An excellent discussion and am enjoying reading the comments. As an undergraduate, I confronted bias, extreme in two cases, in grading during my first 2 years. Only through perseverance and mentoring from two professors was I able to overcome it. Thus, the governing principle in my own teaching career was to let students succeed or fail on their own merits.
Like @einpokum, I abhorred the bell curve believing it to be both antiquated and unfair. One gorilla in the room is the fact that some students on every campus, primarily through various organizations, especially Greek letter ones, have access to exams and papers than non-connected students do. As a graduate student I lost a number of papers graded by professors who left them outside departmental offices. And, as a professor, a couple of papers submitted to me I determined had been plagiarized. Protecting the fairness of student evaluation is an institutional responsibility.
In my own experience, after my 1st year of teaching at the college level, I instituted what I called developmental grading, believing that teachers are responsible for providing a framework in which all students have an opportunity for success. This is a concept I learned when in nursery school from two very intelligent teachers, and it's remained in my consciousness every since. Their application was solely environmental. The approach I used was also environmental, but focused on student evaluation, and was three-phased:
Students took an exam, which was graded by me and later by others as the numbers of students in my general courses increased, from less than 20 when I began, to over 100.
Once grades were known, students could apply to take the exam again for half of the points for each question that they "missed" in the initial exam. These would be added to their earlier grade.
If still not satisfied, they could take an oral exam before the entire class who would then evaluate their performance.
Although there were several who opted for #2, none applied for option 3. I was hoping for at least one to determine how the students would respond.
Only after the 2nd phase was completed did the class go through the exam together to share their thoughts about each question and the expected answers. Later, this was modified for students to go through the exam with student assistants or other professors depending on the course.
A second principle in my evaluation, which students ultimately learned, was that the three previous exams had less impact on their final grade than how they did on a comprehensive final; and, that these exams and the follow-up discussion were designed to help them perform better on the Final. In fact, a student could do poorly on all other exams (barring poor attendance or otherwise failing minimum parameters), get an A on the final and achieve an A for a grade. For some students, no doubt, it appeared as a gift, but they earned a better grade. Why? I believed that finals were designed not to achieve a percentage of a grade in relation to all other criteria, but to test what they comprehensively learned overall about the subject of the course. Grading final exams was one of my base pleasures, as well as the gourmet coffee I imbibed to get me through them. :)
I might add that exams were not the only factors entering into a final assessment, but book reviews, reports about speakers or activities on campus germane to the class or quizzes which they triggered by not asking questions when prompted by my question at the end of each class--are there any questions? Like many of you, I suspect, teaching involves learning complex concepts and reducing them to an 8th grade level so that most first to second year students can comprehend what it is we are talking about.
