The Story So Far....

Here's a post from halfway through my first semester using TBL (March, 2011):
Poor neglected blog. Your blogger has been busy practicing a new teaching technique. Now that I am halfway through my first semester of using team-based learning, I can reflect on what I have learned so far:

• I'm learning to back off. My students have worked together in groups periodically since I started teaching back in the 20th century, and I had developed the habit of circulating while groups work together, checking on their progress and answering questions. Now I'm finding that if I do that, the students ask me things they could ask each other. I want them to ask each other questions, so that they teach one another. So no more unnecessary circulating.
• I'm learning that asking teams to practice skills by writing even brief texts can be counterproductive. Michaelsen points out that writing is inherently an individual pursuit and so team activities should focus more on making decisions than on producing texts, and that always made sense to me. But I thought I could ask teams to write a few sentences summarizing a three-paragraph passage (as a way of practicing summary for research papers) and get away with it. I have found that for such a task, each student will write their own summary with little input from teammates. So next time I'll provide a few summaries and ask them to decide which of them is most effective.
• I'm learning to put the Readiness Assessment Tests in perspective for my students. RATs dominated the first weeks of the semester, as we had a practice RAT the second week, RAT1 the third week, and RAT2 a couple of weeks later. However, there are just 5 RATs in all, and all together they count for just 20% of students' course grades (while the 5 formal writing assignments count for 70%). Students' attention, efforts, and anxiety were disproportionately directed to the readiness assessment tests. It didn't help matters that I used a 105-point scale to grade the 15-question tests. Next time: a 15-point scale and a preventative sermon on the role and value of RATs.

The Nuts and Bolts

Here's a post from early in my first semester of using TBL in ENGL 1311 (now called WRIT 1311):
In team-based learning, students take periodic "readiness assessment tests" to determine how well they have prepared (by doing assigned reading) for the next major unit/topic/assignment. It is important that the tests be graded and returned immediately so that the instructor can immediately clarify concepts/material that students had difficulty with.
To allow for immediate evaluation, RATs usually consist of 15-20 multiple choice questions. But even brief multiple choice tests take some time to grade, unless you have a Scantron machine in your classroom. TBL practitioners have come up with various solutions. Some use transparency overlays; others give students scratch-off answer sheets.
And some use a cordless drill. I'll let the photos speak for themselves.

The student fills in a test answer sheet.

The instructor's key. Shaded cells are correct answers.

The primitive Scantron machine

A big stack of completed test answer sheets, with the instructor's key on top.

The instructor drills through the correct answers.

Instructor quickly grades student's test answer sheet.

How Students See Themselves

Here's another post from early in my first TBL semester:
Michaelsen and others are adamant that practitioners of team-based learning allow students to participate in the formation of teams at the beginning of a semester (although they are just as adamant about instructors controlling team formation, doing much more than telling students to create groups of a certain number). Teams function best when they are diverse so that "member resources" (such as work experience, college experience, access to technology, varying perspectives, etc.) are distributed among the teams.
I asked my ENGL 1311 students to brainstorm a list of factors that affect team performance. One class came up with about a dozen factors. After a complicated vote, we had our top five factors: personality (extroverted/introverted) gender, age, race, and parenthood. I asked students to privately give themselves ten points for each category (e.g., 10 points if you are male, 10 points if you are over 21, etc.). The students stood in a line, with those having the highest scores at one end and those with the lowest at the other end (and yes, I explained that the points were not meant to assign value to one or the other gender, age range, etc.).  The students then counted off (1, 2, 3, 4; 1, 2, 3, 4; etc.) and joined forces with those who had the same number.
I really didn't mean to get so bogged down in describing the details of the process. The point is that students did not form the groups themselves (which leads to groups that are too cohesive--friends stick together, for example, or the older nontraditional students find one another), and I did not form the groups myself (which makes students wonder what the instructor's ulterior motives are, even if the he or she explains the factors used to form diverse groups).
Of course the jury is still out on whether the teams are diverse enough to become lean, mean learning machines. But I can tell you that each team has exactly one male student.