60% of any cohort being "black or female" sounds like the population average or below. 50% of people are female, and more than 10% in the USA are black. A random selection of citizens would likely turn up as 60%+ black or female.

I'm not saying this firing wasn't statistically suspect, since I don't know the demographics of that cohort, but I'm guessing based on this misleading phrasing that it wasn't.