Fixing Flawed Ideas: IIb
Note: I submitted a previous version of this to the Carnival of Education a couple of weeks ago, but Wednesday morning when the Carnival went online, my hosting service belched, and I lost everything I’d posted over the previous 48 hours, including this. I did manage to restore some posts, but not until Thursday, and of course, the URL was no longer the same as that posted on the Carnival. So in response to one of the comments left here (all the comments were lost), I’ve decided to expand this, and submit it to this Carnival.
Some years ago, I was asked (well, you know how that works) if I would teach another course (not a problem, since I’d taught it before, just not in a few years), and if I would take over the development of course materials (in a business data analysis course, essentially applied math). Of course, it wasn’t really a request, so I said sure.
The course was being redesigned which was a good thing, because it was an incoherent mess. It had been passed around for years, and since nobody wanted to recreate it, each time, somebody would drop something here and add something there. The end result was a course that had no goals, a course of disjoint components, a course students hated becuase they couldn’t figure out what they were supposed to be learning, and faculty hated teaching even more.
Like I said, it was a mess.
The first two things that needed to be addressed were course goals and coherence. After a lot of fighting — because everybody had his favorite little pet component — we got that out of the way, and I could start on the materials.
I dug through textbooks, looking for ideas. There is no textbook that by itself would have worked, so that was out — not a good thing, because that meant more work for me. A lot more work. But in the end, it was very fulfilling work, and this is an overview of what I created.
Classroom Cohorts
Even if we had found a usable textbook, well, let’s face it, textbooks suck. The goals and curriculum of the course were broad, yet we covered each component to a fair degree of depth, so it was clear that we needed both individual and cohort assignments. The problem, or one of the problems, with group work is that everybody hates it (except the slackers). And since we give a great deal of weight to student evaluations, I had to come up with a way to make cohorts palatable.
I stole one of my ideas from earlier classes. Instead of creating groups that change from assignment to assignment, we would form cohorts near the beginning of the semester, and students would remain with their cohorts until the end of the semester. Each cohort would create a business identity, to lend realism (and because students really really really like that).
The advantage of cohorts is that you can give students far more complex and time-consuming assignments than you can with individual assignments. And in this course, that was a necessity.
We needed procedures for students to either leave their cohort (this one was easy, but if students left, they had to do all the assignments on their own, so it rarely happened), and we needed a way to control for slackers. We created contracts for the cohorts to sign and turn in with each assignment: The cohort assigned a percentage to each member of that cohort, and in order for a grade to be assigned to the cohort, every member had to sign the contract. So there was no whining later on that somebody got screwed — well, there was, but far less of it than usual.
So if a cohort got 90 points out of a possible 100 on an assignment, and Johnny had been given an 80%, Johnny would get 80% of the 90 points, or 72 points. That worked exceedingly well, and put the burden of negotiation on the students, where it belonged.
Did the scores assigned reflect the amount of work each student did? There’s no way to know. However, for us to record the grade, all members of the cohort — including Johnny — had to sign the form, so Johnny consented to that 80%. If, on the due date, there was a conflict within a group about assigned points and workload, we met with the group and ironed it out.
We had a couple of bleeding-leart leftist colleagues who wanted women-only groups, minority-only groups, and other such identity politics nonsense, so they would feel “safe.” We nixed that idea because in the real world, nobody cares how you feel, much less whether you feel “safe” from imagined threats or not. We also had to battle a push for giving “handicap points” to so-called “disadvantaged” students; we nixed that because life is not a golf game.
We did have a few rare instances of nutball students creating problems, but for the most part, the problems took care of themselves.
The Case for Case
One of the reasons textbooks suck so bad is the way they’re organized. At the end of each chapter there is a list of exercises, all utterly unrelated to any other exericse, like little context-free soundbytes. I wanted to avoid that at all costs, because the one thing our students hate more than anything else is busy work — and unrelated exercises always seem to students to be busy work, even if they are not.
Using a case study creates a context for all the work students do, and pulls everything together. Each assignment, therefore, had to in some way address some problem presented by the case study. Each of the cohorts was a consulting firm, and they had been hired to solve the problems presented by the case study.
However, there are loose case studies — case studies that are sloppily thrown together, and only barely fulfil the functions they are supposed to fill — and there are tight case studies (more about that below). I wanted a wholly case-based class. And this was a problem.
I had written a case, but I had to completely rethink and rewrite it. I needed a case that presented exactly the types of problems we would learn to address in the class, and do so in a way that assignments could build one upon the other, and no other problems, or students could get sidetracked. It was difficult to do, but I managed.
Writing the assignments was tricky, since I wanted one to feed the next. It was even trickier because we presented material in order of complexity, which doesn’t necessarily map onto the order in which you would address them in the course of the problem-solving process. We didn’t have the luxury of assigning projects “out of order” (compared to classroom presentation order) because it was an absolute necessity that at any point during the semester students would be able to see exactly how they were doing — and “out of order” assignments would have delayed testing some components.
The solution was to give individual assignments in “presentation order,” where each individual assignment tested their ability to solve one kind of problem, and combine several problems in cohort assignments. This way, students were being continually evaluated on the skills we covered in class, while we could present thornier, more complex problems to the cohort which integrated two or more of those skills.
This did create a rather bizarre schedule for cohort work, where more cohort work was assigned in the second half of the semester than the first, but we had to live with it. And it enabled us to tweak cohorts after drop and add.
There was also another problem writing the assignments. Each had to be written so that there was only one way to solve the problem set, since we were testing not only their problem-solving skills, but whether they had learned how to use the tools we had taught (note that this is directly opposed to “fuzzy math” taught in the public schools). A couple of times, I had to change the case study information in order to write assignments with only one path to the answer.
In short, writing the case and assignments was a recursive, and not a linear, process.
The Result
We dropped much of the previous course content, because it had little (if anything) to do with anything else in the class, and because in some cases, we could not assume students had the necessary background to understand it. We kept a core which was coherent, and built onto it with other skills that had never been covered before in that course. We ended up with a tight, integrated curriculum.
The cohorts were as much of a success as they can be, given the problems of working in groups. The contracts for percentage allocations worked extremely well — on the first assignment, everybody got 100%, but few cohorts assigned everybody 100% in the following assignments, so slackers did not slide.
We couldn’t have done the course with all cohort assignments, or so we felt at the time. Individual students would not have been evaluated sufficiently. There was discussion about increasing the weights of the two exams to counter grade leveling and the lack of evaluation, but it wasn’t necessary.
Individual assignments also acted as screens for cohort work. Doing well on the individual assignments meant the student had mastered the skills to work well for his cohort. The course was work intensive, but it’s a university. Life is like that, and students expected intensive work — just not busy work.
The final cohort assignment was a formal report and presentation, giving the hiring company (that would be us) the cohort’s recommendations for how best to solve the problems presented in the case. Students turned in the report, and gave group presentations the week before finals week.
Students liked the course, and felt that they were learning valuable skills, where before, students had despised the class (almost as much as those teaching it did). In the end, that’s what really mattered.
Here are some suggestions, if you write your own case study.
Cases That Work
Scope
The primary advantage of a case study is that it provides a context for coursework. One of the primary differences between cases that work and cases that don’t is the scope of the case.
If you’re writing a case study, you should do so with the goal of producing a case study that can encompass as much of the coursework as possible. I say as much as possible and not all because I’m a firm believer in introducing new concepts in terms students are familiar with. In some situations, this could very well be some context other than the case study. However, after you introduce the concept, you should migrate toward the case study as quickly as circumstances allow.
As an example, when I teach descriptive stats, I introduce them in terms of grades — because if students are familiar with anything, it is grades. But once I’m satisfied that the students have “got it,” I will then move from talking about standard deviations (or whatever) in terms of grades to standard deviations in terms of sales or revenues (whatever the application may be in the case study), and from there to the case specifically.
But back to the point. The case should encompass as much of the coursework as possible — otherwise, why write the case study in the first place? This should include materials you use in class, at least as much as possible.
One last point about case scope: You can effectively use a series of “mini-case studies” in class, provided that they all fit within some overall framework. Let me explain. Ages ago while teaching a business communication class, I broke students up into cohorts they maintained all semester. Each cohort was a consulting firm, and came up with a name, a mission statement, business cards, and so forth. Each cohort was “hired” four times that semester, and each “job” was a different mini-case that specified a more narrowly-focused, specific problem set than a regular case study would. It worked, though if I had had the time to do so, I would have written a single case study.
Focus
Perhaps the trickiest part of writing an effective case study is writing one with the appropriate focus. By focus, I mean the problems presented by the case study — unfortunately, not necessarily the same as the problems students will be expected to solve. So before you write a case study for your course, you need to sit down, think carefully about everything you will cover and everything your students are expected to learn, and write a case that incorporates them all.
The case study should not be so general in focus that students read it then wonder what they’re supposed to do, nor should it be so specific that you and the students cannot apply something from your course in the case. So if you teach descriptive stats, your case should, in some way, present problems that will require descriptive stats to solve. If you teach inferential stats such as Chi-square, t-tests, ANOVA, or linear regression, your case should present problems that require these tools to solve.
This is not the really difficult part, at least not in my experience. The really difficult part is keeping the focus tight, so that the case does not present any problems that do not address the course content. The problem is that students get easily sidetracked when doing case work, and if they perceive a problem, they will address it — and if your focus is too broad, they will be wasting their time, and yours.
Realism
It shouldn’t need to be said that the case study should be realistic, but I have seen more bizarre case studies than I would have thought possible. Case studies give a “real world” context to coursework. They should therefore reflect real world situations and problems, situations and problems your students will perceive as realistic.
I find that the Wall Street Journal is a rich source of ideas for case studies. It’s not a copyright infringement to get an idea from the real world, then tweak and add to (or subtract from) it until you have a case that works for your class; after all, if getting ideas from the media were illegal, Dick Wolf would have been out of business years ago.
Of course, I’m operating from my home base, which is business. If you want a case study for a history class, why not get your idea from history? Wherever you get your ideas, if you pull them from the real world, you’re more likely to end up with a realistic case study. If you need to do some research, do it.
One final thing: Once you have begun the semester, do not change the case. If you discover a bug during the semester, deal with it as best you can, but don’t change the case until the semester is over.
Presentation
You’ll need to write up the case study. Keep it as short as possible, and keep your English as concise as possible. The case study is not a vocabulary quiz or a reading exercise; it’s the context for your course. If you can’t write it up in one page at the most, it’s too long and you need to change it.
If you deliver course materials digitally, put a high priority on the case study. Students should not have to click through tabs or navigation menus to get to it; it should be right there, accessible from the front page of your course site. You can make it available from multiple pages on the site, but you then run the risk of having multiple versions if you correct or change it, say, right before the semester begins.
HTML only, please. Do not post a Word file, or worse, a PDF file, because both make the case less accessible.
If you deliver hardcopy materials, put the case at the very beginning of your packet, right after the syllabus. You might consider having it run off on different colored paper, just so set it apart so students can easily find it. However you deliver your course materials, you should explain the case and its purpose in detail in the class syllabus.
You should also explain the case study and the role of the case study in the class to your students the first day, and you should refer to the case study in class as often as possible. Do not assign a case study for “out of class only” work. If you do, chances are at least half your students will have no idea what the case study is, or why it exists.
Writing a case study is no picnic. You have to put a lot of time and thought into it. But using a tight case in your course can give the course a coherence it might otherwise lack, or I should say bring the coherence to the surface. And students like case studies and the sense of realism they lend to the course.