Decision Tree for a 10-Year Old
This may well be the most unusual article I’ve ever come across in any peer-reviewed journal (it was published in an open access journal, so you can get the full text pdf here):
Mullin, Barbara & Roger, commented by Jack Dowie and Rex V. Brown.
Mhairi’s Dilemma: A study of decision analysis at work.
Judgment and Decision Making, 3, (8) (2008), 679–689.
It’s a case description of how a dad (Roger Mullin, a teacher of decision support systems) used a decision tree to help his 10-year old daughter, Mhairi, make a very difficult emotional decision: she had to decide whether to attend a dear friend’s funeral or not.
Decision trees are, in my own experience, one of the hardest tools to teach, as powerful as they are. They involve calculations with probabilities, which people are notoriously bad at – even people much older and more educated than the child in this case. Accordingly, they are not used that often, even by professionals. As Mullin states himself: “For over 20 years of teaching decision making, I have been struck by how resistant many professionals are to analytical approaches. A common complaint, whether from medical practitioners, social workers, police officers, business executives or other well educated groups, is that it is too difficult for their clients or even colleagues to understand and far too difficult to involve them in the process.”
Roger Mullin has my fullest respect and admiration for using this tool, not only with his child, but for a decision so emotional and hard to quantify. In this report (which includes his wife’s perspective and reflections), Roger Mullin makes a very strong case that a formal analysis is not at all incompatible with being sensitive, caring and effective.
Decision Trees Made Easy for the Rest of Us – Video
I have searched widely for good explanations of decision trees, and found many dry and complicated papers along the way. However, here is a video that is refreshingly simple (though not always pretty):
It provides straightforward instructions how to draw a decision tree, and how to calculate expected values with it.
If you give it a try for your own decisions, I’d love to hear about your experience.
by Ursina Teuscher (PhD), at Teuscher Decision Coaching, Portland OR