If it's the study I'm thinking of [1], they essentially tried to measure how much more actual merit hires would need to have to justify the inequality of hiring decisions:

> Despite the confounded nature of merit and social status within measurable prestige, the observed hierarchies are sufficiently steep that attributing their structure to differences in merit alone seems implausible.

It's not the strongest argument, but I'm having trouble coming up with a way to separate it out. You could try to measure the subsequent productivity of professors once hired, but it's likely that scientific publishing and citation also has a prestige bias, so the confounder is back.

[1] http://advances.sciencemag.org/content/1/1/e1400005

I concluded that in that line, the author was referring to the statistics she had previously discussed. But I have to admit that it's ambiguous.

When referring to social "scientists" the quotes should be obligatory