How much of that benefit was mixing up cause and effect? As in, one might hypothesize that having a visually diverse company is a luxury. But we don’t say that wearing a Rolex causes you to be successful.
Second, diversity in e.g. race/gender may be a proxy for some other kind of more functional diversity. In fact, for diversity to matter at all there must be some functional element which can be decoupled from the labels people apply to themselves. The fact that we just assume this latent variable is there without measuring it seemed very unscientific to me.
>for diversity to matter at all there must be some functional element
>without measuring it seemed very unscientific to me.
Well, I think it's accepted that there were "unscientific" elements here. I'm not saying everyone agreed on their value, but they were stated as benefits.
For instance, diversity in viewpoints based on varied experiences would theoretically bring about more interesting solutions. I'm not sure how one would objectively measure that purported effect.
Likewise, helping a company to understand cultural sensitivities among different customer populations, so they don't blunder in language or otherwise.
And, to whatever degree we consider it beneficial for companies to be considered "good corporate citizens" in general, I suppose that would apply here as well—though, again, hard to measure.
So, I agree that say, race/gender are proxies for deeper cultural diversity. I don't know that you would define that diversity as "functional", or that you would need to in order for there to be some benefit.
Second, diversity in e.g. race/gender may be a proxy for some other kind of more functional diversity. In fact, for diversity to matter at all there must be some functional element which can be decoupled from the labels people apply to themselves. The fact that we just assume this latent variable is there without measuring it seemed very unscientific to me.