Residents of a low-lying coastal town get the alert: A severe storm named Hurricane Eloise is coming, and it’s time to evacuate. But three times the number of people will die in Eloise than they would in a storm named, say, Charlie. Why? A slew of psychological experiments and an analysis of historical data suggest that people take female-named and especially feminine-sounding storms less seriously.
All storms used to be given female names, but that changed in the ’70s, courtesy, in part, of the women’s liberation movement. But today, choosing the gender of the storm is a completely random process. The names are picked off a list in alternating gender order. “We were interested in how gender bias impacts things around us, including non-human things like weather systems,” explains University of Illinois at Urbana-Champaign professor Sharon Shavitt, the co-author of a paper on these findings, published in the Proceedings of the National Academy of Sciences. “The alternating male-female names provided an ideal context in which to do this research.”
Even though studies have consistently shown that gender bias can impact employment decisions and judgments of competence, the researchers weren’t prepared for the strength of their disturbing findings. After analyzing death tolls from 47 storms, 17 of which were male-named, the researchers found that when it came to severe storms, female-names correlated with double the mortality rate, on average. With more feminine-sounding names, like Eloise or Cindy, the difference could triple.
The implication of that data could be that, when faced with an oncoming storm, people perceive the feminine as less of a threat. In a series of seven experiments, the University of Illinois at Urbana-Champaign researchers decided to test that hypothesis. They asked test subjects to rate things like storm severity and their likelihood to evacuate based on names alone.
“The data on the storm deaths is correlational. We can’t get around that,” Shavitt says. “On the other hand, when people were asked to imagine a storm with a male name coming for them, we see a difference time and time again–that a female storm seems less threatening and less foreboding. And as a result of that less perceived risk and intensity, we see people less willing to take shelter, more likely to delay evacuating.”
The researchers’ findings could be some of the strongest indicators of the extensive, societal-level damage of sexism yet. But not every disaster expert agrees that hunting for such patterns in large datasets justifies the study’s conclusions.
“I am familiar with this argument, and I think it is genuinely without merit,” writes Louise K. Comfort, director of the Center for Disaster Management at the University of Pittsburgh, in an email. She notes that while she hasn’t yet read the study, correlational patterns in historical storm data don’t necessarily indicate cause and effect. “I have seen no real data to support what is likely a specious argument, more attuned to the sensational effects than genuine measures of impact.”
Others see it as a call for continued research. Tricia Wachtendorf, associate director of the University of Delaware’s Disaster Research Center, points out that factors outside of storm severity–like storm surge or social vulnerabilities–can also greatly impact mortality rates. Then again, she writes, gendered expectations have been proved to have an effect on task delegation and disaster response operations.
“The data from this study and significant findings, including the anticipated evacuation behavior (which may or may not be related to actual behavior) point to the need to explore this topic further,” she writes in an email.
The findings also raise another major question: How could gendered expectations of hurricanes possibly hold up after Hurricane Katrina, one of the most devastating natural disasters in U.S. history, killing nearly 2,000 people, and wiping out more than a million buildings?
“You would think that things would be different, that people would bring to mind more severe, deadly, feminine storms, and maybe they would judge the female storms as more severe than the male storms, but they don’t,” Shavitt says. “Our findings suggest that gender bias could be strong enough to overcome specific experiences with storms.”
Shavitt and her team are continuing the investigation. Currently, they’re studying how gender impacts aid–whether people tend to donate more money to male or female hurricane survivors.