My Cooperative Learning class had an awesome all-day session with Laura Terrill yesterday. It was a bit overwhelming in that she touched on so many different topics--assessment, cooperative learning, course structure, social emotional learning...it was a lot to take in. I was pleasantly surprised because she is not "officially" a CI/TPRS teacher, but almost everything that she said was completely in line with CI methods. I'll post more about that later, but to begin with I'd like to post my take-aways on assessment.
I strive to used Standards-Based Assessment in my courses. I say "strive to" because it is different than how I've been assessed in all my years of schooling, and so it takes constant reflection on what exactly is being assessed any time that I develop a new potential assessment. Anytime that I hear about a great, new assessment idea and run to use it in my classroom, I have to stop, think, and critically reflect on whether or not it is giving me a true read of a student's mastery of Standard X. Since the class was Cooperative Learning, Laura gave us quite a few ideas on how to give group grades effectively. She had some wonderful ideas that would work very well for a more traditional grading system, but I don't see how group grades could ever fit into a standards-based system. Her students all know that the group is responsible for ensuring that each individual in the group can perform whatever task has been assigned. She can then collect just one paper or solicit just one response from the group and give the entire group that grade. (They are always allowed to come in and re-take it individually.)
However, it seems to me that this would not work in a standards-based classroom because the response really becomes a measure of the ability of the highest-performing student in the group. That student can feed the answer to the other students in the group and make sure that they have it correctly memorized. In an ideal world, that student also explains why that is the correct answer to his or her group mates and they glean some new understanding from it, but I would surmise that that is the exception and not the rule. It would make my life so much easier if I could just collect eight papers to grade instead of 32+...this would be a good way for me to give out Citizenship grades (participation falls into that category, which accounts for 5 percent of their overall grade), but that's about it. 
One quick-assessment strategy that she uses and I love is having each student's name on an index card with five rows of numbers 0, 6, 7, 8, 9, 10 on it (the only way you get a 0 is by not answering), and she calls on students to answer questions randomly to assess their speaking ability. I created my own this morning that uses rubrics instead of number scales. The rubric is essentially the same as the one that I use on my free write forms, but some wording is different and I added a pronunciation and speech fluency category. I am going to copy these onto card stock, cut them out, and use them in the same way that Laura does. Any time that you give the students a speaking assignment, you can either call on students or meander through the room with these cards to assess an individual's speaking proficiency. I included lines to write three different dates on it (ideally, I'll assess each student three times throughout the quarter), and I can just write the number of the assessment (1 for the first date, 2 for the second, 3 for the third) in the appropriate box in each category. At the end of the quarter, I can see into which column most marks fell, and voilá! Speaking grade done. Here it is: Speaking rubric.
Another point that she made and I loved was that the worst thing that you can do before an assessment is review. When you review material immediately before an assessment, you place that information into your students' working memory. You want to know what information has made it into their long-term memory. If you really want to know what they know, you should spend the 10-20 minutes before an assessment teaching them new material, and then assess them on their acquisition of the old material. As long as you are still working with material, you really cannot assess it accurately because it doesn't guarantee that it has made the move into their long-term memory.
