WHAT'S IN A NAME? PARENTS AND TEACHERS SHOULD KNOW

Date: Saturday, April 16, 2005

You may recall a small stir last year over an economics paper that asked, "Are Emily and Greg More Employable than Lakisha and Jamal?" The authors, Marianne Bertrand and Sendhil Mullainathan, sent out matched pairs of resumes in response to classified job ads, half with stereotypical black names and half with white.

The presumably white applicants got 50 percent more call-backs than the black applicants, leading the researchers to conclude, "Differential treatment by race still appears to still be prominent in the U.S. labor market."

Very likely so, although there was large variation within racial groups as well; "Brad" got well more than twice as many call-backs as "Todd." And alas for poor "Emily"; despite her position in the paper's title, she got fewer calls than four of the nine African-American female names did.

But a recent paper for the National Bureau of Economic Research by David Figlio indicates that something much subtler is at work. Figlio, from the Department of Economics of the University of Florida at Gainesville, used data from a large Florida school district, not identified for reasons of confidentiality, but including more than 55,000 children in 25,000 families from 1994-'95 through 2000-'01. (The paper, "Names, Expectations and the Black-White Test Score Gap," can be downloaded from nber.org, but it costs $5.)

"I suggest," he says, "that teachers may use a child's name as a signal of unobserved parental contributions to that child's education, and expect less from children with names that 'sound' like they were given by uneducated parents. These names, empirically, are given most frequently by blacks, but they are also given by white and Hispanic parents as well."

In fact, the Emily paper lends some support to that observation. On average, a little over 60 percent of the mothers of Massachusetts girls with the typically black names used in the experiment had at least a high-school degree, while for the white names it was a little over 90 percent.

Figlio continues, "Comparing pairs of siblings, I find that teachers tend to treat children differently depending on their names, and that these same patterns apparently translate into large differences in test scores."

Unlike the Emily experiment, where people deciding whom to call were merely assuming they knew the race of the applicant, teachers know their students individually. And if they treat siblings differently, race cannot be the primary reason.

Figlio identifies four attributes of names associated with low socioeconomic status, including whether they have certain prefixes ("lo-" or "qua-"), suffixes ("-isha"), apostrophes or an unusual number of low-frequency consonants. The more attributes a given name has, the more likely it is that the person was born to a mother without a high school degree or into an impoverished family. Black children are more likely to have such names, but not overwhelmingly so.

In Figlio's data, "a boy named 'Damarcus' is estimated to have 1.1 national percentile points lower math and reading scores than would his brother named 'Dwayne,' all else equal, and 'Damarcus' would in turn have three-quarters of a percentile ranking higher test scores than his brother named 'Da'Quan.' " Those may not sound like huge differences, but they are comparable to the differences seen between names given almost exclusively to white or to black children.

Figlio considers not only test scores, but also the likelihood that a child will be promoted to the next grade (holding test scores constant) and the likelihood that the child will be referred to a school's program for the gifted. Children with lower-status names appear to be held to lower standards for promotion than their siblings, and are less likely to be considered gifted.

For Asian students, however, the pattern is different; those with obviously Asian names are more likely to be recommended for gifted programs than those with more Americanized names, and they have higher math test scores. (Reading scores are higher too, but the difference is not statistically significant.) There's no relation between name and promotion status in Asian families, Figlio observes, "but this is due to the fact that there exists very little variation in promotion status within Asian families."

His results show smaller negative effects for low-status names in schools with larger numbers of black teachers or students. Teachers there "perhaps form fewer preconceived notions about children purely on the basis of their names, and do not adjust their expectations based on names as much as they may in schools where contact with black students and peers is more limited."

Figlio estimates that name expectations may account for 15 percent or more of the gap between black and white test scores. Parents can't reasonably be expected to change the way they name their children, although most parents do consider the likely effects of a child's name and this is one they probably haven't thought of.

But it ought to be possible to change how teachers behave; indeed, it might be enough just to make them aware that this is a possibility they should be prepared for, so they can counteract it. After all, isn't "high expectations for every child" supposed to be the rule rather than the exception?