Back to the roots...
They need to go back to the source of the idea of bimodality in CS ability (http://www.eis.mdx.ac.uk/research/PhDArea/saeed/paper1.pdf)
"The camel has two humps" was a paper which seemed to show a bimodal distribution of ABILITY in people BEGINNING a CS course. The grades weren't the criteria, rather a test of ability given to people who had not as yet attended a CS course.