A while back, a writer I like got in trouble with a lot of people who would otherwise be fans over something she wrote:
In it, Dr. Harris-Perry (who I follow on Twitter) lays out an argument for why white voters, who supported Barack Obama in the first election, may be abandoning him now at a greater rate than they did President Clinton in the 90′s – despite the many political and situational similarities between the two. Given that so many of the ostensible reasons for withdrawing support are balanced between the two administrations, racism may explain, at least in part, any differences in voter support and approval. It’s hard to argue that race and racism have not played a role in this particular presidency far more than in others.
Because I liked both this article and a related one that more closely explored the racial attitudes of Bill Clinton more specifically and liberals more generally, I fired a quick message to Dr. Harris-Perry in support, because I knew that she was taking quite a bit of flack for her audacious temerity to suggest that liberals weren’t the immaculate paragons of fairness that we make ourselves out to be. Basically, just a “hey, I liked your piece in the Nation.”
The problem, of course, is that racism is notoriously difficult to pin down as a single causal factor. Because we’ve gotten so good at obfuscating it through clever language and self-inflicted racial blindness, it’s particularly challenging to detect positively. Usually you have to try and remove all other potential causal factors and then measure the size of a racial disparity and say “well this has to be racism, because what else could it be?” That is far less psychologically satisfying than being able to point at something definitively, objectively racist and say “look, there’s your monster”.
Which is why I find this a particularly fascinating exercise:
Can we really quantify racial prejudice in different parts of the country based solely on how often certain words are used on Google? Not perfectly, but remarkably well. Google, aggregating information from billions of searches, has an uncanny ability to reveal meaningful social patterns. “God” is Googled more often in the Bible Belt, “Lakers” in Los Angeles.
The conditions under which people use Google — online, most likely alone, not participating in an official survey — are ideal for capturing what they are really thinking and feeling. You may have typed things into Google that you would hesitate to admit in polite company. I certainly have. The majority of Americans have as well: we Google the word “porn” more often than the word “weather.”
And many Americans use Google to find racially charged material. I performed the somewhat unpleasant task of ranking states and media markets in the United States based on the proportion of their Google searches that included the word “nigger(s).” This word was included in roughly the same number of Google searches as terms like “Lakers,” “Daily Show,” “migraine” and “economist.”
The cool thing about this is that Google functions like the untramelled id of the internet. It is used away from the prying eyes of social norms, when people are their most natural, disgusting selves. Also, the word ‘nigger’ isn’t exactly a word that isn’t unequivocally racist. I can think of a handful of circumstances wherein someone would we looking for information about niggers that doesn’t necessarily reveal a hatred for black people, but the frequency of those types of occurrences is vanishingly small.
So now that we have a measure of racism, and some data to go with it, what do we find?
Once I figured out which parts of the country had the highest racially charged search rates, I could test whether Mr. Obama underperformed in these areas. I predicted how many votes Mr. Obama should have received based on how many votes John Kerry received in 2004 plus the average gain achieved by other 2008 Democratic Congressional candidates. The results were striking: The higher the racially charged search rate in an area, the worse Mr. Obama did.
Consider two media markets, Denver and Wheeling (which is a market evenly split between Ohio and West Virginia). Mr. Kerry received roughly 50 percent of the votes in both markets. Based on the large gains for Democrats in 2008, Mr. Obama should have received about 57 percent of votes in both Denver and Wheeling. Denver and Wheeling, though, exhibit different racial attitudes. Denver had the fourth lowest racially charged search rate in the country. Mr. Obama won 57 percent of the vote there, just as predicted. Wheeling had the seventh highest racially charged search rate in the country. Mr. Obama won less than 48 percent of the Wheeling vote.
Add up the totals throughout the country, and racial animus cost Mr. Obama three to five percentage points of the popular vote. In other words, racial prejudice gave John McCain the equivalent of a home-state advantage nationally.
Now you don’t have to be an economist to see that there are some weaknesses in this methodology. First, this only measures overt racism, not total racism. People in New York and Pennsylvania might be searching for ‘nigger’ less often because even they don’t use that word (despite having strong anti-black feelings). John Kerry and Barack Obama are not the same person, nor did they run on identical platforms, so using him as a comparison has limited utility. There is no way to measure anti-white racism, as there is no real equivalent to ‘nigger’ that can be used against a white person.
All that being said, while we may not know the exact magnitude of the effect that racism played in the last presidential election (or will in the next one), we can certainly say that it is likely to exist.
Like this article? Follow me on Twitter!
A very interesting study. I can see it has its flaws methodologically* but as a little pointer I think it’s not bad.
I also like that the author calls her subjects “Mr”, as we do to politicians in the UK. Our PM is usually referred to as “Mr Cameron”.
Or “Dave”.
*really a word?
Oh, great, next you’re going to tell me all sorts of Liberals are latent or even overt misogynists, too.
The Google info is very interesting. I’m sure, with some sweet, sweet grant money, one could do a much more thorough analysis of entire phrases and get a more robust picture of e.g. racial attitudes based on both search terms and click-through. Was someone searching ‘niggers’ to do a reverse lyric search for “Still D.R.E.” or to find material published by a neo-Nazi hate group?
I’m not so sure this is true, based solely on the reverse-lyric searches (which I do all the time and assume others do) and prevalence of the word ‘nigger’ or variations in rap tracks. One would HAVE to look at full phrases and which results people clicked through. Of course, I have no trouble believing that racism is still really widespread and influences voting patterns; I do think there are some potentially-severe confounding factors in the Google search data analysis.
I was thinking that too, that it’s probably not vanishingly small. It is still probably a good proxy measurement though.
From the article:
Also, how do you figure that ‘searching for rap lyrics’ translates into a 3 point differential? 😛 I doubt that West Virginia is the rappin’-est state in the union.
This detail alone is alarming. It’s enough evidence to suggest actively racist thought is desired by substantially more than just the members of groups identified by Southern Law Poverty Center. I know that would be too strict a definition of “in fact really racist” folks but I didn’t have the scope right in my head.
(I’m very not surprised porn it way up there)
When you said “we” I thought this post would be reporting analysis of Canadian politics. Still, interesting article. Carry on.
Oh no, I just meant that ‘we’ could use this kind of methodology as a proxy for measuring population levels of racism.
From what I’ve read, I think that people (especially young white people) had quite lofty expectations of what “change” specified, and when that wasn’t delivered, and in fact barely even fought for, in their eyes, the support goes down.
A good example, I think is the war on drugs, which to a lot of people has been amplified even further over the last few years (I’m not sure if this is true or not, but this is the view that a lot of people have).
The other part of it is that there’s a certain entitlement that’s been violated more and more over the last few years…that is that college graduates, especially white and male college graduates, should be able to use that degree to get a “good” job. As such, it’s this entitlement that is taking it more on the chin, than other subgroups who have already internalized that quite frankly, poverty is our lot in life and it’s going to take near impossible amounts of political change to change this.
This isn’t to say that racism doesn’t exist, of course, and the latter study listed goes over some potentially very real effects. But, I don’t think it quite gets to the point on why young white voters are quicker than we would expect to abandon Obama.
Definitely, and I think it is sloppy writing on my part that forges the connection between Dr. Harris-Perry’s point and the Google search paper. All I meant is that it is perhaps possible to measure the extent to which racism plays a role in elections through similar search pattern methods, rather than having to merely infer its existence from other outcomes.
This is a really interesting topic. I was pissed off at Bill Clinton, too– even more so at the end of his second term than at re-election time. Of course, it didn’t matter, because I didn’t have enough money to give anyone at the time. So I can’t compare that situation to my current position of voting for Obama but contributing to progressives like Elizabeth Warren instead. I’ve grown increasingly disillusioned with the Democrats over the last dozen years, and I’m tired of being asked to vote for them because they’re not nearly as bad as the other guys. So I’d like to believe that my response to Obama is merely a continuation of that trend. Still, it nags at me that race might play a subconscious role. Every time I hear someone use the A-word (arrogant) to describe him, I wince. It’s just too close to that other word, the one that starts with u.
When I first saw the study to which you linked, I was surprised that the affect of race wasn’t higher than it is. Then I realized just how large it was in comparison to the normal margin of victory. It takes an amazing person to overcome that kind of handicap. For all my complaints Obama, I surely hope he can do it again.
I wish he was able to do more, but his action on gay marriage and gay service in the military, coupled with the health care and the refusal (not total, but comparative to the Repubs) to engage austerity-only policies have convinced me to vote for Obama.
His willingness to draft Hillary for the position of Secretary of State didn’t hurt. She’s done fairly well in that position, in my opinion.
And, even though this was OMG bad (according to the Repubs) in the last election, I’m happy that a black man is in office. I cried for the last election like a baby (seriously, I was sobbing with happiness that someone not an old white dood won), and I’m liable to sniffle again if he wins, because it’s important to break the white doodz monopoly on power.
I’ll cry again for the first Hispanic/Latino president, or the first president of Middle Eastern descent, or the first president of Native American descent or the first president of Chinese, Japanese, Malaysian, Thai, Korean, etc descent. When a woman is elected, I fully intend to lay waste to a crate of tissues. But my happiness will be greatly enhanced by progressive stances on issues.
I am a little pessimistic about his chances in the upcoming election; his presence in the White House is symbolic of the change in power happening everywhere, and as such has been given unreasonable and untenable expectations. That’s likely to contribute to the volume of search results of that abominable word.
+1000