Recall: the tests are 28 multiple-choice questions. I calculate the grade out of 25. This allows me to ask testing questions, so that a good student can still get an A even after they have got a few questions wrong. I love this system. Fewer complaints, more stretching. My aim with tests are to force students to think hard, generate teachable moments, tell the students and me who is not understanding what (i.e. kick them and me into action) and then, least important, return a grade.

Class average: 79.7% (**C+**).

**A**, 34, **A-**, 18, **B+**, 16; **B**, 16; **B-**, 18; **C+**, 16; **C**, 18; **D**, 23; **Fails**, 11, including 4 no shows.

I have now set nine SC200 Class Tests over the three years it has been running, and only one student in one test got everything. It wasn’t this time. But one student came really close (27/28). Five students got just two questions wrong, and 13 students got just three questions wrong.

This test really, really opened the class out. And in perfect symmetry: 34 students with A’s; exactly the same number with D’s or Fails. When I get a chance I will have to figure out what is going on here. Is it older students or students doing science-related majors at the top? Worryingly, there are some folks at the bottom end with good attendance records.

Also looks like I have a teaching challenge on my hands. This is not a bell curve, not even close. It’s the shape of a shallow dish.