I don’t know which of my fellow media-studies academics, or cultural critics, or amateur statisticians, or anyone else doing statistical analysis of fandom needs to hear this.... but you cannot use AO3 stats in order to make assumptions about the broad popularity of this or that character or ship or attitude in a fandom. Because AO3 is not remotely representative of fandom as a whole!
This is for several reasons, but three big reasons:
- Fanfiction writers and readers are a (tiny) minority of fandom
- AO3's content biases relative to other fanfiction websites
- AO3's recency bias
Let's get into these in more detail under the cut, including some examples that illustrate how each of these factors will bias your results if you are exclusively looking at AO3:
merlin-gifs
