[Update: check out the comprehensive resource.]
Over the weekend I wrote up a mini-thesis on my assessment methods, which, though standard operating procedure at my last school, are pretty foreign here. I gave it a deliberately confrontational title, "How Math Must Assess." SOP here is to use whatever tests the manufacturer supplies and give them whenever the manufacturer arbitrarily decided to divide the textbook. I worry that this arbitrary approach to testing stirs up a lot of hatred for math. The implication I try to avoid, since it's so cocky coming from a third-year teacher's keyboard, is that I know how to do it better.