In recent years, there have been a lot of changes in the whole notion of grades. In the old days, it seemed fairly simple; students did the work, the teacher wrote a few comments and gave the student a letter grade. Today the equation has changed on both sides. Some students (and even some parents) have shifted from a mentality that you earn a grade to the idea that good grades are an entitlement. At the same time, and partially in response to this trend, faculty are thinking about grades in a much more integrated way. In a new edition of their book on the subject, Walvoord and Johnson say that the most important principle is: “Grading should be integrated with everything else that happens in the classroom. The grade is not an isolated artifact slapped on at the end; it is part of a coherent system that includes shaping goals and assignments, communicating with students, helping them learn what they need, responding to them, and evaluating the quality of their work. Integration means teach what you are grading, and grade what you are teaching.” They also suggest that grades have multiple functions, including evaluation, motivation, communication with the student, and aiding student reflection.
Barbara Gross Davis gives some helpful suggestions, including:
Familiarize yourself with departmental standards.
Weight and space various components proportionally to educational goals. If the final exam counts for a significant amount of the grade, students will postpone working on the course material until the end of the semester.
Use absolute standards clarified in advance (rather than grading on a curve).
Coordinate grading with the achievement of course objectives, so that if you say something is an important objective, the grading pattern ought to reflect that.
Anticipate problems: how will you handle missing materials; will your grade reflect improvement over the semester; will you reward effort; how will you deal with complaints (TΦ101 has a good policy for this)?
Special considerations for grading test papers :
Read exams without knowledge of their author (fold the cover of the booklet back so you can't see the author, or use student ID numbers instead of names)
Watch out for the "halo effect." Daniel Kahneman has shown that even if exam booklets are graded blind, graders are typically biased by the first answer they read. So if a student does poorly on the first essay, the later essays are graded lower, and vice versa. The best thing is to grade each question (going from booklet to booklet) rather than grading each booklet.
Go back to the first papers you graded and review them again, after your standards have solidified
If several people are grading, have each person grade all answers to a single question (so grades for that question are consistent).
A few other suggestions and tips:
Some instructors try to give some version of a compliment sandwich, so that the basic format of the feedback has three parts: 1) something the student did well; 2) something the student needs to improve; 3) what the student should keep in mind for the future.
Grading essays is a very tricky business. TΦ101 has a detailed discussion of how to give feedback on papers in our discussion of paper assignments.
Make student self assessment a part of grading, getting students to reflect on their own work as part of the process.
Use an electronic gradebook (on BlackBoard or other course management program) so students can see their grades as they develop. If you take attendance (you should) also record absences, so students see that you know when they are not in class.
Develop rubrics to help students understand what to do and why they received the grade that they got.
Do not put grades on papers where you have extensive comments. Many students will read the grade and ignore the comments. Give the papers out with the comments first, then after the students have read the comments, release the grades on the electronic grade book.
When grading short answer tests, make a mark at the end of the answer. Sometimes students will add additional material and resubmit their test asking for a higher grade. Some teachers make copies of exams before they are returned.
Don’t spend a lot of time correcting grammar in papers; students don’t always profit from those red marks. Some teachers correct the grammar only on the first page or two, and then add a general comment that the student needs work on mechanics.
In his article on "A Triage Theory of Grading," William Rapaport suggests that we and our students spend a lot of energy on the small differences in grading (for example, between an A and an A-), but that, in fact, there is little objective basis for these small distinctions. He proposes that we only give four grades: Assignment done and adequate (3); assignment done, but not clearly adequate (2); assignment done, but inadequate (1); assignment not done (0). These grades can be easily determined and easily explained to students. If we want to cut things finer, we can assign each of these grades to different aspects of an assignment and total (and weight them). TΦ101 has not tried this, but it certainly has an appeal.
This helpful grade-calculator, designed by clever historian Daniel Immerwahr, can be used by both students (to figure out how they would need to do on the next test to improve their grade) and by faculty members to calculate either individual grades or grades for a whole class. It also allows faculty members to create rosters for large and small classes, and has some downloadable excel spreadsheets. A commerical and more extensive grading program from GradeKeeper. This program also does attendance and seating; an individual license is $20.
A Philosophy of Grading:
Although most teacher produce grades all of the time, some fairly basic questions about grades remain unanswered. What, for example, is the purpose of grades? Are they intended as a way to motivate student to learn (by rewarding good learning and punishing failure to learn), or are they primarily directed at providing information for administrators, employers, and graduate students? Is grade inflation a problem? Daryl Close has the most comprensive discussion of grading theory that we have seen. In his article, "Fair Grades," Close rejects the idea that grades are a way to motivate students to learn and argues that the only valid purpose of grading is providing information concerning "mastery of course content." Close lays out some general principles (grades should be fair and consistent and should reflect student competence in the material), and then discusses their implications for a number of practical grading questions. He argues, for example, against using grades to motivate students, or to reward or punish behaviors such as missing class or cheating. He also concludes that it is improper to discount students' earlier grades if they do better later in the course, unless this is built into the course from the beginning and applies to all students (including those who do better in the beginning of the course and worse at the end). Teaching Philosophy 32:4 (2009).
TΦ101 has developed a "Final Examination Grading" that raises some of the most vexing problems about grading (without giving any answers).
Close, Daryl, Fair Grades, Teaching Philosophy 32:4 (2009), 362-398. Walvoord, Barbara, and Virginia Johnson Anderson, Effective Grading: A Tool for Assessment and Learning, San Francisco: Jossey Bass, 2009. TΦ101 has not actually seen this book yet, only read a review of it.
Davis, Barbara Gross, Tools for Teaching. San Francisco: Jossey-Bass, 2009
Kahneman, Daniel, Thinking Fast and Slow, Farrar, Straus and Giroux, 2011, p. 83.
Author: John Immerwahr
Update: July 20, 2012
Thought piece on Grading by John Immerwahr,Villanova University
This work has now been published in the National Teaching and Learning Forum, 19:2, 2010, 4-5. The first several questions are available below, see NTFL for the full article.
Mark Twain supposedly remarked that everyone talks about the weather but no one does anything about it. When it comes to the practice of assigning grades in college courses, the situation is reversed. Everyone does it, but there is very little talk about what the purpose of grades is or how they should be assigned. To highlight some of the more perplexing questions, I offer this exercise.
Final Examination on Grading
Instructions: Read all questions. Cross off any for which you have an immediate answer that you think the majority of your colleagues would also share. If you do cross off a question, add a question of your own to which the answer is either not obvious to you or controversial among your colleagues. Answer your new question and all of the others.
1. Which of these is the most important purpose for assigning grades? Are these purposes consistent with one another?
· Give students themselves feedback on how well they are doing to help them understand their strengths and weakness and to help them decide things such as what fields to pursue.
· Document student knowledge and abilities for use by others, including administrators, employers, and graduate or professional school admissions departments.
· Incentivize students to learn more by punishing them for low achievement and rewarding high achievement.
2. When grading your students, which of these three standards do you use?
· How close their work conforms to your own standards of excellence.
· How well they are doing compared to their peers.
· How well they are doing compared to what they could potentially do, if they put in sufficient time and energy.*
3. Student X writes an excellent paper and gets an A. Student Y writes a poor paper but takes advantage of the offer made to all students in the class that they may rewrite their paper as many times as they wish. After several rewrites (responding to the teachers comments each time), student Y finally produces a paper that is as good as student X’s original paper. Is it appropriate for them both to receive the same grade? Does that send a misleading message to external audiences?
*Robert M. Thorndike, George K Cunningham, Robert L. Thorndike, and Elizabeth P. Hagen distinguish between measuring performance in relationship to perfection, in relation to the performance of others, and in relation to the student’s potential. See Measurement and Evaluation in Psychology and Education. 5th ed., New York: Wiley (1991), 168-169.