A degree of gender bias exists within the hiring process at most organizations in the U.S., and no matter how slight, this bias can have a profound ripple effect on a company’s productivity and workplace culture over time, according to new research.
A new paper published in the Journal of Management on Jan. 13 studied the impact of gender bias, systematic preference or prejudice toward one gender over another when the employee’s gender is irrelevant to the job, on hiring outcomes across a range of hiring scenarios and in diversity-oriented staffing practices.
Jay Hardy, an assistant professor at Oregon State University and lead author of the paper, told The Academic Times that few people would dispute the existence of gender bias in hiring decisions, as human beings are naturally flawed in their decision-making.
But previous research on the subject has primarily examined the presence and the size of a gender bias in hiring, rather than its root cause, its impact on an organization or sustainable solutions to the issue.
And academics continue to debate whether this is a problem worth studying and whether it’s worth it for organizations to invest money in correcting biases if they seem small, according to the paper.
“Of all the things that we think about a person when making a hiring decision, about 4% of our evaluation of that employee is going to be based on simply whether they are male or female,” Hardy said.
For the current study, the researchers analyzed several previous experiments about hiring practices that involved candidate qualifications and gender. They found that overall, men and candidates with higher qualifications were perceived to be more hireable.
The team gathered bias estimates from these experiments to use as a basis for a novel computer simulation that modeled different potential situations that organizations might encounter while recruiting and hiring. In the simulation, the researchers were able to set the amount of gender bias effect and test hiring outcomes.
“The magnitude of discriminatory hiring outcomes associated with even small amounts of gender bias generally proved to be quite substantial,” the authors said in the paper.
The researchers set gender bias effects at 0%, 1%, 2.2% and 4% in the simulation models. The values represent plausible gender bias estimates consistent with data gathered from the experiment analysis. With a 2.2% bias effect, rates of disparate treatment were 13.5% higher than what was observed in the model with a 0% bias. And in the 2.2% bias effect model, women were 49% less likely to receive a favorable hiring decision than men.
In the model with a 4% bias effect, rates of disparate treatment associated with the bias increased by 20.3%, and a woman’s overall odds of getting hired were 60% lower than the odds for a man. Disparate treatment refers to an applicant being treated differently than other similarly qualified applicants based on a prejudice or bias. Discrimination lawsuits against organizations usually pertain to allegations of disparate treatment.
Hardy and the team concluded that there were very few situations in which a small bias effect could be ignored. Even if a hiring bias is measured at less than 1%, it still has an impact on the company in terms of discrimination and productivity loss — so debating the size of a bias is unproductive.
“In nearly every situation that organizations would be in, focusing on the hiring decision, we found these big discriminatory effects that emerged, and these huge dollar value implications of the performance,” Hardy said.
“This little tiny effect, it can’t really be disregarded because when it’s filtered through the decision-making process that organizations typically employ, it can lead to massive rates of discrimination, and it can lead to huge costs for the organizations due to suboptimal hiring decisions they’re going to make.”
In the 2.2% model, for instance, each new hire cost the company an extra $710 per year, and new hires failed at a rate of 16.1%. The 4% model recorded a loss of about $2,125 per new hire per year and a failure rate of 50.2%. New hire failure rates are important to organizations because it’s estimated that replacing a failed hire can cost up to three times the amount of the employee’s annual salary.
The authors explained in the paper that a typical Fortune 500 company that hires 8,000 new employees a year with a 4% gender bias effect in its hiring process can expect the botched hiring of an additional 192 new employees and a loss in productivity of approximately $17 million per year.
As a solution to hiring bias and inequitable workplaces, some companies use a “targeted recruitment” approach that actively encourages increasing the number of diverse, qualified candidates in a hiring pool. Hardy said these methods can produce more equitable ratios within a workforce to an extent, but they can’t fix the overall problem without addressing underlying bias.
“Given these results, we advocate for a ‘no tolerance’ policy when it comes to the question of how much bias can be reasonably tolerated in the evaluative process, as our simulation findings show that a failure to address even small amounts of existing biases can have substantial negative consequences on hiring outcomes,” the authors said in the paper.
Hardy urged organizations to recognize that bias is a natural consequence of the human brain and is inevitable. And hiring managers should find solutions to their problems that will have a lasting impact, he said.
Companies large and small have revisited their recruitment practices in recent years, and the issue particularly came to the forefront last summer during the large-scale racial justice movement in the U.S. that followed the death of George Floyd at the hands of Minneapolis police.
It’s a common problem for an organization to be more diverse in the bottom levels of its workforce than in the upper management levels, and Hardy said slight, subtle biases in hiring are likely tipping the scales behind the scenes and producing these unfair outcomes.
“I hope that we can remove some of these evaluative emotional judgments and conversations around [this topic] and stop making this such a political issue, and focus on it as a problem that needs to be solved,” Hardy said.
The study, “Bias in Context: Small Biases in Hiring Evaluations Have Big Consequences,” was published in the Journal of Management on Jan. 19. Jay Hardy, of Oregon State University, was the lead author. Richard F. Martell and Andy Olstad, of Oregon State University, and Kian Siong Tey, Wilson Cyrus-Lai and Eric Luis Uhlmann, of INSEAD, served as co-authors.