Homework Exercises

[Nga Sagastume]

Iam currently taking computer science coursework as part of a computer science related degree at my University. I was wondering if I should upload the solutions I make for various trivial homework assignments (ie. not more than 20-30 or so lines of code, on basic/fundamental topics) to my GitHub account, or if I should only upload larger course/personal projects to it.I have only been regularly coding for about a year, and a large portion of the code I have written was for a company where I was under an NDA, so I was wondering if this would be an appropriate way to add more breadth to future job applications.

Many professors would prefer that you not post solutions to homework assignments to public Github repositories. Why? If they ever reuse those problems in a future semester, then by making the solutions publicly available you are creating a temptation to cheat for other students, and that can be detrimental. So it would be a kindness to avoid publicly posting solutions.

From my experience, I'd bet that most professors would love your enthusiasm but be very reluctant to allow it because of the issues with plagiarism and academic integrity. If you discuss it with the professor you may find that they are open to extra credit projects or something to give you some opportunity to add to your repositories. Go into the discussion fully expecting it to be a long shot.

I've had quite a few syllabi state that storing class work in a public repository would be considered a breach in our academic integrity policies, but at the same time I've also been given permission from other professors to use GitHub public repositories on a number of projects. It usually came down to if the mission of the project is unique to the student/team. When every team is creating a one of a kind project, it usually becomes less of a problem to create a public repository. The best luck I've had getting public GitHub repository approval was with professors that offed independent study extra credit opportunities and when we had group projects.

For instance, we were once asked to model a toy ontology language in a programming language of our choice, and I modelled it in Coq, the proof assistant. I do think there is some value in me having published that solution on GitHub; if only for the reason that it raises my incentive to improve it when I can spare time.

Whether professors generally mind students publishing their solutions, is apparently highly culture-dependent.The other answers by D.W. and Joshua096 report on the default in their experience being yes, i.e., you should always ask professors before hand.

In contrast, in the German academic culture I experienced, the default was that you can share anything. In fact, our student council publicly collects old written/oral exam questions. And professors know this; some even recommend you to take a look at them. Rarely do they kindly ask you to not publish things, knowing they couldn't/wouldn't want to go down to legal path of forcing you anyway.

I have been wondering a while why many universities in the US have graded homework in the STEM field.I completely understand the grading for mid-term exames or final exams. In those exams you have to prove that you have actually learned something in class and that you can work on topics related to that.

However, this does not quite apply for exercises. In my imagination exercises are for revising and practicing content and methods of the lectures. They are also useful to find topics which you might think you have understood, but really you haven't. In my mind this learning environment should be free from pressure to "perform" and "produce results", but instead should be open and honest so that the learning process can be most effective. By grading exercises you create pressure that the students should not learn something (and sometimes fail), but that they should already know all that stuff.

EDIT

Maybe I was too vague in my question, but I am interested why US universities often have graded feedback. I fully acknowledge the usefulness of regular feedback during studying. So I am interested in the reasons why this feedback counts towards the final grade.

Now there are some great answers, each giving different reasons. I feel it will be tough to select an answer, because if would look like I chose it to be the correct reason. Thus I will simply take the highest ranked answer in a couple of days, to mark this topic as solved and thank all contributors.

You are quite correct however in perceiving that this is bad. It creates an unhealthy confusion between formative and summative assessments, and is generally bad for learning as compared to an ideal situation in which students have a stress-free period dedicated exclusively to learning and getting feedback, followed by exams meant to test their knowledge and assign them a grade.

However, such is the culture in the US. Students are generally stressed and chronically overworked, and the ideal conditions that I described above as being most conducive to learning simply do not exist. Moreover, in an environment in which all or most instructors grade homework, any instructor who decides to deviate from this social norm and not grade her students' homework will know that that would cause her students to focus their time and energy on the coursework for their other classes, which would mean they would end up not learning the material for her own class at the level that she wants them to learn it. So instructors are essentially forced to comply with this norm whether they think it's a good idea or not.

Dan Romik's answer that if homework wasn't graded, many students wouldn't do it is at least partly correct (although I can't help wondering whether overall comprehension would increase if homework wasn't graded in any class, allowing students more flexibility to focus their study where it could do the most good. One student's vital lesson is another student's busywork), it misses one very important distinction.

Homework and tests measure two very different skillsets. Timed tests are good at measuring how well a student understands the basic concepts but frankly terrible at judging how well a student can combine/use/apply those concepts in novel ways. Homework allows students to apply effectively unlimited time and resources to any given problem, which is terrible for testing comprehension of basic concepts (they could simply look up the answer) but can be an excellent way of testing whether, given appropriate time and resources, a student can apply their learning to more difficult problems.

If you think about it, graduate school embraces this dichotomy as well, no matter the country. One may consider the thesis somewhat equivalent to a very large, involved homework assignment and the thesis defense equivalent to the final exam. Now consider two students: one student writes an absolutely groundbreaking thesis but through stress, tiredness, or for some other reason completely flubs his defense, while the second student writes an extremely mediocre thesis but absolutely nails the defense. Which of these students would you consider more worthy of the degree?

I was a math professor in the US most of my career. I didn't grade homework in lower-level courses like Calc 1, 2, 3, or Differential Equations or Linear Algebra. My syllabus listed selected problems from each section, and the students were encouraged to ask me or the TA for help if they got stuck. If they wanted to learn the material, they did the homework, and most students figured that out quickly.

For upper division courses, ones with proofs, then I collected and graded homework. This was because learning to write good proofs takes practice and lots of feedback. This had nothing to do with pressure and performance, but just that the student was attempting a proof and I was critiquing it. If I gave 8 out of 10 marks for the homework, then the student knew about how well he was doing.

One additional reason: if homework (and other such activities) are not graded, then it follows logically that the grade depends entirely on exams. Many other countries have a culture of high-stakes exams, so this is not a problem. In the US, however, high-stakes exams are increasingly viewed as stressful, discriminatory, and arbitrary, and so are becoming increasingly rare. Rather, students generally like being able to earn points through homework, participation, projects, or other "offline" activities. Instructors who fight this system and insist on high-stakes exams will not make themselves popular....and since "forcing" students to complete homework generally results in better outcomes anyway, this is a battle that few choose to fight.

One view is that the final grade is not only a measure of competence in the subject matter but also a measure of skill acquired through practice (via homework). In this sense, we can think of homework as a "laboratory component" for a "theory" course and thus it deserves some credit.

Another point of view is that students tend to only work on things that earn them credit. It is a separate and more philosophical question to be debated elsewhere whether this is something they should be conditioned to do!

Also, there are many, many students who just can't perform well in the environment of a test, no matter their level of familiarity of the material. And if class time is taken up with lectures instead of working through problems, then without homework that is required to be done, that familiarity will be poor for most. Combine that with the extreme consequences for failure in American education and anything that would increase the chances for student failure with no benefit is passively malicious.

A good professor in the US should be trying to help students succeed if they can and giving them other places to turn competence into grade points is part of that. It does mean that those with disabilities that cause them to forget assignments end up with poor grades though.

3a8082e126