STEVE INSKEEP, HOST:
Here we are at the start of hurricane season, which runs from June to November. And this year, in addition to storing food and water, some researchers suggest people should also pay attention to the psychology of storm names. NPR social science correspondent Shankar Vedantam joins us each week on this program to talk about remarkable phenomenon's in social science and research. Hi, Shankar.
SHANKAR VEDANTAM, BYLINE: Hi, Steve.
INSKEEP: OK, storm names, hurricanes have names, Bob or Federica or whatever - why should we pay attention to that?
VEDANTAM: Well, the National Hurricane Center comes up with a list of hurricane names each year and it's a predetermined list. So in 2014, the Atlantic hurricanes, for example, the first one's going to be called Arthur, the second one's going to be called Bertha, the third one's Cristobal.
INSKEEP: Boy, girl, boy, girl.
VEDANTAM: They alternate male-female. Now, there's new research that shows that those names might play a role in the amount of destruction that the storms cause. I spoke with social psychologist Sharon Shavitt at the University of Illinois at Urbana-Champaign. And along with her co-authors, including graduate student Kiju Jung, they analyzed the destruction caused by Atlantic hurricanes between 1950 and 2012. Even after eliminating storms like Katrina, which caused unusual amounts of damage, they find that hurricanes with female names cause more deaths than hurricanes with male names. Here's how Shavitt put it.
SHARON SHAVITT: If you took a severe hurricane with a highly masculine name like Charlie, and you change that hurricane's name to Eloise, our models of the archival death data suggest that that would triple its death toll. It would take it from about 15 people dying to about 42 people dying.
INSKEEP: Wait a minute, random experiment here - many hurricanes - and you're saying the female hurricanes kill more people?
VEDANTAM: That's right. Now, obviously the names of the hurricanes don't determine how destructive a storm can be, Steve, unless we believe in the supernatural, which we don't believe on MORNING EDITION. What Shavitt thinks is happening is that when a storm is named Eloise, people take it less seriously than a storm named Charlie because their unconscious gender stereotypes tell them Eloise is supposed to be gentler than Charlie. Now, there may be tiny differences in willingness to evacuate or take precautions, and as a result you may start to see increased fatalities. So in some ways, this might be an example of unconscious sexism at work, except the people who are hurt by it are the very people who themselves hold the bias.
INSKEEP: OK, so hurricanes don't kill people, sexism kills people - who hold sexist beliefs. That's the theory anyway. But is there really hard evidence besides the stance that she mentions?
VEDANTAM: Well, the short answer is there is not direct evidence. Destructive hurricanes are very rare and we don't know whether people actually behave differently in hurricanes depending on the name of the storm. What Shavitt and her colleagues did was the next big thing. They conducted a series of laboratory experiments, where they asked volunteers to rate the risk and danger of storms with different names. Here's the interesting thing - when you ask people to rate the difference between Hurricane Arthur and Hurricane Arlene, most people will say, well, obviously there's not going to be a difference. But if you ask them to rate the storm's danger without alerting them to the fact that you're thinking about the storm's names...
INSKEEP: Just look at the category it's in or whatever other factors you may have...
VEDANTAM: ... And if you present the storms one by one and say how scared would you be about Hurricane Arthur - it's listed as a category 4 hurricane, here's where it's going to show up, here's where it's going to hit. And you ask people to rate how scared they are by it, something very interesting emerges. Here's Shavitt again.
SHAVITT: What we found is that, systematically, when people were asked to imagine a male-named hurricane like Christopher or Alexander or Victor, they felt that it would be more risky, more frightening, more intense and they were more motivated to take shelter than if they were imagining Hurricane Christina or Alexandra or Victoria.
VEDANTAM: You know, I should say, Steve, that between 1953 and 1979 all storms were given only female names.
VEDANTAM: And Shavitt and her colleagues find that even if you look only at storms with female names, storms with more feminine names like Cindy tend to be taken less seriously than storms with names like Bertha.
INSKEEP: Wow, so I'm thinking about this - when they started alternating male names in there that was a significant advance then in hurricane safety, it warned people more strongly against hurricanes at least half the time.
VEDANTAM: Well, the thing is that the lining up of male and female names is not going to line up with the severity of storms. And so here's the problem, you could switch all the names to male names and potentially people will take it more seriously, but much of the time you're going to be crying wolf because most hurricanes are not going to be so destructive. The better answer here, Steve, is for people to recognize that without their being aware of it, subtle factors like a name can unconsciously bias them.
INSKEEP: Which is why his Twitter handle is @HiddenBrain. NPR's Shankar Vedantam. Thanks very much.
VEDANTAM: Thank you, Steve.
INSKEEP: It's MORNING EDITION from NPR News. Transcript provided by NPR, Copyright NPR.