There's a strong tendency, in some environments, to view equality in terms of equality of outcome. E.g. women are not and will not be equal to men until they compromise 50% of engineers, 50% of CEOs, etc. Coupled to this is an assumption that any inequality of outcome is indicative of bias or discrimination, and thus positive discrimination is valid to rectify it.

The notion of population level differences is not good for such people because it would men that disparities of outcome may exist without bias. It's extra uncomfortable for places that engage in discrimination to rectify disparities of outcome, because it fractures the justifications for these policies. E.g. if we accept that women might make up 20-25% of engineers without societal bias, the discriminatory policies to try and hire 50% women doesn't very fair any more.