In Academic Year 2013-14 the grades given in the Arts & Humanities Division at Dartmouth were on average 0.29 points higher than in the Science Division on a 4-point grading scale, and in 1977 the difference was 0.25 (you can see the data here). The Social Sciences Division has always been sandwiched between the two, with grades about 0.05 points higher than the Sciences on average over the years. Finally, grades in our Interdisciplinary Programs show more variability over time, but are more similar to the Arts & Humanities than the Sciences and Social Sciences (again, see the data here).
Discussions about grade inflation typically devolve into a shouting match between the Sciences and Humanities, which is completely wrong. Grades in all Divisions and in all departments within them over the past 50 years have risen at nearly identical rates (again, see the data here). Let me say that again: grades in every Science department at Dartmouth are increasing at the same rate as grades in every Social Sciences and Arts & Humanities department and in every Interdisciplinary Program over the past 50 years.
Our committee believes that these differences in grade distributions among academic disciplines are natural. Our effort is decidedly not to equalize grades across Departments or Divisions at all. I think very good educational reasons exist why grades in the Arts & Humanities and Interdisciplinary Programs are higher than in the Sciences and many areas of the Social Sciences. Let me outline some of them here, but in so doing challenge everyone in every discipline to return to more rigorous grading standards. If we do, our students will be better and accomplish more.
The main differences between the sciences and the humanities that people point to with respect to grading are: we give exams and they write papers, and we have facts and they have opinions. I don’t think these differences are as big between the sciences and humanities as people imagine. However, I do think the reasons for our differing grade distributions are buried in these seeming differences, but it’s a much more nuanced difference.
Grading in the sciences is somewhat simpler because we tend to grade based on exams, and a lot of the questions on our exams involve knowing specific facts, hypotheses and theories. The formula for an ellipse is pretty much set. You can’t quibble about the definition of the Ideal Gas Law. Electron transport in the mitochondria of eukaryotes operates under a strict set of biochemical principles. In the Humanities, the subject matter tends to require less overt expression of such facts. Students aren’t typically graded on their abilities to recite Shakespeare’s 18th sonnet, express the major features of Hammurabi’s Code and apply it to present-day situations, or provide a detailed timeline of Alexander’s conquest of Persia.
Thus, less grading in the humanities involves “you’ve got your facts wrong”. (If I’m wrong about this and what follows, I hope you will correct my errors in the Comments section below!) However, most science exams involve some questions of analysis and application where multiple answers may be completely correct, and so grading in the sciences also often involves some degree of evaluating speculation and opinion.
I think the main difference is due to the fact that one can bump up against the limits of knowledge and practice in a discipline much faster in the humanities than in the sciences and many social sciences. Most scientific disciplines are bottomless pits of ever more complex theories and facts, whereas many humanities disciplines have a fixed amount of material with which they work. We have only so many writings of Plato and Aristotle to study; a fixed number of events happened in the Civil War; human language has a finite set of rules and patterns.
As a result, many humanities courses bump up against the limits of “I’m teaching you absolutely everything there is to know”, and if students can develop excellent mastery in that, then so be it. Increasing the rigor in these courses would involve having students applying the same kind of scholastic analyses and techniques that Humanities professors apply in their own research. In fact, this is what most Humanities faculty are doing in their classrooms. Moreover, student works can approach and sometimes surpass what is published in scholarly journals or what is presented in major art galleries and performance halls. Faculty should actually target some assignments towards such ends for students as producing papers that could be published in scholarly journals or presented to the public.
Skills of writing, reasoning and analysis in Humanities courses also tend to be more transferable across disciplines within the Humanities than in the sciences.
Grading in the sciences and humanities are also different beasts. Grading in the sciences is mainly about structuring one’s exam. Having talked to many other scientists over the years about this, I think my general process of developing an exam is very similar to most science professors.
I write three types of questions for exams. 50% of the questions are the soft-balls where I’m just trying make sure the student was paying attention. If a student can’t answer these, then they’re going to get a D or E on my exam. If they can, they’ll at least probably get a C.
The next 30% of questions are more challenging. These require the student to think harder, and to apply their knowledge to familiar situations. I try to set the difficulty of these questions to a level that the good students in the class will be able to answer with hard work and diligence, but many will not be able to. These are the questions for me that separate the B students from the C students.
Finally, I make the last 20% of the questions very challenging. These are questions that I think only the very best students will be able to successfully and completely address. These are the questions that discriminate the A students.
I know of a few universities in the country where faculty basically dispense with my first two categories and jump straight to the hardest level for almost all the exams and grade accordingly. I would argue that students at these places are getting the best educations in the country. The rest of us can’t jump to there, but we can start moving in that direction. For example, I would love to dispense with the 50% softball questions, but most students at Dartmouth would freak out if 50 of 100 points was a high score on an exam (I had one professor in college that gave exams like this and I studied like crazy for that class). Given their responses to exams with an average score of 70 vs. 80, I’ve seen some freaking out already.
We in the sciences need to step up our standards by increasing the percentages of these more difficult questions on exams, and holding ourselves to more rigorous standards of what scores on exams represent as grades. From analyzing grades I gave 15 years ago to grades I gave this year, I can see the cause of my own grade inflation. What is now the lower end of the B range for me used to be B-, and my current B- used to be C+. In intend to fix that.
I would also challenge my Humanities colleagues to think seriously how they challenge students in their grading practices. In a large faculty forum a few years ago, I heard a senior Humanities professor say that when he first arrived at Dartmouth, he could read a student paper and clearly decide what grade to assign. He said he couldn’t do this anymore. Students now come and challenge him on the grades they get on papers and assignments, and he has a hard time explaining to them why he gave them the grades he assigned.
In my opinion, this professor was shirking his most important responsibility to his students: holding them accountable for their work, and clearly explaining the shortcomings of their work. These are the most important teaching moments.
Having an argument in your paper is not the basis for giving a paper an A. Having an argument in your paper should be the basis for not getting an E (the failing grade at Dartmouth) on your paper. Grades of A-D should be based on the quality, structure and depth of the argument presented. You get more work and struggle from students for better arguments if you reserve A’s for truly exceptional ones. For example, could you imagine with a little more work, this student’s paper being published in a scholarly journal? Is that the appropriate standard for an A paper?
Also, as students revise papers, just don’t mark up the papers and grade on whether they took your corrections or not. Make the students do the heavy lifting of figuring out how to revise, based on your comments of what’s wrong. Discuss with them how to fix it, but don’t fix it for them, and then essentially grade yourself. I teach writing classes, and I know of what I speak. It’s very easy to say, change it like this and then grade them on whether they did it. In that case you’re essentially grading yourself. Allowing students to revise papers does mean that grades in the class might be higher, but part of this is for the students to learn how to revise and edit – not simply do what you say. Grade them on how well they revise and edit as well, because those are also very important skills they’re supposed to be learning from your class.
The problems caused by grade inflation will not be solved by making the grade distributions in the humanities and sciences match. The problems caused by grade inflation will be solved when we all return to holding students to rigorous academic standards in all our courses.
Update added 12 May 2015:
Here’s what we’re talking about! Students from Dr. Tom Luxon‘s Shakespeare I class (English 15) presented their scholarship developed as an assignment in the class at the 2015 Medieval and Renaissance Forum at Keene State College. Here’s the photo of the students from the Dartmouth Now article.