In your rush to attack hooande for not reading the entire article, you neglected to read the entire comment and missed the part about organized group effort. If you consider the prospect of being a teacher who is gay, Muslim, an immigrant (or one from the wrong country) etc. it's quite easy to believe that the risk isn't limited to a single student in one class.

Fortunately the authors weren't so careless — that concern is obviously key to the repeated cautions about assigning too much weight to such tests. I'd like to believe anyone setting policy would be similarly slow to jump to conclusions but this is a field notoriously prone to chasing fads and quick fixes…