Andy Resnick said:
A guy from U. Minnesota told us that rather than randomly grouping students for small-group work, they intentionally put together teams consisting of a good student, a mediocre student, and a poor student. My comment was that groups then can't be assigned until well into the term.
In the med classes, we have tried various combinations of assigning groups. We've tried it based on their previous semester scores (since we teach second semester), and we've tried it completely randomly, and we've done it alphabetically, and we've done it so that they have the same group all semester, and we've done it so they switch groups two or three times a semester.
Others who use team-based learning for a variety of subjects in a variety of schools and levels have also looked at the impact of group assignments.
Basically, there are a few things to avoid, but otherwise, it does not matter. You want to try to balance for prior experience in a subject. So, if I'm teaching anatomy and some students in my class have taken an anatomy course before, I would want to break them up among the groups. Doing well in another subject is not necessarily a predictor of doing well in the subject I teach, though the students who do very poorly in their first semester may indeed be weak students who will do poorly second semester too, so we split them up if they all end up assigned to one group. Otherwise, random assignments work fine.
The risk of trying too hard to balance groups by putting a good, middle and poor student into each group is that you set up a situation where the really good student ends up carrying the whole group and the group let's them do all the work because they get things right.
I would not use the top students as examples. First, I don't like putting any student on the spot in the classroom. It makes the environment very uncomfortable. Second, I'm not sure that the class watching a good student breeze through a problem is necessarily helpful for them to learn how to work through the spots where they got stuck. Having a good student demonstrate how to do a problem is not really different from me demonstrating for them. Instead, having a mediocre student do the problems probably works better. They aren't so lost as to not even be able to start or to look like a fool in front of the class, but will have places that they can't just figure out the next step on intuition alone, which presents a teachable moment to explain to the class HOW to figure out what that next step should be. I also would not want to make one student appear to be the "teacher's pet."
Yes, there are students who only associate with other students not doing well in the course, and who think nobody passes and it's all impossible. To let them know that it is not impossible, and because they do need the feedback of how they are progressing compared to the rest of the class, for each exam, I post the exam average. Because some students were getting into that rut this last exam, thinking nobody could have done well, I posted what the high score was too. (This time around, someone DID get a 98%.)
After final grades are announced, I post a grade distribution curve so they can see that a lot of students do get As and Bs, the majority get Cs, and some do get Ds and Fs in spite of everything.
I'm actually going to try something a little different on the student evaluations of the coures this year. I want to see how those evaluations vary by where students sit in the classroom...just roughly divided by front, middle and back of the room. To do this, I'm putting on a question about clarity of visual aids (i.e., text on slides), and then asking them to tell us where they sit in the classroom so if improvement is needed, we know more about where the visibility drops off. But, I think we all suspect, but nobody quantifies, that the students who sit up front do better and pay more attention and generally like classes better than those sitting all the way in the back of the room. I think this is important to identify, because the classroom I teach in has double the seating capacity of my class size (to leave us room to spread out during exams), and that means nobody HAS to sit in the back if they can't clearly see the lecture slides from there. So, if the students in the middle can see, and those in back can't, then that tells us something about their motivation to really pay attention to the lectures. (I'm also doing this because I KNOW the other lecturer in the course uses slides with text that is too small to see all the way in the back of the classroom; I've pointed out to her to be careful about her font selection because of the size of the room, but she hasn't done anything about it, so I want her to get that feedback from the students too.)
Anyway, I really am enjoying this particular discussion, because the general statement that Andy Resnick made about too many instructors wanting to just blame the students is certainly consistent with my observations. I'm glad to see DenverDoc starting up a discussion that recognizes that when students aren't doing well in a course, it means the instructors need to do something different to engage them, get them caught up, explain things better, etc. I cringe when I hear other faculty who have been teaching a long time who get their evaluations and say things like, "What do the students know?" or "Oh, they always complain," or "Don't worry about what the evaluations say, they don't mean anything anyway." Now, granted, I have seen bad evaluation questions that really aren't going to be helpful other than as a means for students to vent their frustrations (one they ask our med students is "Was the material presented in lecture of an appropriate level for first year med students?" Huh? In that case, yeah, how would they know if it wasn't? That's something faculty peers should be addressing, not students). But, when evaluations are consistently bad, it means you've really missed the boat on how to approach that class.
Some of the other faculty in my department have shared with me the critiques on their teaching evaluations, because they are sitting there laughing about it, not taking them seriously and just venting their frustration at having to be evaluated by students. Since we team-teach in that course, I've sat through their lectures and been in the labs with them. You know what? I agree with the students in most cases.
By the way, last year, I knew I was going to be taking over as coordinator for the course I'm running this year. Since the previous coordinator couldn't care less what was happening by then, just happy to no longer have to coordinate it for the future, I got free reign to write up the questions for the course evaluation. I KNEW there were problems with the course, knew students didn't like it, knew there was VAST room for improvement. So, I didn't write the standard questions. I didn't need those to know they were going to rate things as horrible. Instead, I wrote questions that really got to the points that were important...how can we improve? So, instead of them telling me they didn't like lab (attendance already demonstrated that), I asked questions like, "Given the choice, would you prefer an unstructured lab or assignments due in every class?" "Did you usually work with the same group of students for every group activity, or did you change groups each time?" "Would you prefer to choose your own groups, or have them assigned by the instructor?" "If a new textbook were chosen for the course, should it have more or less detail?" "If a new textbook were chosen for the course, should it have more or less text?" "If a new textbook were chosen for the course, should it have more or less pictures?"
I actually got a lot more constructive criticism and good insight from the students from that approach than when we use questions like, "Did you like the book assigned?" The responses were not all what you'd expect either. The students WANTED in-class assignments. They WANTED attendance taken. Overall, this year students are more engaged in the class, they show up, even those who aren't doing well are trying (except maybe 2 or 3 who are still just drifting). Exam scores are up. The students all seem to be having fun and enjoying learning, even as they are struggling.
I also have put up a "suggestion box" on my course website, where students can anonymously give feedback for improvements all during the term. I do tell them that the two things I can't change are to teach less material or to make the exams easier. But, I tell them that if they have suggestions of ways we can help make it easier to learn all that material or do well on the exams, that will certainly be considered.
There is one aspect of the course that I let them decide how it will be done, and those are the team activities they do. On the first day of class, we did an exercise that was not graded for them to get an idea of what the activities would be like. Then, I let the class vote on how many points future activities were worth (I gave them a range to choose, of 1 to 5 points per question...this wasn't a "sky's the limit" question). When I started out the course, I was doing it as all-or-nothing. They weren't thrilled with that, because they lost a lot of points if the got close but not quite. So, based on their feedback, I started giving partial credit for close-but-not-quite answers. This has actually meant the in class assignments got harder, because I can't just give them answer choices that are right or wrong, but have to have a lot more shades of really close but not exactly right for some reason to have a basis for awarding partial credit. They really seem to enjoy that. However, I then got feedback from some groups that they weren't always in agreement on the answers, and half the team would overwhelm the other half and give the wrong answer, which didn't make the students who knew the right answer happy. So, I've added a new component to this "game" which is that they can have a mutiny in the group, but they better be absolutely sure they want to disagree with the group before risking mutiny, because if they choose to disagree with the group and are wrong, they will lose all the points. (I won't give them partial credit on those because I don't want it to be an individual assignment where they all just give their own answers and stop working as a team, so there needs to be a big risk to disagreeing with the group that someone must be very confident the group is wrong before dissenting.) Overall, these activities really work out to be a very small percentage of their grade. But, giving them ownership of that aspect of their grade really has given them a pretty positive outlook on the course.
NEVER before in all my teaching have I had students come up to me and say, "I understand WHY we are doing these activities, but..." when they had something to critique or suggest about it. I'm going to add more of these activities each year, because they really do learn from them. It just wasn't feasible to do them every lecture the first year because it takes time to write really good questions for that sort of activity.
I make them all case-based and clinically relevant...things like, "You are the first to arrive at the scene of an accident" and then I go into an explanation of some injury. I.e., "You see a large chunk of metal protruding from the lateral side of the victim's arm..." and have them answer questions about the anatomy...are there major blood vessels at risk and which ones, what nerves could be damaged, what muscles might be damaged, what functions might be compromised and what future disabilities might this patient face if the damage can't be repaired, etc.
You can do the same thing with some basic math too if that's what you need to teach them...I might ask something like, "A physician has ordered a patient be given an i.v. drip of a medication at a rate of 10 mg/h. You hang a 250 ml i.v. bag of saline and set the flow rate as 20 ml/h. The medication is dispensed from the pharmacy in a vial containing a solution of 250 mg/ml. You add 1 ml of the medication to the i.v. bag. A few hours later, the patient dies as a result of an overdose of this medication. Who needs to call a lawyer, you or the prescribing physician? Explain."
I have NO qualms about mentioning things like patients dying or malpractice lawsuits when posing questions to my students...nursing, med or dental. They all need to be aware that they are not learning this stuff as busy work, but because people's lives depend on them getting it right EVERY time, not just 70% of the time.
), because they can figure them out just fine when they really put their collective minds to it. But, then they screw up the simplest questions on exams. 
I tell them this bluntly, because they wouldn't want a nurse treating them who doesn't know that stuff either. 
