Rick L. Danheiser
In July I began my appointment as Chair of the Faculty. Joining me as faculty officers are Professor Duane Boning of EECS as the new Associate Chair, and Professor David Singer of Political Science as the new Secretary of the Faculty. We are honored to have been elected to serve as your faculty officers for the next two years.
Regular readers of the Faculty Newsletter are aware that it is the privilege of the Chair of the Faculty to prepare a column for each issue of the Newsletter. As I write this, my fellow officers and I have only been in office for a few weeks and we are still in the process of identifying key issues and setting our priorities for the next two years.
Some of the topics we have been discussing and hope to consider in the coming year include the Institute's international engagements, faculty benefits (including availability and support for childcare and for faculty housing), promotion and tenure, graduate student tuition, the new Schwarzman College of Computing and its impact on the Institute, and last but certainly not least, the General Institute Requirements and plans for a task force on the undergraduate academic curriculum.
In my future columns in the Newsletter I intend to address these more substantive matters, but in this first column I would like to focus on a more mundane, albeit important and timely topic here at the beginning of the semester: term regulations.
Term Regulations: Read and Obey!
Every semester the faculty officers receive a number of complaints from students – either directly or via the Undergraduate Association (UA) – about classes violating term regulations. The UA maintains a website through which students can report violations with the option to remain anonymous, and the UA Committee on Education sends an email to all undergraduates several times over the course of each semester summarizing the term regulations and reminding them of the process to report violations.
One of the responsibilities of the faculty officers is to enforce these Rules and Regulations of the Faculty. At the beginning of each semester I send an email to all instructors summarizing the key rules that affect how we teach. These fall into three categories: (a) grading guidelines, (b) restrictions on assignments and exams, and (c) the scheduling of exams, quizzes, and review sessions. Full details on the rules and regulations that affect teaching can be accessed via the Faculty Governance website. In this column I would like to discuss some of the most common violations that have come to the attention of the faculty officers in recent years, with a focus on rules that impact the assignment of grades.
Grading on a Curve
. . . is strictly forbidden! For a few reasons I have devoted most of this column to a discussion of this particular rule. First, the mistaken view that grades are curved is widely held among students. I have frequently encountered eyerolls and skepticism when informing students that grading on a curve is prohibited at MIT. And the fact is that curving (and the related "scaling" of grades) does happen, as we know from complaints received by the faculty officers in the past.
Some faculty practice "curving" simply out of ignorance of the MIT rules, but a complicating factor is that grading on a curve is an accepted and common policy at many other institutions. Some new faculty join us having been students or having taught at places where grading on a curve is general practice and they assume that curving grades is legal here too.
A source of occasional confusion with regard to complying with this rule is that various people have different interpretations of just what constitutes grading on a curve. Aware of this confusion over the definition of curving grades, the faculty officers proposed expanding and sharpening the wording in Rules and Regulations with regard to grading this past spring. After receiving input from relevant committees (FPC, CUP, CGP, and CAP), we introduced the new wording at the Institute Faculty Meeting on March 20 and it was approved by vote of the Faculty on April 17. The new wording in the section on grading (Regulations of the Faculty Section 2.62) states:
"The grade for each student shall be determined independently of other students in the class, and shall be related to the student's mastery of the material based on the grade descriptions below. Grades may not be awarded according to a predetermined distribution of letter grades. For example, grades in a subject may not be allocated according to set proportions of A, B, C, D, etc."
The definitions of grades are then outlined in section 2.62.1:
A Exceptionally good performance, demonstrating a superior understanding of the subject matter, a foundation of extensive knowledge, and a skillful use of concepts and/or materials.
B Good performance, demonstrating capacity to use the appropriate concepts, a good understanding of the subject matter, and an ability to handle the problems and materials encountered in the subject.
C Adequate performance, demonstrating an adequate understanding of the subject matter, an ability to handle relatively simple problems, and adequate preparation for moving on to more advanced work in the field.
D Minimally acceptable performance, demonstrating at least partial familiarity with the subject matter and some capacity to deal with relatively simple problems, but also demonstrating deficiencies serious enough to make it inadvisable to proceed further in the field without additional work.
I hope that the revised definition is clear! In explaining the MIT grading policy to my own classes I have always found it helpful and clarifying to tell the students that what this boils down to is that every student in the class has the potential to receive an A (always the hope of the teaching staff!) although everyone could potentially get a C . . . it is just a matter of how well they have mastered the material.
I also hope that everyone agrees that this is the only policy that is fair and makes sense at MIT and that to grade on a curve would be contrary to Institute values. Why should the evaluation of one student's level of mastery of the material in a class be influenced by the level of mastery demonstrated by other students? Do we not want to encourage students to focus on mastering the subject, not on competing with their classmates? Criterion-based grading avoids pitting students against each other and leads to a heathier atmosphere in which the students can appreciate that the instructors are rooting for all of them to get an A.
From the point of view of instructors, it is certainly much easier to grade on a curve, which is perhaps why it is so popular at other institutions. In some universities curving is regarded as a tool to fight grade inflation. A curve can also be a means of ensuring fairness in cases where classes are divided in various sections with different instructors and different exams (perhaps offered in different semesters). Grading on a curve then ensures that students are not disadvantaged by taking a class with a particular instructor, or in one semester rather than another.
Of course implicit in our MIT policy is the assumption that instructors can distinguish between (for example) A-level and B-level performance. The pithy MIT criteria quoted above are helpful in defining grade borders, and I have always found it pretty clear how to recognize what constitutes A-level work in a course. In the words of Justice Potter Stewart (concerning another matter), "I know it when I see it." Admittedly, however, defining grade borders can be challenging, especially for less experienced instructors, and consultation with colleagues can often be helpful in this connection.
OK – now time for a quiz! In each of the following scenarios, are we dealing with a violation of the MIT policy on grading or not? Answers are provided at the end of the column. Warning: some of these cases are not very clear-cut and are included to stimulate discussion.
1. An instructor defines the grade borders for the first several assignments in a class based on the MIT grading criteria with the result that 25% of the class fall in the "adequate" or lower range (C or below). Just prior to drop date nearly all of these students drop the class. The remaining students continue to perform at a "good" (B) or "exceptional" (A) level. At the end of the semester, the instructor assigns 10-20% of the class a grade of C.
2. An instructor analyzes the grade distributions in previous semesters for a subject and finds that in each case the distribution was 25% A, 50% B, and 25% C/D/F. To ensure fairness relative to prior years, the grade distribution for the new semester is set to correspond to this distribution.
3. Upon examining the distribution of scores after an exam the instructor chooses the letter grade borders so that they fall at breaks between significant numbers of scores in the list.
4. After composing an exam, an instructor decides that a score of 80% represents exceptionally good performance, superior understanding, etc., in other words, A-level work. Upon grading the exams, however, the instructor finds that one question was worded in a confusing and misleading fashion and most students answered it incorrectly. Consequently, the instructor revises the A/B border to 75% to take this into account.
One more recommendation with regard to prudent policy. It is my experience that it is best to provide students during the course of the semester with a letter grade for each exam and assignment. This allows students to have an accurate idea of what letter grade they are headed for and minimizes disputes over grades at the end of the semester.
The Take Home Messages
• Don't grade on a curve! Review and understand the MIT policy on grading and consult with colleagues (and the faculty officers) if you have any questions.
• Clearly communicate your grading policy to your students. We need to disabuse students of the notion that they are being graded on a curve and that a classmate's success can detrimentally affect their own grade in a subject. In my larger classes I have found many students to be obsessed with learning what the class average was on an exam. I inform them of the average, but I always use this as an opportunity to reiterate and emphasize that the class average has no bearing on the letter grade assigned to their numerical score.
• In the case of subjects taught by different faculty in different semesters, communication between instructors is essential to ensure that students are being evaluated on a similar basis.
I have run over the space allotted to me by the editors, but I would like to close with a few words about other aspects of the term regulations. As Associate Chair of the Faculty over the past two years it was my observation that the most common violations of term regulations involved (a) scheduling the due date for an assignment after the last day of classes, and (b) having an assignment due in a class with a final exam after the "Last Test Date". These are violations of Regulations 2.52 and 2.53, respectively, and typically are due to instructors who simply are not familiar with those regulations. Occasionally, however, the violation has arisen due to questions concerning the definition of what constitutes an assignment, and this is a question that will receive attention from Faculty Governance in the future.
The answers to the quiz follow:
1. I consider this to clearly be a case of curving grades and a violation of MIT policy. I provided this example because it highlights a major and common concern of those students who have the impression that some instructors at MIT grade on a curve.
2. This is curving and violates MIT policy on grading although I appreciate the instructor's intent. It is not necessarily the case that grades were curved in the prior semesters, but in any event two wrongs don't make a right.
3. A gray area. If the letter grade ranges are set primarily based on the MIT criteria, and then fine-tuned with reference to where there are breaks in scores, then this might be acceptable.
4. Another gray area, but I would argue that this does not constitute a violation of policy since the grade borders are still being set based on the instructor's judgment of the level of mastery of material.