Can we make our students into superforecasters?

Poster_of_Alexander_Crystal_SeerAs long-standing readers might recall, I read. A lot. But I don’t tend to talk/write about my reading much, either because it’s not immediately relevant or because the rest of my life is so rich with incident. Probably the former, then.

However, I’ve just ploughed through Philip Tetlock & Dan Gardner’s “Superforecasting“, which does offer some interesting perspectives that might be tied to our work in the classroom.

Tetlock will be known to you for the meme that most predictions are no better than chance in their accuracy. As he points out, while that might be true in the aggregate, it’s not true for ever forecaster, and this book (which follows up the Expert Political Judgment that set out the (rigorous) basis for the meme) explores those people who have demonstrated a consistent and measurable ability to outperform both the metaphorical ‘crowd’, but also other mechanisms, such as forecasting markets.

Tetlock’s core point is that such ability is not innate, but learnt and learnable. And that’s what interests me here.

Over the past couple of weeks, I’ve been teaching students about negotiation theory, before I drop them off into the metaphorical deep end of actual negotiation. One of the core skills I’ve been trying to stress to them is the need for preparation, which forms the bedrock of good practice.

However, while it’s easy to say “be prepared”, it’s often hard to know how to prepare, especially if the situation is a complex one. And I can see utility in Tetlock’s approach as a way of improving the chances of having as good an oversight of the preparatory phase as possible.

In practical terms, that means doing things like breaking down intractable problems into tractable sub-problems, looking at the situation from as many different perspectives as possible and engaging in constant checking of your practice and biases.All the kinds of things we might be encouraging students to do as active learners, in fact.

And perhaps that’s the point. Learning is a memetic process, where we draw analogies across from one place to another, hopefully casting light on the way. Tetlock’s not interested in this book about effective actors, although he does mention several examples of how his core ideas get operationalised, but about our ability to forecast events. I’m less interested in that, but can see how I could use it to improve students’ performance elsewhere.

Tetlock talks about the rise of the ‘mission command‘ model in the military, the notion that what matters is the underlying intent, which local commanders can then find the most appropriate way to achieve it. This replacement of the strict hierarchy of command-and-control with something much more flexible and adaptable is not so different from the shift towards active learning. We set out learning objectives and students find their own path towards them.

I wouldn’t want to stretch this too far, but it makes the point that we should be casting our nets as widely as possible when developing our practice, because the analogies can take us to places that might never have considered before.

Leave a Reply