So this study is saying "gender is a factor in how people see storms being dangerous" and "people who see storms as dangerous are more likely to pack up and leave". It is not saying "Hurricanes with female names kill more people because people don't respect them" as I believe it has no real data to show this, only to show the above. Nowhere does it say that is a significant factor. The article also mentions that all hurricanes used to be given female names, and with modern perception of hurricanes, it is likely that since male hurricanes are later, and lower number of deaths are later, it may be that that is a far larger driving factor. The data uses hurricanes from 1950 on, where the practice of having male hurricane names came after 1970. In other words, this study is showing that people find things with female names as less threatening, not that people actively are killed by hurricanes because of the gender of their name. The writers did mention a response to this rebuttal, saying that they measured the "femininity" even before 1970, not just a binary "male or female" system, however, the hurricanes will still skew female for the early, more deadly, hurricanes. Finally, I believe they used a modern group of people to determine femininity, not people from those eras, and many names change if they are either gender over the years, so results are not accurate for past decades, or not as accurate. Laboratory experiments indicate that this is because hurricane names lead to gender-based expectations about severity and this, in turn, guides respondents’ preparedness to take protective action.