(September 2000) Racial discrimination is a topic that undergraduate sociology students expect to study. Few of them, though, expect to encounter it as part of a class project. So when students in my research methods course conducted a housing audit, they got more out of it than they had anticipated.

I came up with the idea of using undergraduates to carry out a housing audit partly because, at my university, pressure to involve undergraduates in research is mounting. But I wanted to go beyond assigning research for research's sake, which students see through and which doesn't generate data for social scientists. The audit project proved so successful that it could be applied widely in classrooms throughout the country to produce valuable data, sharpen students' research skills, and teach them a life lesson.

For many years, I have followed the results of racial housing audits. The design of these audits is simple: Teams of white and black auditors are assigned similar identities and characteristics. Realtors are usually sampled from some listing, such as the real estate section of a major newspaper. The auditors then visit selected agents to inquire about the availability of housing. After the encounter, the auditors fill out forms describing their treatment. When all agents have been contacted, the investigator compares the findings of black and white auditors to see whether they were treated differently. Systematic differences in treatment are taken to reflect racial discrimination.

My plan was to have students do a housing audit to study racial discrimination over the telephone. Many African Americans speak a dialect known by sociolinguists as Black English Vernacular, and even more speak standard English with an accent that most listeners identify as "black." If people can attribute race by voice alone, phone-based discrimination can occur.

The analysis revealed phone-based racial discrimination. Compared with whites, African Americans were less likely to speak to a rental agent (agents could screen calls using answering machines), less likely to be told of a unit's availability, more likely to pay an application fee, and more likely to have credit mentioned as an issue. These racial effects interacted with and were exacerbated by gender and class. Typically, students posing as lower-class blacks (speakers of BEV) experienced less access to rental housing than those who represented middle-class blacks (speakers of BAE), and black females experienced less access than black males (see figure below). In general, students playing the part of lower-class black females were most disadvantaged.


Percent of Callers Who Reached Agent and Were Told Apartment Was Available

Apartment Renting

Urban Affairs Review, forthcoming.


Students enjoyed participating in the study and writing the papers, and they learned something not only about research, but also about the reality of race in America (see "For More Information"). After the end of the term, I worked with the postdoctoral fellow to refine the analyses, and together we wrote a paper that is forthcoming in Urban Affairs Review. Our experience suggests that telephone audits constitute a potentially cheap, easy, and efficient way of measuring and studying processes of racial discrimination in urban housing markets. All that is needed to accurately measure racial discrimination is access to a local newspaper or rental guide, a telephone, and people capable of using linguistic styles associated with race and ethnicity.


Douglas Massey is Dorothy Swaine Thomas Professor and chair of sociology at the University of Pennsylvania.


For More Information

Students' Reactions

The reaction of the students to the experience varied:

  • White students were shocked at the degree of discrimination against blacks. Most said they wouldn't have believed it beforehand.
  • Black students were not surprised. It simply confirmed their experience.
  • The women in the class were most shocked. Although black women expected the racial discrimination, they were dismayed at how much worse they were treated even than black men, and they were horrified at some remarks made to them, which hinted at sexual promiscuity, drug use, and welfare receipt. No other group had to endure this kind of stereotyping. White women were shocked, and then outraged, that they were treated differently than white males. Like many women of their generation, they assumed that gender bias was a thing of the past.