Researchers have claimed that people are more likely to prepare for a hurricane with a male name than a female name due to implicit sexism, in a study published in PNAS which is, more than likely, incorrect.
The study, "Female hurricanes are deadlier than male hurricanes," attempts to figure out why (slightly) more people die in storms with female names than in storms with male ones. In order to test the theory that this is because people are pretty fundamentally sexist ("warm fuzzy, Tabitha won't hurt us as much as mean old Harry, so we won't take precautions") the researchers asked six groups of volunteers a number of hypothetical, storm-related questions that hinged on each storm's names. For example, volunteers who saw lists of hurricane names without descriptions assumed those with male names would be stronger. Equipped with more detailed information on the storms, they still thought "Alexander" would be more dangerous than "Alexandra," and so on.
But a few methodological flaws hamper the theory. National Geographic's Ed Yong spoke to the National Center for Atmospheric Research's Jeff Lazo about what went wrong. Lazo explains that the data set used by the researchers is skewed because male hurricane names weren't used at all until 1979 — by which time storm deaths had decreased significantly. The researchers write that they split the data set in two, pre-1979 deaths and post-1979 deaths and got comparable results, but those new sample sizes are too small to judge by.