I agree that this is necessary. It’s also done in a haphazard way that involves setting points or weights that seem ‘good enough’. My argument (which I will develop later on) is that we can decide what our priorities are in determining achievement levels, and the weights can develop from those priorities.

This is a totally a valid question, but I’m excluding it for a practical reason: we don’t need the number of rubric categories to match our levels. We make categorizations of students all the time based on our experience, and the mismatch in dimensions here forces us to make choices on what matters. I’ll be clarifying my reasons further in my next post.