Statistical analysis showed that the death toll for severe hurricanes with more feminine names was much higher than that for hurricanes with masculine names—and it all comes back to implicit biasby Pragya Agarwal / September 18, 2019 / Leave a comment
On 5th September, the UK Met Office along with Met Éirenann, the Meteorological service in the Republic of Ireland and their new partner the Royal Netherlands Meteorological Institute (KNMI), revealed the storm names for 2019-20.
Having names makes it easier for people to keep track of severe weather warnings and also increase the awareness of any impending severe weather condition. In the UK, the practice of giving storm names only started in 2014, although it had been widely used in the USA since the 1950s—when the hurricanes were only given female names, due to their unpredictability. Fortunately, we have moved away from that explicitly sexist practice—UK storm names alternate between male and female—but the politics of storm names haven’t yet died down.
It might seem like a fun and innocuous thing to personify a weather phenomenon with a name, but research shows that names matter. In my upcoming book SWAY with Bloomsbury Sigma, I investigate in detail how our unconscious biases are formed, reinforced and activated. My research shows names are rich sources of information. They can signal gender, class, race, and can also evoke a feeling of comfort, familiarity and glow of warmth (or the reverse). Names are, therefore, subject to stereotype, and they can activate the implicit gender bias which runs through our society. When people see a typical ‘feminine’ name, the bias and stereotypes associated with the gender are also transferred to the name-holder.
A research study of hurricane names and corresponding fatalities, published in the Proceedings of the National Academy of Sciences of the United States of America in 2014, showed that even in the case of a natural disaster, its assigned name results in it being judged as per the social biases and expectations for that sex. The researchers did a historical analysis of 94 actual Atlantic storms between 1950 to 2012. Kiju Jung and his collaborators at the University of Illinois at Urbana-Champaign and Arizona State University took the storm’s names and gave them to a group of participants—who didn’t know that they were hurricane names—and asked them how masculine or feminine each one was, as well as the potential risk. From this independent coding, a masculinity-femininity index (MFI) was developed. Statistical analysis showed that the death toll for severe hurricanes with more feminine names was much higher than that for hurricanes with masculine names.
Over six experiments, researchers showed that hurricanes that had explicitly female names were judged to be less risky by people than those that had male names. When asked about a male hurricane, like Alexander, people predicted a more violent storm than when asked about a female hurricane, like Alexandra. They were more willing to evacuate to avoid Hurricane Victor than when it was Hurricane Victoria. The more masculine the name, the more respect the hurricane drew. The ones who read the scenarios with male names said that they were more likely to evacuate than those who read identical scenarios with female names. People assumed the hurricanes randomly given feminine names to have a less violent impact, and therefore prepared less for the severity of the impact—which, the researchers suggested, was the likely reason for more fatalities from hurricanes with feminine names.
The participants were being led by their unconscious bias to equate masculine names with strength and aggression and feminine names with warmth and gentleness. The bias is not limited to people who hold explicit gender beliefs, but also those who consider themselves egalitarian. These are the traditional masculine and feminine frameworks that our society operates under which create these implicit biases. When we see anything labelled with other categories, our responses to such an event or entity is informed by our mental representation associated with this category. These mental representations then often influence our reactions. For instance, if a hurricane was named after a flower, we would naturally assume it to be gentler than one that is named after a reptile.
As the global climate changes, the frequency and severity of such hurricanes and storms are likely to increase. This research shows that subjective risk perception and communication of a weather phenomenon is shaped by the implicit biases, and has implications for meteorological agencies, science communicators and policy-makers.
But what the research also shows is that name and gender bias is pervasive in our society. Perhaps it is time to remove the gendered naming of the storms. Rather than using ‘he’ and ‘she’ for storms, we need to employ non-gendered assignations; at least until we can move away from associating certain names as gender-specific, and associating certain behaviours to a specific gender.