Traditional hiring processes can lead employers to rely on imperfect signals when narrowing a candidate pool. A manager with little more than a standard application may lean toward candidates from the best schools or those with impressive professional networks, and that may give undue advantage to those higher on the socioeconomic ladder and exclude candidates who could be better fits.
Conducting in-person interviews with more candidates could in some ways help ameliorate the problem, suggests research by Washington University’s John Barrios, University of California at Santa Cruz’s Laura Giuliano, and Northwestern’s Andrew J. Leone. However, in-person interviews can also lead to unexpected results and raise different biases, according to their research—which was conducted before COVID-19 forced many such interviews online.
The researchers exploited a dramatic shift in how the job market works—or had been working, before the pandemic—for newly minted accounting PhDs seeking research-oriented academic positions. In 2013, the American Accounting Association introduced its Accounting Rookie Camp program. During a two-day conference, university recruiters could mingle with, interview, and listen to research presentations from dozens of PhD students. The process allowed recruiters to obtain more information about a wider swath of candidates and consider some who might not have previously made the cut for in-person interviews. (The 2020 Rookie Camp moved online.)
Barrios, Giuliano, and Leone collected data on the career and educational histories of all 2,270 accounting doctoral graduates from 102 academic institutions in the United States from 2005 through 2015. They boiled the sample down to 1,741 individuals, obtained data on their publications and academic advisers, ranked their degree-granting institutions, and estimated their gender, race, and national origins on the basis of their names.
They find that Accounting Rookie Camp reduced the importance of the ranking of a candidate’s PhD program and adviser connections. The weight accorded to publishing papers in the highest-profile journals also decreased, though overall publishing became more important. That may be because recruiters spent less time noting the prominence of the journal and focused more on the content of the work, the researchers write.
The program seemed to open the door for candidates from lower-ranked schools to obtain higher-prestige placements, and those new hires were shown to increase both the number and quality of publications compared with those hired before creation of the Accounting Rookie Camp program, the research demonstrates. But the new system did nothing to reduce disparities in gender and racial-minority placements, the researchers find.
Moreover, the process may have actually put certain underrepresented groups at a disadvantage, especially those with Chinese surnames, argue Barrios, Giuliano, and Leone. Such candidates experienced a decline in placements at high-ranking universities. It’s possible that the interactions at Rookie Camp changed recruiters’ beliefs about Chinese candidates’ research productivity, or magnified language or culture differences that recruiters viewed negatively, the researchers suggest. Even Chinese students who adopted American-sounding first names, who historically received high-tier placements, performed worse after the implementation of Rookie Camp.
“We saw a leveling of the playing field in the sense that people from Penn State were competing for jobs with people from Columbia, whereas before, you wouldn’t really see interactions between job candidates at certain universities,” Barrios says.
“Academia is a club, and the selection into the club ends up being a function of who is in the club already,” he adds. “You would imagine biases would go away if you are forced to talk to an individual, but this speed-dating format may not be as beneficial for certain candidates where certain biases may be expanded.”
And as the results also apply to early-stage online interviews, there could be unintended hiring consequences as companies do more of those.