It would be nice to see some control as well, like filtering out movies that do not make back their budgets (within reason), as they are probably just terrible movies regardless of diversity characteristics.

Or maybe a minimum gross filter of 500K?

I added filters for minimum 10M domestic revenue and released 2000 or later for various data fidelity reasons. (I do not have access to budget information so cannot account for that) Sample Size is still 2,012 which is good enough.

Spoilers: yes, movies which star male actors (fitting this criteria) earn significantly more than female leads on average, and it is statistically significant.

Additionally, there is no difference (practical and statistical) between RT/Metacritic scores of movies with male leads/female leads.

Which is odd because the box office os 50/50 on gender.

http://www.mpaa.org/wp-content/uploads/2015/03/MPAA-Theatric...

Thanks for that link. I'll include it in the post.

Is there any way to control for "hype budget"?