Phys.org

This is correlation and not necessarily causation, but it does line up with an important argument for DEI programs.

The argument is that DEI programs are for raising the bar; not lowering it.

This is based on the assertion that ability is equally distributed across different groups but opportunity is not. If you take that as a given then it stands to reason that overrepresentation of one group over others would be caused by less capable members of the overrepresented group being favored over more capable members of the other groups.

So, raising the bar would correct for that and result in greater diversity.

That being said, whether or not the test itself is biased would play a critical role in whether or not you actually see that corrective effect.