> They were talking about how to gauge student progress when the students had taken a 100-question test, and a 50-question test. None of the three teachers real

> They were talking about how to gauge student progress when the students had taken a 100-question test, and a 50-question test. None of the three teachers realized that you had to scale the test scores before making a comparison between the two tests. That's not necessarily true at all. It's a function of the difficultly of each question. Possibly it could be generalized to the difficulty of each test -- where perhaps the questions on the 50-question test were twice as difficult as the questions on the 100-question test. The best way to resolve your concern is to never assign a percentage per test -- just assigned points to each question based on difficulty. Then accumulate points and take an average only at the very end. But, short of that, taking a weighted average of the two tests is not necessarily any more correct than just taking the average of the percentages.