Here's a personal confession that may (or may not) offer something you can adapt to your own purposes. A recent newsletter that came by my desk had a brief squib on leadership. Counter-intuitive though it seems, said the article, you can get better improvement from people by focusing on their strengths than on their weaknesses.

Obviously this can be applied to our advising and mentoring of students. But I have also been interested in exploring ways to apply it to grading. I tried the following experiment in my "problem" course, intermediate French. Intermediate French is a very difficult course that asks students to make a giant leap in proficiency and that requires so much of students that I must constantly battle student fatigue and discouragement. I told students that they were being asked to progress in a variety of language skills, all of which are factors in being language competent. But since people are not equally skilled in every language task and yet can still be perceived as language competent, so I felt that I should adjust my grading to recognize and validate their individuality. So, the 2 main elements of the course are worth 20% and 30% of the student's grade. Which element is worth 20% and which 30% depends on the student--I assign the 30% to whichever skill is stronger for each particular student. I have done the same thing with the 2 secondary elements of the course, worth 10% and 15% of the grade.

The result of shifting my grading system slightly to recognize student individuality and individual capacity was very satisfying. Not only did the students start working harder, but the energy in the class dramatically improved. Students were no longer focusing on negatives, but rather exploring the ways to make use of what they can do right. What I have liked best about this adaptation to my grading system is that I now have students who are less focused on not making mistakes and more focused on positive development. They look more like risk-taking learners which, in language, can be a very good thing indeed.

I hope this gives some of you some ideas about things you can try in your "problem" courses.