Okay, random post about the stupidest thing ever to get excited about. But what the heck:
This is W2MI. It's a simple matrix for evaluating behaviour or performance in relation to any standard—so you can use it for anything from setting HR standards to testing religious zealots. Examine any criterion and evaluate the subject's behaviour following the matrix. Zero points for something completely wrong, two points for behaviour that could act as a model for others, and one point for anything in between that is neither wrong nor at model level. The final level, "inspiring", is essentially off the chart—you do not reveal it to the subject unless he or she qualifies for it.
The psychology is simple but brilliant, both for the evaluator and for the evaluated subject. Naturally, nobody wants to be "wrong". This is nice and simple for the evaluator: you don't have to explain why something is wrong. It just is based on the criteria. You can argue about whether or not something should be wrong, but the key point is that you will not argue about the word "wrong". It is worth zero, meaning there is no encouragement whatsoever to engage in this behaviour.
Being graded as "well-intentioned" is encouraging, but innately implies something lacking. Similarly, the evaluator can easily check: if he or she is making excuses for the subject, or is inclined to offer a pass for difficult circumstances, these conditions still fall under this category because the behaviour is not a model to be followed by others (philosophy buffs will note the shout-out to Kant here—this is precisely the idea!).
So wrong behaviour is worth nothing, and well-intentioned behaviour is worth 1 point. Then model behaviour is worth double well-intentioned behaviour. Finally, inspiring behaviour is worth triple well-intentioned behaviour, knocking everything else out of the water.
Inspiring behaviour is that which inspires others to improve themselves, and even inspires the evaluator. Model behaviour itself does not inspire—it's just a good example to follow. Inspiring behaviour encourages others to not only follow a model, but expand on it, explore, and test the limits not only of themselves, but of the model they have aspired to.
4 levels. Evaluate a selection of questions, and boom, you have a rating of your subject based on your criteria. Ideally, you should set your "maximum" score, if you have such a thing, as roughly 10% beyond a perfect "model" score—to get a maximum score, you must be a model in nearly all criteria and inspiring in several. So, for example, if you have 5 questions, the perfect model score is 5 x 2 = 10, so your "maximum" score should be 11 or 12. Again, you don't tell your subject about the "inspiring" level—at least not until they max out on "model" scores and are wondering why they are not at perfect yet. An ideal test would go up to an unneven number, with upwards of 100 points to conceal the impossibility of scoring perfect without inspiring behaviour.
When I first developed this, it was Unexcusable/Excusable/Model/Inspiring. The modified nomenclature was developed with Cedric Julius, and he introduced me to the impossible-to-achieve maximum score concept.
Boring? Maybe. But it will be very interesting to apply.
Update: Some interesting back-and-forth on this with Amorous Route author Scott Donald on FB. Posted as a series of comments below, originally from Dec. 9-15, 2011.