Human-Computer Interaction for Development (HCI4D) is a major research area at CHI 2017. It is focused on designing and developing technologies for under-served and under-represented populations around the world. We spotlight an important example of this work and look at the broader contributions of HCI4D at this year’s conference.
Georgia Tech Research Examines a Government Mandate for Panic Buttons in Mobile Devices and the Need for a Better Answer
Story and Visualizations by Joshua Preston
Illustrations by Z.P.
When walking down an unfamiliar or isolated street you might on occasion take out your cell phone and pretend to talk on it in order to deter any would-be criminals. It turns out this is a universal instinct, one of many that women in New Delhi employ when in public spaces, where they often face pervasive sexist attitudes and violence. This year, a mandate by India’s government goes into effect for cell phone manufacturers to include a panic button on all new devices in an effort to curb increased violence against women.
Georgia Institute of Technology researchers interviewed and surveyed women and men in New Delhi and other parts of the country to understand how effective such a measure might be. They found that a number of interconnecting factors influence people’s personal sense of safety.
“We studied all forms of violence women face in public spaces, but our focus is on real or anticipated situations that might bring to test how well the panic button would work,” said lead faculty member Neha Kumar, assistant professor in the Sam Nunn School of International Affairs and the School of Interactive Computing at the Georgia Institute of Technology.
CHI 2017’s HCI4D papers cover a broad range of challenges. Click to interact and see overlapping areas of study in the bar chart.
Isolation was the most cited reason New Delhi was considered unsafe past 9 p.m., when there is a considerable drop in the number of people in public areas. Respondents said large crowds and high-density areas also presented their own dangers, giving individuals who harassed women a cloak of anonymity. But crowds of women in designated women’s compartments on the railway system were deemed safer with little threat of harassment or assault.
Smartphones allowed women to hold strangers accountable for their actions, primarily while traveling. When using Uber-like services or auto rickshaws, interviewees said they would take photos of the driver and license plate and send to family members before getting into the vehicle, as well as send updates of their location during the trip.
Building a Better Solution
Many women said that a panic button on their phones wasn’t beneficial to their current technology practices, one reason being that the phone might not be immediately accessible during an assault. Also, law enforcement, while seen as an effective deterrent against criminals, wasn’t trusted to always respond to assaults for a variety of reasons.
“Our findings highlight that women’s sense of safety is incredibly complex,” said Kumar. “There are multiple factors that exert power over women, affecting their sense of safety and behavior in an emergency situation.”
Researchers found, based on the results, that the mandate for a panic button in phones might ultimately be ineffective in preventing assaults because it does little to interact with, and even misunderstands, the social power dynamics and factors that play into women’s public safety.
“Some participants felt that the impact of sexual assault was too significant to risk, and they valued proactively taking safety measures much more than using a reactionary tool such as the panic button,” said Naveena Karusala, a Georgia Tech computer science major and lead author on the study.
HCI4D ranks 9th globally in research at the conference. Click to interact and see the top ten HCI fields ranked by country.
The researchers developed redesign options for the panic button to help address systemic issues likely to impact its effectiveness. To ensure accessibility, they recommended a stand-alone wearable device, something that connects to a smartphone or an accessory to ensure that the panic button is almost always reachable. To avoid accidental activation, the researchers say that utilizing a series of multiple quick presses or a single long press could be an option.
Making it more functional is another priority, such as programming the panic button to update a person’s social network with his or her location if necessary and alerting family as well as authorities when pressed.
“Our design recommendations are informed by a nuanced understanding of women’s perceptions of personal safety, ensuring women are the ones speaking and designing for themselves,” said Karusala.
But researchers warn that different functionalities and form factors alone aren’t sufficient to make a panic button effective in the sociocultural context of New Delhi. They say it is imperative that the government increase the accountability of police in terms of responsiveness to emergencies for the system to help mitigate public violence against women.
“This is just one effort to help address the larger social issue”, said Kumar. “But I think the best metric for progress is women being free of the undue burden to ensure their personal safety in public spaces.”
The research, “Women’s Safety in Public Spaces: Examining the Efficacy of Panic Buttons in New Delhi,” is being presented this week at the Association for Computing Machinery’s Conference on Human Factors in Computing Systems (CHI 2017) in Denver, Colorado.
Thirty-four authors published HCI4D work in this year’s main technical program at CHI. Click to interact and see author details.