I don't know Teachstreet - but checking my theories/tests, it does fit the model of what has been hit. I took a snippet af exact match content from a Teachstreet directory listing and put it in google: http://www.google.com/search?client=safari&rls=en&q="Get+ready+for+the+SATs,+connect+with+a+math+tutor+or+improve+your+LSAT+score"&ie=UTF-8&oe=UTF-8
So what I see is Google taking note of exact;y what it would take note of - trying to put the original source of the content at the top and penalize (ignore) content tat mimicks the original.