I’m kind of obsessing about complexity theory right now (Dave Snowden’s Cynefin Model mostly), and looking at simple, complicated and complex systems. I had a lot of conversations about this last weekend, and have been thinking about it a lot.
A couple of upfront disclaimers — first, I’m just learning about this, so I don’t pretend to really understand this stuff. It’s my interpretation, but it wouldn’t surprise me at all to know I’m getting the details wrong. Second, I’m not digging into Chaotic (for now at least). Third, there’s a much longer post on this brewing, and I have more questions than answers right now.
So — let’s apply this to the question of school testing, for example:
Simple things (with explicit rule sets) are probably fine to assess via multiple choice tests. MCQs for multiplication tables? Sure! No problem!
But complicated things (e.g. the subtleties of designing a scientific experiment) and complex things (e.g. problem-solving skills) do not have explicit rule sets, and are therefore NOT appropriate topics for a really reductionistic assessment methods.
School testing models are trying to squeeze all the ambiguity out of the system by trying to control every variable. You can do that with simple and possibly with complicated systems (though it’s an insane amount of work — witness the amount of procedural documentation in the air safety industry, or the nuclear power industry in their attempt to eliminate all ambiguity. It’s usually only justifiable when people’s lives are at stake).
But you can’t (by definition) eliminate all the ambiguity in complex systems. E.g. you can teach principals for problem-solving, or a process, but how it gets implemented depends on the context, which you can’t control. That’s where teachers, with their personal judgment and ability to adapt, become really important. It’s one of the limitations of computer-based instruction.
People don’t like not having control. School testing is trying to exert control by pretending that everything can be put in the simple box, so it can be measures using simple, objective measures. But it just doesn’t work.
I think there’s some real value in having a good way to assess whether or not you are dealing with a simple, complicated or complex situation, and adjusting not only your assessment, but also your learning design for that. Working on this, but if you know of anything really useful, please let me know.
A couple of good resources:
Addition: This article is a pretty perfect case study of this Poet: I can’t answer questions on Texas standardized tests about my own poems