Glenn McDonald believes that music is the thing that humans do best, and that you don't have to understand people to want to hear them sing. At Spotify he variously attempts to quantify awe, turn arguments about counting into conversations about love, and extract new particles from the Large Genre Collider. At previous Pop Conferences he has discussed music cartography, algorithmic morality, patterns of protest and escape and disgust, and other pretexts for symphonic metal.

Panic, Death and Other Gender Patterns in Spotify Listening
In September 2016 Spotify added a third gender choice, labeled "non-binary", to the service's signup form in the US. In the first year of this possibility, over a million new Spotify members selected non-binary. By November 2017 the non-binary signup rate was at about 2.5%, and this group was collectively streaming over 20,000 days of music per day.

This has thus produced one of the largest datasets yet compiled on the listening behavior of people who identify as non-binary, and we may now begin to attempt to answer questions that until now were mostly hypothetical, like:
   -How different is their listening from the full population, or from listeners who identify as male or as female? (They listen to 30% more music, for one thing.)
   -How is their listening different when it's different? (They like Panic! At the Disco a lot.)
   -How does non-binary gender identification relate to age, and is non-binary listening different in different ways at different ages? (Yes.)

And having asked these questions about listeners who identify as non-binary, it makes sense to also ask them about people who identify as female or male. Plus:
   -What artists and genres skew most distinctively towards listeners of a particular gender? (The most dominantly male-listened genre is progressive deathcore.)
   -How do listeners' genders and performers' gender relate, and how does this relationship differ across genres? (The most female-listened major artist is the all-male K-pop boy band BTS.)
   -How do Spotify demographics differ across geography, and is that interesting?
   -Could these insights be used in music programming and personalization to confront and reduce gender bias?

These and many related questions will be partially answered or earnestly evaded in a statistical tour, voluminously illustrated with numbers and music, of what we think we've learned so far.