How to write a teaching case
I am currently – with colleagues Mikael Lönnborg and Gerhard Schjelderup – editing what we hope to be a book of Scandinavian teaching cases. In a meeting in Stockholm recently, I was asked to explain what it takes to write a teaching case. I gave my opinion, we had a very interesting discussion. Here is my (very rough and off the cuff) opinion about what it takes (in reality, how a teaching case differs from a research case).
Why are you writing this case?
Cases are written for a teaching purpose – and to write a teaching case, you need to have a teaching objective in mind. It is not enough to have an interesting company. Even the best company story needs to ...view middle of the document...
Make sure you tell it logically and cooly, not leaving anything out, but also conveying the complexity of the situation.
0.5 page: Conclusion, typically with the protagonist wondering what to do, often with some sort of event (board meeting, etc.) where he or she has to present a solution to the problem.
Most cases are just that – one case. You can have a B case and even a C case, but keep them short, since they have to be handed out and read in class. The B case should explain what the company did and perhaps introduce a new problem, the C case, if necessary, should bring some sort of closure, explaining what eventually happened. In my experience, it is very hard to get discussion after a C case – the students become exhausted. As a novice case writer, especially if you are writing about a company with a long history, it can be tempting to create a long string of small cases, but in practice this seldom works well – for one thing, it forces the discussion into a very predictable path.
A good case should be a description of an interesting situation, frequently a decision point – and nothing else. This means that there should be no theory and no discussion of the case in the case itself. Save that for the teaching note, or write a separate academic article about it. Not only does this make the case more realistic, it also means it can be used for more purposes than the one initially envisioned. This can be quite challenging for the traditional academic writer – but ist is actually good practice to only present the facts (though, of course, which facts you choose to present constitute a discussion of sorts).
When teaching students how to analyze a case, I always start by saying that for most business situations, if is useful to begin the analysis with the assumption that people are not stupid and not evil. Consequently, when you write a case, make sure it has no heroes and no villains. If a case has a clear-cut hero or villain, it is a sign that you have not done enough research. Write things so that the students can see the issue from many perspectives.
A really well written case has dramatic structure – there is a beginning, a middle that builds up the story, and a really compelling issue at the end. The best cases are almost like a detective story, where you have to dig deep into the analysis to find surprising and sometimes counter-intuitive conclusions. One example of a “detective story” case is Fabritek 1992*, a very old (first published 1969, rewritten by Jan Hammond) case about a quality control issue in a small mechanical workshop. (Hat tip to Robert D. Austin, eminent case teacher, for making me aware of this case and showing me how to teach it.) The case is excellent because it starts with the company (strategic level), proceeds to describe a new situation and a new process (organizational or business logic level) and then introduces the problem (operational level.) Analyzing the operational...