that make their kids? Before you answer, consider this: Their second son is the spliting image of my husband....

Now mix into that my adopted Colombian son, who married a girl from Tennessee with Cherokee blood...

FYI: I refused to place "race" on questionaires since they started asking in the 1970's...
8.14.2005 2:52am
In corporate HR jargon, "diverse" is the polite way to say "minority". It has nothing to do with the Webster's definition of "diversity" at all. I deposed a "Chief Diversity Officer" of a F100 company and caught on when he started used the term to refer to a single person - "she is diverse" - to say "either non-white or handicapped."
8.15.2005 10:28am
If you have in a room a black person who is part white and has passed for white her entire life, a similarly situated white-hispanic, and a similarly situated white-asian, and all these individuals were raised in a white suburb by your typical Republican parents, do you have diversity? Heck, toss in a white gay republican to round off the mix.

Diversity is much more than just where your genes come from.
8.15.2005 10:29am
Larry (mail) (www):
From the plig:

a group of people who think and vote like me but are of a different race or gender; process of admitting upper-middle class blacks who blew off expensive prep course because too busy with frat activities and thus bombed the LSAT getting into non-TTT schools
8.15.2005 10:49am
james (mail):
Eventually the U.S. will obtain a large enough population of ethnically Hispanic peoples that they will be treated just like every other ethnic group that is mostly racially white . As a nation full of multiple racial and ethnic groups, it is a mistake to focus on race and "diversity". It encourages one or more groups to view their fellow citizens as outsiders.
8.15.2005 11:39am