Why Unqualified Candidates Get Hired Anyway
Why do businesses evaluate candidates solely on past job performance, failing to consider the job's difficulty? Why do university admissions officers focus on high GPAs, discounting influence of easy grading standards? Francesca Gino and colleagues investigate the phenomenon of the "fundamental attribution error."
People make snap judgments all the time. That woman in the sharp business suit must be intelligent and successful; the driver who just cut me off is a rude jerk.
These instant assessments, when we attribute a person's behavior to innate characteristics rather than external circumstances, happen so frequently that psychologists have a name for them: "fundamental attribution errors." Unable to know every aspect of a stranger's backstory, yet still needing to make a primal designation between friend and foe, we watch for surface cues: expensive pants—friend; aggressive driving—foe.
A new research paper demonstrates that the fundamental attribution error is so deeply rooted in our decision making that not even highly trained people-evaluators, such as hiring managers and school admissions officers, can defeat its effects.
"One of the consequences is that you end up admitting people who should not be admitted, and rejecting people who should not be rejected."
Inflated Applicants: Attribution Errors in Performance Evaluation by Professionals paints a picture of businesses repeatedly promoting or hiring less-qualified managers who benefit simply by being associated with a high-growth group, or universities who weaken their academic achievements by admitting students who have earned high GPAs in high school without considering the ease of the schools' curricula. The research, published this week in the journal PLOS ONE, was conducted by Samuel A. Swift and Don A. Moore, University of California at Berkeley; Zachariah S. Sharek, Carnegie Mellon University; and Francesca Gino, Harvard Business School.
"Across all our studies, the results suggest that experts take high performance as evidence of high ability and do not sufficiently discount it by the ease with which that performance was achieved," the paper reports.
Most research in this area has focused on defining the fundamental attribution error rather than actually observing its practical effects. The researchers decided to investigate whether business executives and admissions officers—people whose jobs require unclouded judgment—are as susceptible to the error as the rest of us. If this were the case, their research could be the crucial first step towards helping businesses and universities make smarter recruitment choices.
"Nobody before us really looked at the consequences for important decisions like selection and admissions," says Gino, an associate professor in the Negotiations, Organizations & Markets unit at HBS. "And you can imagine that one of the consequences is that you end up admitting people who should not be admitted, and rejecting people who should not be rejected."
The results are enough to spook anyone who has ever sent off a résumé or college application: not only were the studies' subjects unable to counteract this correspondence bias, they remained susceptible to it even when warned explicitly of its dangers.
MAKING THE GRADE
Excited to begin a study with such distinctly practical footing, the researchers devised a set of experiments to measure executives' and admissions officers' resistance to this phenomenon.
The first study asked professional university admissions officers to evaluate nine fictional applicants, whose high schools were reportedly uniform in quality and selectivity. Only one major point of variance existed between the schools: grading standards, which ranged from lenient to harsh. Predictably, students from "lenient" schools had higher GPAs than students from "harsh" schools—and, just as predictably, those fictional applicants got accepted at much higher rates than their peers.
"We see that admissions officers tend to pick a candidate who performed well on easy tasks rather than a candidate who performed less well at difficult tasks," says Gino, noting that even seasoned professionals discount information about the candidate's situation, attributing behavior to innate ability.
Similar results can be seen for the second study, in which the researchers asked business executives to evaluate twelve fictional candidates for promotion. In this scenario, certain candidates had performed well at an easier job (managing a relatively calm airport), while others had performed less well at a harder job (managing an unruly airport).
As with the admissions officers, the executives consistently favored employees whose performance had benefitted from the easier situation—which, while fortuitous for those lucky employees, can be disastrous on a companywide scale. When executives promote employees based primarily on their performance in a specific environment, a drop in that employee's success can be expected once they begin working under different conditions, Gino explains.
Having determined that executives and admissions officers are equally likely to commit these logical fallacies, the researchers turned their attention towards ruling out alternative explanations, and verifying these results outside of the lab. The third study in the series asked subjects to look at people's scores in a game, and guess how well they would do in the subsequent round.
"We wanted to see if we could eliminate the bias by providing people with full information," Gino says.
But even when subjects were told that these scores were subjective, and that some games had been easier than others, the participants consistently betted on the players with the highest scores—and were consistently disappointed when their choices went on to lose the second game.
FAILING THE TEST
The implications of this judgment (or lack thereof) are clearly reflected in the real admissions records of graduate business schools, which tend not to consider situational influences in their evaluation of applicants, and are more likely to accept students from schools with lenient grading norms. Applicants with a GPA that is one standard deviation above the mean, 0.17 points, are 31 percent more likely to be accepted than denied or waitlisted, the authors report in the paper. Furthermore, according to the National Association of Colleges and Employers, 58 percent of employers won't even consider candidates with a GPA of 3.0 or lower.
These practices endanger not only the quality of the universities but the futures of the students themselves, according to the researchers.
Gino reports being surprised at how difficult it was to counteract the fundamental attribution error, and, particularly, how strongly its effects could be seen in these records.
"We thought that experts might not be as likely to engage in this type of error, and we also thought that in situations where we were very, very clear about [varying external circumstances], that there would be less susceptibility to the bias," she says. "Instead, we found that expertise doesn't help, and having the information right in front of your eyes is not as helpful."
Yet hope is not lost. Having identified the pervasive, often harmful effects of this universal error, the researchers' next steps will work toward formulating new methods of personal evaluation, which may help universities and businesses safeguard against misguided recruitment efforts.
In the meantime, staying mindful of the fundamental attribution error can, if only in hindsight, provide a humbling reminder of the limits to human perception, and perhaps—with enough reinforcement—teach us not to make the same mistake twice.