Why female engineering students help reproduce a structure of bias.
“A little more gender balance…would be nice, but at the same time, I want the best people to be there,” writes a young African American woman studying engineering. “I really hope that I haven’t gotten where I am…because of my gender. I’d like to think I earned it.”
Nobody likes to imagine they’ve achieved success because of an unfair advantage or that they were hired for reasons unrelated to the job. Rather, we like to believe it was pure ability – merit – that landed us the corner office and the fat salary.
Meritocracies are predicated on the belief that only the best are chosen and that hard work and talent are always rewarded. If we presume that talent and hard work are not gender specific, then why is it that assumed meritocracies show extraordinary imbalances between men and women?
Surprisingly, part of the answer is the assumption itself: Assuming a setting is a meritocracy can blind even those experiencing discrimination to its actual inequalities. Even worse, blindness prevents change.
The engineering environment is one such setting. Although engineering may perceive itself as upholding the ideals of a true meritocracy, this self-perception actually prevents members from recognizing the numerous instances in which meritocratic values fall short.
The number of women studying or working in the profession is dismal: 19 per cent enrolment, 14.8 per cent of working engineers, and an attrition rate as high as 50 per cent. So what is it about engineering that maintains extreme imbalances despite decades of effort toward gender parity?
If you can’t take the heat, get out of the kitchen
To find an answer to that question, Brian Rubineau, Associate Professor of Organizational Behaviour at McGill University, and co-authors Carroll Seron (University of California-Irvine), Susan Silbey (Massachusetts Institute of Technology), and Erin Cech (University of Michigan) studied a group of female engineering undergrads at four universities. Over four years, the team collected diary entries from the students and then analyzed the data. The results were revealing.
“Individual experiences of blatant sexism were written off as aberrant, one-off run-ins with jerks, and not representative of the profession,” said Rubineau.
This, he explains, is a common reaction among those working within an assumed meritocracy who believe they are there because they deserve to be and so are invested in overlooking, diminishing, or justifying instances of bias.
“Fake it ’til you make it.”
Paradoxically, adopting this interpretation provides an ego boost to those suffering discrimination: A mantra among the female students was “Fake it ’til you make it.” Attrition is considered a personal failure rather than due to structural problems in the profession. The ones who could “hack it” did so proudly and, as a result, considered themselves superior.
If it’s not broken, why fix it?
So, how does one untangle such a Gordian knot and introduce diversity when the people best suited to identify bias are opposed to change?
Rubineau points out that the most successful changes tend to come from within but adds, “Engineers listen mostly to engineers and trust engineers, and those people won’t say that engineering is not a meritocracy. The urgency to change within the profession is low.”
It’s a complicated problem and those within the culture are usually blind to it. As an example of how deeply hidden this bias can be, Rubineau cites Amazon’s foray into using artificial intelligence as a recruiting tool.
Intended to eliminate subjective opinion and filter out the top candidates, Amazon’s AI system mined the resumes of the previous ten years’ successful hires for patterns. Given that the majority of those resumes came from men, it now seems obvious that the AI would necessarily reproduce the corporation’s existing inequality—but that outcome was not anticipated.
“It hired almost only men and did so without conscious awareness or animus in its bias, and that’s exactly what we’ve been seeing in engineering.”
“It hired almost only men and did so without conscious awareness or animus in its bias, and that’s exactly what we’ve been seeing in engineering.”
If the current system is assumed to be meritocratic, or perfect, then any deviation from it is seen as inappropriate. As a result, the status quo is perpetuated.
As do many people, the women in the study firmly believed that affirmative action programs or enforced quotas would result in the hiring of poor-quality candidates. But, Rubineau notes, there is no empirical evidence to support this belief and plenty of data showing that the opposite is true.
Take, for example, experiments in which diversity was forced through quota systems. The results show that women do no worse than men, that talented women are not in short supply, and that there’s no kind of sacrifice in quality.
But resistance to corrective measures to promote gender diversity remains strong from all sides. As one woman in the study wrote, “I do not want to be a pity hire.”
See no evil
Rubineau and his co-authors provide important insights into why past efforts have failed. They also offer guidelines for new approaches to achieving diversity in hiring as well as maintaining it through employee retention.
Significantly, research clearly indicates that recruitment campaigns will not intrinsically result in a diverse workforce. Business owners and hiring teams need to acknowledge that simply assuming an ideology of meritocracy while maintaining existing methods of identifying talent will replicate disparity. Further, as the attrition rates for women in engineering indicate, even if greater diversity among hires is achieved, relying on them to affect structural change is misguided.
Some organizations are trying approaches that mask an applicant’s identity to root out hidden bias when hiring. One approach is to redact resumes so that they contain nothing indicating the gender or ethnicity of an applicant; another is to hold blind auditions.
Research conducted with orchestras has shown that 50 per cent more women are hired when musicians audition behind a screen, unseen by the hiring panel.
In a similar vein, the Silicon Valley start-up GapJumpers has created online technology that replicates the benefits of blind auditions, enabling managers to test the skills and abilities of applicants and then rank them without the risk of hidden bias. GapJumpers claims that, when using their technology, 60 per cent of the top talent comes from marginalized groups.
But after hiring there’s retention, and that can prove to be an even greater challenge. Those committed to improving diversity within their corporate structure should first consider the complexity of the problem and not resort to single-effort solutions. Transparency and accountability are key. However, teams would benefit from remembering that every employee or recruit has a range of skills that may not be easily assessed vis-à-vis those of others. Talent that is right for one management level may not work at another level.
Most important, don’t leave the job of changing the culture to those most impacted by bias. Although they may be best suited to recognizing it, they are also the most at risk and not likely to increase their vulnerability by speaking out.
Above all else, perhaps a good first step is to admit that, though the pursuit of meritocracy is a worthwhile goal, we’ve still got a long way to go before truly achieving it in the real world.