New Wharton, HBS Research Reveals Last Interviewee of the Day Often Fares Worst
A recent research paper co-authored by professors from the University of Pennsylvania’s Wharton School and Harvard Business School reveals that MBA applicants interviewing at the end of a day after a series of strong candidates are more likely to receive a lower evaluation than they would otherwise.
Terming the phenomenon “narrow bracketing,” the researchers conjectured that interviewers charged with evaluating a subset of applicants in a given day tended to try to provide evaluations in keeping with the overall expectations for the entire interview pool.
“For instance, an interviewer who has already highly recommended three applicants on a given day may be reluctant to do so for a fourth applicant,” wrote Uri Simonsohn, an operations and information management professor at Wharton, and Francesca Gino, an associate professor of business administration at HBS. Data from more than 9,000 MBA admissions interviewers supported the prediction.
The paper, entitled “Daily Horizons: Evidence of Narrow Bracketing in Judgment from 10 Years of MBA-Admission Interviews,” was published in the February 2013 issue of Psychological Science. The researchers found that a similarly qualified applicant who interviewed on the tail end of a series of top-scoring applicants got worse than expected evaluations. In the reverse situation, when a similarly qualified applicant interviewed after a group of weaker competitors, that applicant got better than expected evaluations.
Simonsohn and Gino hypothesize that an interviewer may be reluctant to give high ratings to the last candidate in a string of highly-rated candidates because he or she knows that only a certain percentage of individuals are accepted into a program.
“An interviewer who expects to evaluate positively about 50 percent of applicants in a pool may be reluctant to evaluate positively many more or fewer than 50 percent of applicants on any given day. An applicant who happens to interview on a day when several others have already received a positive evaluation would, therefore, be at a disadvantage,” they write.
The problem with this narrow bracketing behavior is that the subsets arbitrarily created by the scheduling of the interviews should not influence the interviewers’ judgments. “While the merit of an MBA applicant may partially depend on the pool of applicants that year, it should not depend on the few others randomly interviewed that day,” Simonsohn and Gino write.
Though they studied MBA applicant interviews specifically, the researchers say that the phenomenon of narrow bracketing is not confined to academic admissions. Indeed, the same dynamic could play out whenever individuals are spreading similar decisions out over multiple days – be they decisions about applicants for loans at a bank or candidates interviewing for a job.
“In any setting where people have to make a large set of judgments that is broken down into a small set on the same day, you might see the same thing,” Simonsohn notes.
So what can applicants do to counteract the narrow bracketing effect? Not much, Simonsohn and Gino say. “There’s no magic in this for the user. You can’t see who you’re competing against and often can’t control the timing of your interview…. When the candidates are spread out over weeks and weeks, your competition is the entire applicant pool and not a subset of that. But in reality, your competition is drawn from two pools — everyone and the other applicants who get interviewed that day.”
The researchers do, however, offer suggestions for low-cost, low-risk ways companies and universities could seek to control for the effect, such as having interviewers enter each applicant’s scores into a spreadsheet or database program that would help them monitor the results of their interviews over time, taking the focus off a given day’s candidates.