Names, race, and economists
Last time, I talked about economist Steven Levitt's take on baby name fashion in the book Freakonomics. Names have suddenly become a hot topic with economists -- they seem to have wrenched the field out of the hands of psychologists and sociologists. Their single hottest subject is the "consequences" of having a distinctively black name. As the title of one paper asks, "Are Emily and Greg More Employable than Lakisha and Jamal?"
In 2003, a pair of researchers from Harvard and the University of Chicago sent out hundreds of resumés with either white-sounding or black-sounding names. The "white" resumés received 50% more callbacks for interviews, a seemingly dramatic consequence. Yet that same year, a different pair of researchers from, yes, Harvard and the University of Chicago, looked at life outcomes of people based on birth certificate data from the State of California. (The certificates indicated the parents' education level and other socioeconomic cues.) This study found no independent effect of distinctively black names .
As it happens, one of the authors of the second study was Levitt, who summarizes the result in Freakonomics. He dismisses the resumé study and all other field simulations, claiming "the audit studies can't be used to truly measure how much a name matters, the California names data can."
I read both papers when I was researching my book, The Baby Name Wizard. My initial reaction was that both painted names with a rather broad brush. All "black" names aren't created equal. Take two examples from Levitt's "blackest names" list, DeShawn and Terrance. Both may send the same skin-color signals, but they send very different cultural signals. (Just as, say, Beatrix and Shyanne are equally white names that send different cultural signals.) Look at Emily and Lakisha, from the title of the resumé paper. Emily, an old familiar classic, is the #1 name in America; Lakisha, an invention of the 1970s, has never cracked the top 1000. How can you compare such wildly different names and expect a pure reading on the effects of race?
Enter an economist from neither Harvard nor Chicago: David Figlio of the University of Florida. In 2004 Figlio looked at children in a large Florida school district, tracking signs of teachers' expectations of individual students: whether children were promoted to the next grade, for instance, or recommended for gifted programs. He rated names both for their racial makeup and their socioeconomic makeup. (By analyzing variables like parents' education level and economic status, he found that certain name characteristics were typical of a disadvantaged household.) And he focused especially on pairs of siblings, who had the same family background but often very different styles of names.
It's a remarkable bit of research wizardry, teasing apart the effects of names, race and socioeconomic status -- even the effect of the family that raises you. Figlio's findings showed that, indeed, all "black" names are not treated the same. A name like Dwayne, which was strongly African-American but carried no socioeconomic markers, didn't affect teachers' expectations. But a name like Da'Quan, with multiple signals of economic status, did. Teachers, consciously or not, drew inferences about the child's background and potential based on these naming signals. In Figlio's data, a pair of brothers named Dwayne and Da'Quan could expect subtly different treatment in school, which translated into different levels of scholastic success.
It's a useful demonstration for prospective name-and-number-crunchers that names carry a rich web of connotations. People are extremely sensitive to names' nuances: history, popularity, spelling, punctuation...everything speaks to our mental models of names and culture. There's a reason that parents agonize for months over name choices. It's not just a black or white question.