1. According to research, it is too difficult for highly trained examiners to measure English proficiency spanning more than 9-10 levels. (Covering a scale of no ability to highly proficient.)
2. Even measuring 10 levels professional examiners can be wrong 27% of the time.
3. On such a scale, for classes that meet once or twice a week for 45-minutes, it may take a full year of training or more to improve one level
4. Factors weigh on the whole process which can cause inaccuracies. During some research on IELTS training it was found that after a 3-month intensive training candidates could improve half a level. However, some candidates actually scored at a lower level at the end of the training than they did at the beginning. Reasons for this were the state-of-mind of the candidate on the test day, familiarity and lack of familiarity with the subject matter of the tests, faults in the testing system (IELTS), etc. So it would be possible to test your student at the beginning of the course and then the student could do worse at the test at the end of the course. How would you give a grade in this case?
Alternatively, test the students on the course material at the beginning of the course. If the students are properly placed in the right level classes they should score very low on such a test but after training on the material should score very high at the end of the course.
In that it is too late at this point, the next best thing is to review the material we taught the students and find some distinct and important points that we can test on. However, if the classes consisted of "well, what does everybody want to talk about today?" it may be impossible to test the students.
Finally, the idea of scoring by student attitude or participation in the class or other student behavior focused method, to my mind, but perhaps many would disagree with me, is fraught with the most disadvantages. Such a criteria makes the test exactly one of classroom behavior, not learning or language acquisition. As the focus of most of us teachers is on the student centered classroom we are constantly questioning ourselves if we are meeting the needs and capturing the interests of our students. Some days we do better than others. Some students (especially the very bright and the very dull) are more easily bored than others. We are tempted to punish our troublemakers and reward our so-called 'good' students by the score we would give them. But again, we would be scoring the student on their behavior which is perhaps not the best way to reflect their learning in our class.
The irony is that for many schools it really doesn't matter what score you give the students because the school often does not treat the oral English class taught by a foreigner as a 'real' class. These grades often don't show up as part of their year-end scores.
Nonetheless, I think we should all be careful in how we go about making these kinds of decisions. There are certain 'automatic' impulses that we should be aware of and question. How often do we do things because that's the way everyone else does them? How often do we do things because that's the way it was done when we were students?
These things I have said in the above are with the realization that they may not exactly apply to the certain aspects of the current discussion. Without complete understanding of the situation I know I may be misunderstanding some things. I am just trying to explore various factors and considerations for the purpose of reflection.
We are the teachers and in a position of power with the students and the school. It is a great opportunity to for us to explore and discover the best ways to do things for our students and ourselves.