Panos’s recent blog post on “Why I will never pursue cheating again” is a detailed description of how he dealt with copying in his class and why he’s tired of it. It seems that many a teacher immediately identifies, and so do I. Here is my fantasy way of totally solving the problem: stop giving grades.
For me, the worst part of the academic experience is giving grades and everything that surrounds them: writing exams, grading them, dealing with appeals, designing student assignment according to their gradability, making sure that every student task is legally well-defined, dealing with cheaters, etc. I would estimate that more than half of my teaching time goes to grading-related activities and less than half the time goes to real teaching. I would estimate that the situation for students is about the same: more than half of their time is spent on optimizing their grade and less than half of it on optimizing their understanding, the latter smaller half including also time that both improves grades and knowledge.
Beyond being an amazing waste of time, the whole grading thing is extremely harmful to the educational atmosphere: most teacher-student interaction is devoted to grade-related issues which immediately puts them on confrontational mode. The atmosphere in class deteriorates to students being only interested in what’s “going to be on the exam”, forcing the teacher to structure all material in well-defined testable modules, rather than optimize for educational value. As grades become the most central part of the educational experience, even students who used to be interested in learning stuff, are trained to focus on grades only. An “optional” homework exercise is ignored unless it comes with a grade bonus (in which case, it is treated as mandatory, in light of the grade maximization goal). The list goes on and on.
There certainly are some reasons why grades are beneficial, but I think that their total weight is not sufficient to justify the great costs. I would say that the main reason that we grade is since we really are not in the business of providing an education but rather in the business of providing degrees. To award a meaningful degree you need to grade. To deliver a meaningful education you do not. I know that I shouldn’t complain since the degree-awarding business (aka higher education system) is really a great one to be in, but I do wish that someone would design an education-delivery one.
Turing's Invisible Hand 
> I would say that the main reason that we grade is since we really are not in the business of providing an education but rather in the business of providing degrees.
Though taking into account the post you linked to, even the point of a degree is cast into doubt if people just cheat their way to getting it.
Noam, I think one solution is to outsource the grading related part of the schooling and focus on education.
Suppose you are teaching an undergraduate algorithm’s class. You teach and give assignment for educational purpose but you do not decide their grades.
Suppose another professor teaches the same class at another university. This professor decides the grading for your students, and you decides the grading of her students. This can be feasible if all the grade related assignments are submitted in a digital format.
he students could still argue about their grades, but it would be more laborous, and hence they may only do so in fewer cases. Also, any enmity would be towards a remote professor (and her name could kept confidential if helpful).
What makes the situation worst is that the students are costumers at one hand paying money to the university and on the other hand they have strong organizations that they can use to pressure the university. This makes the bureaucracy of grading and pursuing cheating very difficult, punishments are too small, occasional, and very difficult to get. There was a student in our university that took the matter so far that lawyer and university’s governing conceal got involved. We give these students certificates for what they don’t really know. The other interested party is the industry where these degrees have value and universities degrees are a way of ranking possible candidates for them, but there is not any corporation between universities and industry about what does a degree means, there is only an indirect feedback through reputation which is too weak to be an incentive for universities to fight cheating.
Here is a suggestion: there should be no punishment for cheating during the course, but suspected cases should be taken into account to give a grade solely based on the opinion of the teachers about the honesty and other social factors about the student and this should be put in their degree. I think that would be a reasonable incentive not to cheat.
Part of teaching is a mechanism design problem – how to give incentives tor students to learn the material well and to enjoy doing so, so that they will continue their studies. As onerous and unpleasant as grading can be, the incentives of grades do achieve these good learning aims in the right context.
In a context where grades are a serious barrier to entry to some desired club, or where even a passing grade is quite uncertain, the grade incentives can be detrimental in the sense you describe. The grade incentive so dominates that anything to achieve good grades is optimized, whether or not that involves learning.
When the grade stakes are much lower. the incentives align quite well. Grades are seen more by students as diagnostic of their understanding. A personal grade goal is then just enough of an incentive to push them learn the material better (rather than some outside stick) and they seem to appreciate this push.. This incentive is useful because of how humans optimize – think of how much better students learn when they are taking a course for credit rather than just auditing it.
I have taught in both of the above environments and the difference is like night and day. One does have to make sure that grade expectations are clear so that there is sufficient correlation between students’ understanding of their grades and their understanding of the material. Also, even in the low grade stakes context one has to be careful to set things to have sufficient reinforcement that students feel that they have some reasonable possibility of mastering the material (and that this mastery will be reflected in their grades).
An “optional” homework exercise is ignored unless it comes with a grade bonus (in which case, it is treated as mandatory, in light of the grade maximization goal).
This can be made to work. I give such optional material on virtually every assignment and do give a grade bonus but do it so that it is not treated as mandatory:
I promise in advance that
* nobody’s grade can decrease because someone else did optional work that they didn’t because I calculate all grades first without counting the optional work, which is recorded separately
* one can get the top grade without doing any optional work
* the total impact of all optional work on grades is very limited.
The optional problems I assign are usually harder than the regular exercises. With these incentives I get an interesting but limited cross section of students doing the optional work. For the students who are already at the top and who do a lot of it, it typically has no impact on their grades at all but it keeps them engaged. On the other hand it also energizes some students at all levels.
With all this positive stuff about grades, even in the low stakes context, designing graded material, and grading it, is still the least desirable part.of teaching. Part of the difficulty is that mechanism design requirement that grades should be correlated with understanding\. Nothing in teaching is worse than grading a problem where one has screwed this up.
I share Paul’s experience.
Another useful trick is allowing collaboration between students for solving assignments and as long as they understand the solution and explain it to me it is OK to get the solution from anywhere they like (Internet, other students, …), they just need to write themselves and separately without copying each others papers. If you suspect a student is cheating you can ask him to come to the office and orally explain the solution.
Problem solving ability of students also need to be evaluated, but that can be done by in class tests or exams where cheating can be made much more difficult.
It’s a tough question. It seems to outsiders like there’s a simple answer: “make the classes more interesting”. But anyone who has ever done any kind of instructional design knows that this is ridiculously hard to do. I’m not saying it shouldn’t be done, just acknowledging that it’s rough.
I don’t claim to have all of the answers, but I have had a moderate degree of success. I teach courses in web development, from Freshman to Senior level, for a B.S. degree in the same. Students generally enjoy my courses, while also acknowledging that they are challenging. And I’ve almost completely eliminated cheating.
But I haven’t done it by eliminating grading. I’ve done it by eliminating guidelines. The assignments in the courses I’ve designed follow a common pattern: “here are the objectives you need to prove competency in, now you figure out the best way for you to show me that you’ve got them down”. I also mix in non-coding work with the code to ensure that students can relate the code to real-world situations.
For example, one of the courses I teach is an intro to programming that uses JavaScript. For each assignment, the student needs to take a story and tell that story using console.log and the required language elements. So if the assignment covers variables, data types, and output, their code can be anything that tells a story as long as they use variables, specific required data types, and output. The story must be something personal to them. I get stories about hockey, pizza, Stargate SG-1, and a dozen other themes. There’s no cheating, and each student gets to write something that means something to them. And because there’s no linear way to get to the One True Solution, each student has to find their own path, proving that they aren’t just regurgitating syntax.
There are still grades, but it’s almost more checklist style: did you get the required elements and did you use them correctly? Screw all of that canned stuff where they need to write a Fahrenheit/Celsius converter, or a Fibonacci function. Show me that you can code, not that you can Google for the right code.
Grading that many unique works may sound like a chaotic nightmare, but it’s no worse than digging through code that should all be the same. It takes the same amount of time. Better yet, I enjoy grading because every one is different and far more amusing because the students *want* to have fun with it.
There’s an analogous problem in research – we spend too much time weighing up the merits of each others’ papers.
A quick question: Do you look at candidates’ CGPA while accepting graduate students to work with you?
Sure. If the signal is there, I use it. But’s it’s far from being my main consideration.
If the colleges were better, if they really had it, you would need to get the police at the gates to keep order in the inrushing multitude. See in college how we thwart the natural love of learning by leaving the natural method of teaching what each wishes to learn, and insisting that you shall learn what you have no taste or capacity for. The college, which should be a place of delightful labor, is made odious and unhealthy, and the young men are tempted to frivolous amusements to rally their jaded spirits. I would have the studies elective. Scholarship is to be created not by compulsion, but by awakening a pure interest in knowledge. The wise instructor accomplishes this by opening to his pupils precisely the attractions the study has for himself. The marking is a system for schools, not for the college; for boys, not for men; and it is an ungracious work to put on a professor.
I have homeschooled both of my children and I don’t test and don’t grade. My daughter’s college entrance exam was the first actual “test” she’d ever taken.
I have the advantage, though, in having more of a master/apprentice relationship with my kids, one that most teacher’s simply can’t afford to cultivate.
“We” grade to check if your education-delivery is working and for whom.
“We” grade to find out if they are “getting it” and if they aren’t, it’s time for a better-different education-delivery.
“We” grade to track progress.
Most businesses look at degrees, not necessarily grades, unless they suck, which means “you didn’t get it”.
I didn’t have the greatest grades in the world. Something like a B- average, and I have had jobs in a number of fortune 500 companies because the degree got me the interview and my personality won the job. Grades were never an issue.
I am amazed at two key concepts that are ignored in this article:
1. Motivation. Without extrinsic motivation most students won’t apply themselves, or to the degree otherwise that they would. (We can talk intrinsic motivation but, let’s be realistic here) Grades provide a tangeble means of measuring performance, which forms part of the motivation mechanism. Only commenter Paul Beame really alludes to this.
2. Statements such as, “… more than half of their time is spent on optimizing their grade and less than half of it on optimizing their understanding” skirts the fact that teachers have the ability / onus to grade for demonstration of understanding. For example, grade to the ability to apply / demonstrate what is learned, rather than simply memorization (where such learning is lost soon after the exam).
I completely agree. I know I and many of my fellow students wouldn’t have been motivated to do half as much work if we weren’t going to be tested on it.
Also, you’re forgetting the research that shows doing an exam is a better learning experience than doing study (http://www.sciencemag.org/content/early/2011/01/19/science.1199327.abstract). That is, you _need_ to do assessment in order to learn things properly.
Sure assessment is a PITA, but it is valuable, and indeed essential for good learning outcomes.
[…] read the whole article, click here. I’m always glad to see professors evaluating this problem. To me the answer seems clear: […]
> even the point of a degree is cast into doubt if people just cheat their way to getting it
I strongly disagree.
Unlike research, where understanding matters, — industry is all about “getting the best score, given a set of rules and restrictions”. And that’s exactly what students learn for their degree: to be inventive and competitive (against the teacher), to wage an arms race, to shoulder their way through; basically, to succeed outside of the ivory tower that the professors may think universities are.
IMO, Panos Ipeirotis and the like are misunderstanding the point of undergraduate studies. As Noam mentions, it’s *not* about delivering knowledge. That’s what graduate studies are for.
@ A. Skrobov. “…misunderstanding the point of undergraduate studies… it’s *not* about delivering knowledge. That’s what graduate studies are for.”
Please elaborate. I’m interested in hearing your take on the point of undergraduate studies. Thank you.
Academics spend half their time evaluating other people and half their time being evaluated by other people. This includes students, conference PCs, journal reviewers, grant proposals and on and on.
The same effects that grades have on teaching, they have on research (maximize publications, grants, etc).
The whole system is rotten.
What would happen if you stopped grading? Like, what happens in classes where you announce at the start that everyone will get an A? Would you be subject to institutional sanctions? Would students get uncomfortable? Would opportunistic absentees fill up the roster?
Abstract:
A large emphasis on grades does not encourage understanding of material. Therefore, providing incentives for students to learn and distinguish themselves outside of the classroom is an absolute must!
Body:
Students purchase degrees with the expectation that the degree will lead to a job. Universities must introduce a grading system as a quantitative form of quality assurance(in addition to admissions selectivity, etc.) so that companies will continue to hire their graduates. Then, universities can boast about their placement rates and sell more degrees. In the short term, cramming and cheating work well for many students in this type of system and going above and beyond is not rewarded.
I think that universities and their students would benefit tremendously from inviting industry to take a more active role in education(more than just internships and a job-fair or two). After all, the reason students are buying these expensive degrees is to find a job. Company-sponsored competitions/clubs/seminars are fantastic for inspiring students to actually problem-solve and apply knowledge rather than memorizing facts.
This sort of industry involvement would essentially be a hyper-extended interview, where recruiters get to observe first-hand the character/ability/leadership of potential employees. Likewise, the students would evaluate potential employers and have an opportunity to stand out. Greater industry involvement would also reduce the emphasis on grades and decrease incentives for cheating/cramming/whining to the dean about a justly awarded bad grade. Maybe this is already being done to a greater extent than I realize, but I haven’t seen it very often. Thanks to anyone who read this wall of text.
*Note: Evan, I recently took an abstract algebra class where we only received one grade the entire semester. For homework we were assigned every problem in the textbook(Herstein). Anyone who earnestly put forth effort and sought to understand the material received an ‘A’. I enjoyed that system, but I think it is only a matter of time before a lazy student receives a grade he/she is not happy with and brings the discipline-hammer down on the prof. for unfair grading practices.
**Appendix:
In primary and secondary schools, a different type of grading system(test scores) determines how much state and federal funding is received. Because there is an upper bound on grading scales, the best way for schools to maximize funding is to dedicate resources to increasing test scores of the underachieving students…as they are the furthest from the upper bound and not as subject to diminishing returns. Obviously, this impacts the high-achieving students adversely. Sadly, I don’t see this system changing anytime soon.
I want to say how important it is that students purchase degrees. Employers are the end-user and professors function as quality-assurance technicians. This is not an inherently bad or ‘evil’ arrangement, but we ought to recognize it for what it is and make the best of the situation by inviting companies to take a more active role in the development of the degree-wielding graduates they hire.
If universities are selling degrees, which constitute (among other things) certification and QC for organisations that hire graduates, then it would be quite irresponsible for a university to get corporations strongly involved in the QC process. After all, in this view, the main purpose of the university is to perform this important function so hiring firms don’t have to do it themselves. This is supposed to be one of the universities’ core competencies. It would be one thing to ask some hiring firms to spot-check the quality of their assessments, but to get them involved wholesale calls into question the whole arrangement.
I was recently shopping for a new pair of in-ear headphones. Many manufacturers offer a line of custom in-ear monitors(IEMs). The monitors are molded in an impression of the customer’s ear and the buyer gets to specify color, cable mounts, frequency response, etc. Then, the manufacturer is responsible for the production and quality of the IEM. Although the upfront cost for a custom IEM is much higher than the cost of a universal IEM, the consumer is sure to enjoy a higher level of quality and be happy with the product for many years.
I advocate a similar approach in education. A major difference is that students play a leading roll in their own development and are not inanimate objects like IEMs. Therefore, it would be silly if corporations did not work closely with students and universities to ensure the best fit.
Also, in my IEM analogy, the money makes one less stop than in the education system…but I don’t think this extra step makes universities wholesalers and students retailers. Students usually don’t carry large inventories of degrees for companies to choose from.
Without grades, we’re just going on someone’s word “I approve that this person knows what they’re doing”
Headings and standards give us goals to work toward. A balance needs to be achieved between breaking a subject down into bits and sections, and going too far with scoring.. more room needs to be given to explore and develop real world understanding.
Some teachers do it well, but they are held back by curriculum requirements. Some teachers use the curriculum requirements as a way to excuse their teaching – “covered all the bullet points, it’s their fault they don’t understand”
There should be an area in academics which studies how to improve the way we teach. Assimilates new studies and pushes out new ways to teach people and recognise their ability..
We need a “this person is so passionate about this topic, they actively pursued interest and i believe they are an asset to anyone who hires them” certificate.. signed by epic grand-master people approved.. somehow..
Back to the guild days :D!!