June 18, 2019
In 2015, NASA celebrated over 50 years of spacewalking. Three years later, in March 2018, the agency called off the first all-female spacewalk due to a shortage of smaller-sized spacesuits. The walk-back sparked a Twitter storm of women sharing stories of their own ill-fitting work uniforms and oversized ‘standard’ gear. "But," as one woman tweeted, "it's not just spacesuits. It’s public spaces like bathrooms, cars, cockpits, office air conditioning, microwave installation heights, Oculus, military fatigues…an endless list.”
Back in December, personal experience set me on a trail of research that ultimately revealed the phenomenon of patriarchal coding. Here's what I learned: 1) I'm not alone in thinking today's VR headsets weren't designed for women. 2) The majority of products and systems we use every day are actually designed by and for men. 3) Patriarchal coding affects almost every aspect of women’s lives. 4) It's unintentional for the most part. Sexism is so ingrained in our society that women’s unique needs and biology (like the fact that we have breasts) are excluded from reality, even of the virtual kind.
Emerging technologies are no exception: Wearable devices such as body-worn sensors, exoskeletons, and VR headsets - which experts hail as the future of safety and training - exhibit coded patriarchy. One-size-fits-men wearables risk further alienating the female workforce when all signs indicate that the future of work requires women to succeed. We cannot allow workplace wearables to succumb to the biased design thinking behind everything from automobiles to office buildings if we hope to improve job prospects and working environments for women.
The history of man
Women and the female perspective are largely missing from world history (as is the non-western point of view), so it's not much of a surprise that women are entirely absent in the research underlying the foundations of modern life. The star of the show is “Reference Man,” a 154-pound Caucasian male aged 25 to 30, who has been taken to represent humanity as a whole when it comes to designing products, planning spaces, and setting standards. For example, medications are tested on men even though women process drugs differently, and standard vehicle seating position was determined using a male crash-test dummy, though women drivers sit further forward and are thus 47% more likely to be seriously injured in a car crash. (Side note: In 2011, the U.S. finally introduced a female crash-test dummy, though not an anthropometrically correct one.)
Beyond product ergonomics
It's both frustrating and dangerous that so many gadgets we use today are one-size-fits-men. Developers have also failed to factor women into software design, which is why navigation apps provide the quickest and shortest routes to a destination but not the safest. Consider voice recognition and other AI tech we use to interact with our devices and make decisions: Google’s voice recognition software is 70% more likely to accurately recognize male speech, and Apple's Siri, when first launched, could assist with a heart attack but did not understand the phrase "I was raped."
Last year, Amazon had to scrap an experimental recruiting tool that taught itself to prefer male candidates for software development and other technical jobs. How did this happen? The computer model was trained to observe patterns in resumes from the previous ten years. Tech is notoriously, overwhelmingly male, so of course most of the resumes were submitted by men. What’s alarming is that in a 2017 survey, over half of U.S. HR managers said they would make artificial intelligence a regular part of HR operations within five years. That means women will have to combat unfair algorithms in addition to unconscious bias in order to advance in the workforce. IBM CEO Ginni Rometty says it’s up to businesses to prepare a new generation of workers for AI-driven changes to the workforce. In a world in which AI will impact every existing job, the fact that women and minorities are disproportionally left out of the early stages of the tech is quite tragic.
The data gap at the heart of the workplace
Occupational research has traditionally focused on male workers in male-dominated industries. Few studies have been done on women’s bodies and work environments, so there is little occupational health and safety data for women and other gender identities. Years of relying on data from male-only studies is why most professional uniforms are designed for the average man. It's why trends like the increasing rate of breast cancer in industry remain a mystery, and why serious injuries in the workplace have gone down for men but are increasing among female workers. Women have been entering the workforce at twice the rate of men for the last three years, yet we're essentially clueless about women working in male-dominated industries.
Diverse data can be critical when it comes to body-worn sensors that track biometric and environmental factors in the work environment. Say we're tracking exposure to a certain chemical. When a reading reaches the threshold, a companion app sends an alert to the wearer. But how is that threshold determined? Differences between men and women - hormones, immune systems, body fat percentage, etc. - influence how chemicals are absorbed by the body. Without female data, the danger zone is likely to be higher than the toxin level to which a woman can be safely exposed, putting her at greater risk. At this point, it would take a working generation of women to get any usable data on long-latency work-related diseases.
No PPE for you
Construction is one of those male-dominated industries in which standard equipment and PPE has been designed around the male body. Although there's little data on women in construction, a study of union carpenters did find that women have higher rates of wrist sprains and nerve conditions than their male counterparts. To comply with legal requirements, many employers just buy smaller sizes for female employees, but scaled-down PPE doesn’t account for the characteristics of a woman's body (chest, hips, etc.). How can a woman be expected to perform at the same level as a man if her clothing and equipment are a hindrance? Already a minority in the sector, women don’t usually complain about ill-fitting PPE or oversized tools. Instead, they make their own modifications using duct tape, staples, and the like. (Dust and hazard eye masks designed for the Reference Man also put many men of color at a disadvantage.)
Of course, it doesn’t have to be this way. Standards can change. A standard-sized bag of cement, for instance, could be made lighter for a woman to easily lift it. Exoskeletons may be another solution. For Jane Henry, the answer was going back to the drawing board. Jane's SeeHerWork is an inclusive clothing line for women in fields like construction and engineering, both of which have lucrative, equal-pay careers and massive labor shortages.
Designing the workplace
Surprise, surprise: Men are the default for office infrastructure, too, from the A/C (feeling cold hurts productivity) to the restrooms (a single restroom with urinals serves more individuals). According to the Bureau of Labor Statistics, women represent nearly two-thirds of all reported cases of carpal tunnel syndrome, which indicates that workstations are less ergonomic for women. Open office plans (lots of glass, identical desks, group spaces) are conducive to socializing and breaking down hierarchies, right? No, they actually encourage sexist behavior. A 2018 study documenting the experiences of women in an open office designed by men found that the lack of privacy created an environment in which female workers were always watched and judged on their appearance. Designers are starting to use virtual reality to determine factory layouts, workstations, and even assembly processes, but it's a safe bet they're not deliberately putting headsets on women to get their input.
I spoke with Janelle Haines, Human Factors Engineer at John Deere, about her experiences using VR - mostly on males - to evaluate the ergonomics of assembly.
“Fitting the job to the worker hasn’t [always] been a focus. Even in the last fifteen years that I’ve been studying ergonomics, there has been a huge shift in learning to focus on ergonomics. It has become a kind of buzz word…There are some jobs that have been at John Deere for years and years, since we started building combines, that aren’t a great fit for women, but going forward with new designs we’re using VR to make sure the workstations and what we design do work for women.”
Ergonomics aren’t a new area of study, but Janelle's experience shows a promising shift in thinking and a deliberateness that’s necessary to address inequality.
The future of work: Uncomfortable = unproductive
Women typically have smaller hands and wrists than men. Because of this, women usually need both hands to use the standard smartphone while men can use it one-handed. This kind of oversight may carry into the next wave of mobile: Wearable technology. Muscle mass distribution, vertebrae spacing, bone density, finger length...biology and physiology matter when designing devices meant to be worn on the body.
Women are already at a disadvantage in VR. As far back as 2012, researchers found that men and women experience virtual reality differently. According to a growing body of research, depth perception is to blame: What creates a sense of immersion for men is motion parallax or how objects move relative to you. For women, it's shape-from-shading, meaning an 'off' shadow can ruin the experience for a female user. Since shape-from-shading is more difficult to render in VR, most VR tech uses motion parallax. Why does this matter? VR will be crucial for filling millions of vacant positions and upskilling the workforce as automation advances. Women's inferior experience of the virtual world adds one more roadblock to their advancement in industry.
Stop defaulting to men
The long legacy of researchers overlooking or not wanting to pay for testing on women has serious implications in the era of Big Data, when millions of connected 'things' are generating massive amounts of data. It’s much bigger than a spacesuit: Feeding biased, incomplete data into powerful algorithms that make important decisions for us will exacerbate inequality, put certain populations at risk, and encode prejudices into our future. The answer (like the answer to many issues of inequality) is more diversity in the labs and back rooms where the future is being designed, tested, and engineered.
In writing this article, I drew heavily on the efforts and writing of a number of inspiring women, including: Caroline Criado-Perez, author of Invisible Women: Data Bias in a World Designed for Men,” Abby Ferri of the American Society of Safety Professionals, and Rachel Tatman, research fellow in linguistics at the University of Washington.