1.1 Introduction

The issue of gay death in tv became particularly relevant in the media landscape in March of 2016 when tv drama ‘The 100’ killed off its most prominent lesbian character right after she got into a relationship. This was quickly pegged as an example of the ‘bury your gays’ trope in which gay characters are killed off immediately after they enter relationships or immediately after they come out.

TV fans didn’t take this all lying down, though. Groups of queer fans orchestrated huge social media campaigns against shows that killed off queer characters, some ran a fundraiser for American queer helpline The Trevor Project that raised over $160,000USD, and the charity “LGBT Fans Deserve Better” was formed with the express purpose of promoting better queer representation.

A few brave journalists decided to tackle the problem from a statistical angle. Vox journalist Caroline Framke figured out that 10% of the deaths in the 2015-16 TV season were queer women, which seems pretty high. Marie Bernard and Heather Hogan of Autostraddle worked together to make a cool infographic of the overall state of queer representation in TV. In the academic sphere, Meredith Bartley and Kira Deshler each took a shot at calculating the percentage of queer female characters who died (~40%). LGBT Fans Deserve Better also released some statistics, though this happened after I started my project and therefore didn’t influence the development.

Sadly, every analysis has its flaws. The 40% death statistic quoted by both Bartley and Deshler is unrealistically high; it comes from looking at all deaths in the world as a proportion of american characters, rather than in comparison to all the characters in the world - I calculate the correct version of this statistic later on. We also currently have no articles discussing long-term trends in queer death, such as whether more characters are dying in recent years than in the past, which makes it difficult to know whether 2016’s high death rate is a fluke or part of something broader. The biggest problem, though, is that all the statistics we have right now could just be due to random chance. We know that 10% of dead characters in 2015 were queer women; we don’t know if that’s unusual enough to point to a broader issue with TV. There are types of statistics - called hypothesis tests, and explained down in the Analysis section - that can tell us whether the percentages we find are caused by chance or not, but nobody’s used these tests to check their results yet :(