John Metcalfe was CityLab’s Bay Area bureau chief, covering climate change and the science of cities.
A flawed study claims people tend to view feminized hurricanes as less risky and are therefore more likely to be killed by them.
Say a powerful hurricane is barreling down on your community. Would you be more likely to evacuate if it was called Fay or Omar?
Probably Omar, claim researchers behind a new and somewhat problematic study about the psychology of hurricane names. People who hear about storms with masculine monikers like Arthur and Kyle perceive them as riskier and potentially more destructive, according to the study's authors at the University of Illinois at Urbana-Champaign and elsewhere. But hurricanes with feminine names like Dolly and Hanna register as less threatening, they say, making people more apt to hunker down and get wiped out by the monster tempests.
How storm names might affect our threat assessment and, in the end, survival odds is a matter of gender expectations, the researchers allege. Here they explain how it might go down in our brains (I've removed citation numbers for clarity's sake):
Research shows that women and men are socialized to have different social roles and self-schemas, in turn generating descriptive and prescriptive expectancies about women and men. Men are often expected to be strong, competent, and aggressive, whereas women are often expected to be weak, warm, and passive. Men are more likely than women to commit violent behaviors, and thus males are perceived to be more strongly associated than females with negative potencies such as violence and destruction. We extend these findings to hypothesize that the anticipated severity of a hurricane with a masculine name (Victor) will be greater than that of a hurricane with a feminine name (Victoria).
To test whether this theory holds water, the researchers asked people to rate how "intense" they thought the names are for 2014's Atlantic hurricane season, which began on Sunday. They also how at risk folks would feel if they were in the track of a storm named Christopher or Christina, and how they'd react under voluntary-evacuation orders for hurricanes Danny or Kate. By and large, people responded with greater alarm and willingness to quickly leave town for masculine storms – something the researchers deem a "hazardous form of implicit sexism."
So is it time for the World Meteorological Organization to abandon its yearly grab-bag of boys' and girls' names? The researchers suggest that it's worth considering. By their calculations, having female names in currency might be causing larger death tolls: Changing a storm's name from Charley to Eloise almost triples the fatalities it's apt to cause, they say.
Or maybe not. Some background on the hurricane naming system: The international community has used distinctive storm names for decades, the theory being it makes them easier to communicate about among the wide network of weather stations and ships. And in the 1950s and '60s, these names were always female. "In that era when political correctness had never been heard of, the exclusively male meteorological community in the USA considered female names appropriate for such unpredictable and dangerous phenomena," explains the U.K.'s Weather Online. "In the 1970s the growing numbers of female meteorologists began [to] object to such a sexist practice, and from 1978 onwards girls' and boys' names alternated." (That was for the Pacific hurricane season; the Atlantic season, the focus of this study, switched over in 1979.)
The researchers say that after performing an archival study of 92 landfalling hurricanes from 1950 to 2012 (using primarily NOAA's storm data), they found that "hurricane fatalities established that severe storms with more feminine names are deadlier." Except for nearly three decades of this period, only female names were in circulation. I asked one of the study's authors, Sharon Shavitt, to break down the differences in fatalities for male-and-female storms throughout history, and here's what she wrote back:
Since 1950 there were 47 storms that caused over $1.65 Billion in damage (normalized value adjusted for today's dollars). $1.65 B is the median value of normalized damage, meaning that half are above/half are below this level of damage.
There were 17 total "male" storms. They caused 391 total deaths, for an average of 23 deaths per storm
There were 30 total "Female" hurricanes. They caused 1354 total deaths, for an average of 45 deaths per storm
In other words, for these damaging storms, female named storms killed about twice as many people as those with male names did. And when you take into account the femininity/masculinity of the names,* the differences are greater, as discussed in the paper. For severe storms, highly feminine names can increase the death toll by 3-fold or even 5-fold if you compared the most masculine possible name with the most feminine possible name.
* What Shavitt is referencing here is another part of the study that ranked hurricane names along a scale, with most masculine on one side and most feminine on the other. But even if storms with more feminine names indeed have caused bulked-up death tolls, you have to wonder why the years when only these names existed got included in the study.
Lead author Kiju Jung did try to account for these two different naming systems. But that tactic posed a problem, as National Geographic's Ed Yong explains in an enlightening take on the study:
Jung’s team tried to address this problem by separately analysing the data for hurricanes before and after 1979. They claim that the findings "directionally replicated those in the full dataset" but that’s a bit of a fudge. The fact is they couldn’t find a significant link between the femininity of a hurricane’s name and the damage it caused for either the pre-1979 set or the post-1979 one (and a “marginally significant interaction” of p=0.073 doesn’t really count). The team argues that splitting the data meant there weren’t enough hurricanes in each subset to provide enough statistical power. But that only means we can’t rule out a connection between gender and damage; we can’t soundly confirm one either.
For what it's worth, hurricane names are retired "if a storm is so deadly or costly that the future use of its name on a different storm would be inappropriate for obvious reasons of sensitivity," according to the National Hurricane Center. Since 1979, 53 names have been retired; 25 of these names were female, and 28 male.
There's a lively comments-fest going on at the Capital Weather Gang about this perplexing study. But whether you buy into it or not, it's worth noting that a ton of things beside "gender expectations" affect death rates during a storm disaster. Here's the take of Bill Read, former director of the National Hurricane Center, who writes:
While the gender bias is likely real, I don’t think it plays a significant role in human response to an approaching landfall. The test conducted for the study involved people who were not under the stress of an approaching hurricane. As quoted in the article, while necessary to eke out the gender difference, it leaves me with the need to know if is this factor significant, or is it very minor in the mix of all other societal and event driven responses. My experience with Rita (massive (over) response to evacuation orders) and Ike (less than ideal response) is a point in fact. In the case of Rita (sweet female), the events three weeks earlier due to Katrina were cited as a contributing factor to over reaction. For Ike (bad boy male), the horrific evacuation for Rita was cited as a reason for under response. I used to think, and still do with caveats, that a more important driver is how strong the storm is at the time action is required. Rita was a Cat 4 heading to 5 when decision time came. Ike was a Cat 2. These two real world events had exactly the opposite response one would expect from the gender bias paper.