Also, the Bechdel Test is incredibly misleading and distracts from the actual problem IMO, which is simple underrepresentation of women in movies. If only 20% of movie characters are women, probably 4% of dialogue (20% of 20%) would end up being women talking to each other. Solve representation, not content.
It's informative to realize that a substantial portion of movies also fail a gender reversed version of the Bechdel test (though sure, smaller than the original test)... Not only does it reflect the representation in main roles, it also counts films which are about men and women interacting as fails.
Lets see, last night while you were sending the comment I was watching "turning red" -- AFAICT it fails it: There was no scene where two men talked to each other about something other than a woman.
Most movies have most or all of their dialog about the important characters. When the important characters are disproportionately men then thats what the dialog will be about. When the subject is about the interactions of some differently gendered characters, then the dialog will often be about a different gender than the character talking.
The prior posters point was that it reflects the under-representation of women as substantial characters more than anything else-- I'd also add that it also reflect the further under-representation of women in movies that aren't about relationships. We don't need to invoke a complicated test to observe that women are underrepresented in serious roles and stereotyped into certain kinds of plots when they do have significant roles.
https://bechdeltest.com/view/6242/mad_max:_fury_road/
Also, the Bechdel Test is incredibly misleading and distracts from the actual problem IMO, which is simple underrepresentation of women in movies. If only 20% of movie characters are women, probably 4% of dialogue (20% of 20%) would end up being women talking to each other. Solve representation, not content.