When developing IIBA’s competency model, we took a close look at several models of skill acquisition and eventually settled on using the Dreyfus Model. I was amused to learn that it has been most heavily applied (and validated) in nursing, especially as my wife is currently working on her Ph.D in that field! Sadly, her research doesn’t have to do with education and so I can’t make her do my job.

Anyway, last night I was thinking about the relative importance of various learning and improvement methods to people at the various levels, and how much those things will affect how effective a person is. I think it looks something like this:

That is to say, at the Novice level you are almost entirely dependent on some kind of training or education to help you figure out what to do (I’m including mentoring in that). As you progress through the Advanced Beginner and Competent stages, your own experience becomes more important until it becomes the primary predictor of effectiveness.

Finally, though, experimentation with new ideas starts to become a factor. Most of the people I know who are really good at whatever they do try new approaches just because. Self-education seems to be more important than training at those levels of performance, although formal training doesn’t stop being useful. I’m not sure if that’s a limitation of training itself or a reflection of the reality that there are rarely enough practitioners seeking training at the expert level to create the market demand (or for that matter enough trainers).

This model also has interesting implications for certification programs. Many of those programs aren’t very clear about what level of competency they actually are certifying. For example, I see a lot of discussion/dispute in the Agile community over the Scrum Alliance certification program, which is based on a course lasting a few days. Is that a legitimate approach? It can be, if you are aiming at the Novice or Advanced Beginner level. However, if people think that they’re getting Competent or Proficient practitioners out of a course they’re going to be disappointed. At the other end of the scale, I’m not clear on whether it’s practical to try and certify Expert practitioners.

I’m going to be spending more time investigating the practical implications of the Dreyfus Model. The reason I’m thinking about that, of course, is that it has real implications for professional development at different stages of a person’s career. The mechanisms that are good at one level won’t work at others.