UI study: Don't be misled by hurricane's 'femininity'
Watch out for Hurricane Melissa. Mitch, as well.
A new study by behavioral researchers at the University of Illinois concludes that hurricanes with feminine names are likely to cause more damage than hurricanes with macho names — probably because girly storms are seen as less threatening.
In six different experiments, researchers found people were slightly more likely to flee a storm named Christopher than Christina, Victor than Victoria, Alexander than Alexandra and Danny than Kate.
And that can have deadly consequences: the more feminine a name, the more people it killed. An analysis of more than six decades of death rates from U.S. hurricanes showed that severe hurricanes with a more feminine name resulted in a greater death toll.
Both men and women rated female names as less frightening. "These are totally randomly assigned names. They don't predict the severity, but they do appear to predict people's response to the storms, how risky the storms are perceived to be," said co-author Sharon Shavitt, a UI professor of marketing.
The study, which didn't involve any experts in meteorology or disaster science, was published this week in the Proceedings of the National Academy of Sciences.
Shavitt talked about the findings Tuesday with The News-Gazette.
What drove you to choose this particular topic?
We're behavioral scientists. We look at how people evaluate things and how they make decisions or perceive risks. ... In particular, we're interested in how gender bias affects the way people judge things around them, including nonhuman things such as weather systems.
We know these kinds of implicit biases routinely affect the way actual men and women are judged in society, but now it appears that these gender stereotypes could be having deadly consequences.
Hurricanes started occasionally being named after men in 1979. How can you draw conclusions from the years prior, when we had only female-named storms?
First of all, our analysis primarily focused on the femininity or masculinity of names, not only on male-female as a binary category. Even during the female-only years, the names differed in their degree of femininity.
If you compare Ione, which is a female name during that era, relatively gender-neutral or less feminine, to Diane, a female name during that era which is highly feminine, that illustrates my point. (Both hit North Carolina in 1955, the costliest hurricane year in history, but Diane caused significantly more damage.)
Even going back to 1950, the first three years of the data set, they used a naval alphabetic system — Able, Baker, Charlie, etc. — as code names of hurricanes. It might not surprise you that King, one of the hurricanes early on in the data set, is perceived as very masculine.
The nature of the name makes a difference for severe hurricanes, the ones high in damage. It doesn't make a difference in less severe hurricanes, where death rates are low to begin with. But for severe storms, where protective action has the greatest potential to save lives, that's where we saw the degree of femininity of the name predicted its death toll.
What do you think the study tells us about hurricane preparedness and the practice of naming hurricanes?
We will leave the policy decisions to the policy experts. But we think that this research has something to add to the conversation about social factors that predict responses to hard warnings. Meteorologists have called for more consideration of social science factors to better understand how people react to hazard communications. And we think it is worth considering this finding and following it up.
We also want to encourage people in the paths of storms to think consciously about the storm names, so they can remind themselves that the name a hurricane is assigned means nothing about its severity.
Are we likely to see more male-named hurricanes as a result? Or perhaps names like Voldemort or Godzilla?
That's been suggested by a number of people who've read the study. It's a good question. Probably this possibility should be looked at. At the same time, it's not as easy as giving them all scary names. Storms differ a lot in their severity. ... The key is to provide relevant information.
Would the National Hurricane Center/World Meteorological Association be open to a different naming system?
I think they should study it. And any name or label that's assigned to storms going forward probably should be pretested to see what responses it elicits, rather than just moving to a different arbitrary system.
If you name them all Voldemort, our data would suggest that it would work to make people take the storms more seriously. But if you give them all scary names, you would probably be scaring people equally across every storm, instead of giving information that's key to the storm's characteristics.
Have you ever lived through a hurricane?
I have not. I've lived in the Midwest most of my life. We've had tornadoes, other severe weather, but no hurricanes.
2014 hurricane names
The National Hurricane Center started using female names for Atlantic storms in 1953, then began alternating with males in 1979. Lists of names are now maintained by the World Meteorological Organization. With Atlantic hurricane season under way, here are the 2014 names: