Rejection rate might give a clue to intent - but even then not really cause it would be hard to tell whether the difference in rejection is justified by other non-racial factors. I wrote the article to show the effect of the policy not the intent of the implementation and for that the metrics I used were appropriate