Although I admire your desire for researching more variables you are failing to account for all the anecdotal evidence from years over years of the same scenario with almost no variation: player is accepted, valued by peers, gets items. The second mic shows a female voice the game changes. Players start killing the player nonstop, loot them, ask in COMM “tits or GTFO”, saying that they will rape them, checking their profile for pictures and start PMing asking for nudes.
This level of harassment is meant to make the player quit.
I am from before games were skewed towards boys (good old 80s) and as early as 1994 I already avoided using gametags with female nicknames. I avoid multiplayer like the plague unless I’m sure I can play at that server without being annoyed to death by a teen with Tourette’s.
Yes, you will find some level of racism with strong accents but younger males are the same demographic as the harassers. Their voices are pretty much the default. I have a hint that the data on such a graphic would show a very high bar for female voices in general, followed by a proportionately small bar on male voices with foreign accents and an almost flat bar on young male voices. Of course the plural of anecdote isn’t data. If you do the research please link me to your findings, I would love to see the results.