"In the absence of [Variable], the results would be the same across [Variable]."

Technically correct, but the statement is meaningless. The assumption that we can reliably design statistical controls for something as pervasive to society as racial differences is rather optimistic, to say the least