Summary: Students who have never taken a particular course often can get "credit by exam". However, it's extremely common for credit by exam to be disallowed to students who have previously taken and failed a course. For example, see the policy of the University of Mary Washington. Is there pedagogical reason for that different treatment of students who never took class vs failed the same class?

In my own experience, I have seen way too many undergraduate students that I thought could have passed if they had had two or three more weeks to study. They seem like an ideal use case for credit by exam:

  • Fail the course.
  • Get some extra tutoring over the break.
  • Sit credit by exam at the start of the next semester and earn passing marks.
  • ???
  • Profit Graduate!

This argument applies especially for lower-division undergraduate courses (e.g. Chemistry I, Introduction to Psychology, Survey of US History, etc.) that are especially commonly offered in credit by exam format. These courses typically require a relatively shallow level of mastery of a wide body of knowledge, which can be easier to test for on exams rather than projects or term papers.

Question: Are there pedagogical reasons for not allowing students who have previously failed a course to sit a credit by exam examination while allowing students who have never taken the course to do so?

One obvious argument could be that credit by exam assessments may not be as rigorous as actually completing the full course (attending all lectures, completing all homework, participating in class discussions, etc.). If that is true, then I really wonder why credit by exam should even be offered at all - it doesn't make sense to single out some students who are allowed to follow an abbreviated path and others who are not.

To be clear, I'm not asking for a link to your institution's credit by exam policy nor am I looking for counterexamples (universities where students can get credit by exam for courses they have previously taken and failed). What I'm looking for is a pedagogical argument for having this policy. In other words, if I was in charge of setting credit by exam policy at my institution (I'm not), why would I want to go to the trouble of barring certain students from taking advantage of it rather than just saying that anyone who can pass the exams deserves the credit?

In response to comments, I am asking from the perspective of the USA.

  • The 2 camps: I've noticed a battle raging here in the US between camps over whether degrees ought to be about competence or completion. The competency camp claims that requiring formal study is unfair to those who can't afford to "go to college" but could learn on their own or at low-cost, unaccredited community learning centers. The completionist camp argues that without the "these hallowed halls" experience (being fed scraps of learning by social superiors, microwaving another bowl of ramen before rushing to another 7 AM lecture in an unheated lecture hall in the middle of a New England snowstorm, etc.), degrees are meaningless.

  • There are also curriculum control issues: In competency models, curricula are set by administrators, accreditors, or regulators and professors are expected to "teach to the test". On the other hand, competency supporters will respond that in the completion model, there is no guarantee that the professor will adhere to any standards at all.

We're getting a bit far from the original question, but a big reason for the competence model is the rampant credentialism we've seen here. With professions increasingly requiring a bachelor's degree or even a master's just to get in at the ground level, there are widespread calls to give these degrees to anyone who can demonstrate that they can do the job. Whether this is good or bad is beyond the scope of this question.

  • 3
    A non-pedagogical reason is so that students do not realize that credit by exam is harder than the course; excessively hard credit by exam is economically incentivised. Commented Aug 22, 2021 at 21:55
  • 1
    Are you talking about credit in the sense of getting academic credits toward the number required (e.g., 120) for a degree, or credit in the sense of satisfying some gen-ed/prereq requirement? (I.e., the latter doesn't count toward the total number of classes you need to take, which is AFAIK the norm for what happens when you test out of a class without taking it)
    – Kimball
    Commented Aug 23, 2021 at 0:45
  • 2
    Some rules are mandated by the accrediting bodies of universities and professional schools (law, medicine, etc). I suspect that the accrediting bodies require that a failed course be repeated in order to receive credit. I did once know that my university's accreditor required that all failed course appear on a transcript. They could not be removed if the same course was repeated and passed. This isn't an answer, but it does tell you how deeply such a rule can be embedded in academia. Commented Aug 23, 2021 at 15:12
  • 1
    @David I suspect that too, but I'm really asking whether there is an underlying pedagogical justification for accreditors having such a policy. That is, if I actually had authority to grant credit to anyone who dared to show up and pass my exams, why might I consider actually barring students that I previously flunked from doing so? Commented Aug 23, 2021 at 15:37
  • 1
    @Robert Columbia The authority comes from the institution. Formal procedures to do this will save you. Informal procedures will only get you in trouble. In today's world, formal approval of the Department head and the Dean are necessary to make this possible. Otherwise, I have never heard of any college or university that does this. Commented Aug 23, 2021 at 15:54

7 Answers 7


Short answer: A short exam cannot test mastery of all the skills taught during a course. When such exams are offered, they are normally offered only to students who have successfully completed an equivalent course; in these cases, the exam is used only to "spot-check" a rough equivalence rather than to rigorously probe all the course aims.

Longer answer...

The entire higher education system is largely based on completion instead of (or at least, in addition to) competence. Many people who have never been to college are more knowledgeable than degree holders, but this by itself does not entitle them to a degree. Maybe it should be otherwise, but that’s how it is.

Given this, it is self-consistent that test-passing in lieu of course-completion is largely restricted to lower-level courses that the student has completed elsewhere. The key fact in these cases is that the student has already successfully completed an equivalent course. It is not that passing the exam replaces completing the course; rather passing the exam is merely confirmation that the completed course was sufficiently similar to the equivalent college course. This is the theory at least; I realize that some colleges decide to save time and avoid arguments by just accepting exam scores without checking transcript/syllabi (the number of students who can pass such an exam without completing an equivalent course is probably very small in any case).

Now, you ask for a pedagogical reason why students shouldn’t be allowed to test out of courses. For some classes, it’s true that no short exam could possibly replicate the experience of taking a class. For example: laboratory sessions, research papers, and discussions are important and cannot be replicated in a short written exam. Similarly, in some upper-division science courses, the homework is much harder than the exams. Indeed, it's possible to pass the final exam but fail the course because of bad performance on homework, discussion, lab, etc. In “skills” classes like writing or algebra, however, I agree that disallowing credit by exam is difficult to justify from a pedagogical standpoint.

Finally, you raise an interesting point about final exams: why not allow students to take or retake final exams weeks after the course ends? Setting aside the concept of “placement exams,” it seems like courses that require a final exam do not need to schedule the exam within days of the course ending. I can think of no pedagogical reason against this, but it would be a headache from a scheduling point of view. Further, it would in practice encourage / require students to study during their break periods, which may have deleterious effects.

  • 3
    Interesting point about asking or not asking for a transcript. When I took advantage of credit by exam 20 years ago, I was not required to provide proof that I had completed similar coursework, but in fact I had (but couldn't do a direct transfer for credit since it wasn't accredited, hence the exam). Commented Aug 22, 2021 at 21:11
  • 7
    While my experience might not be the norm, when I took courses like basic English & history by exam, prior coursework wasn't required. It was more that they were at a level which a well-read person should be able to pass from common knowledge, yet were required - apparently because a lot of freshmen weren't well-read :-(
    – jamesqf
    Commented Aug 23, 2021 at 4:20
  • 3
    Agree, there are certain "basic skills" courses (writing and algebra come to mind) where passing an exam seems like a better metric than completing a course. One of the colleges I taught at required almost all students to take a writing course unless they passed a very difficult exam (which few students passed); however, the students who completed the course did not have to retake the exam (and indeed, most still would not have passed). But this is perhaps not an exception to the rule, since virtually all students have past coursework in writing and algebra.
    – cag51
    Commented Aug 23, 2021 at 4:53
  • 8
    "First, I would point out that the entire higher education system is largely based on completion instead of (or at least, in addition to) competence." This may be location specific. In our country the long (for centuries) tradition is that the lectures are voluntary. You only have to attend the exercises/laabs/problem solving/discussions... Many courses are lecture only and you are not required to attend. You only have to pass the final exam. Commented Aug 23, 2021 at 7:28
  • 3
    "it would in practice encourage / require students to study during their break periods, which may have deleterious effects" Please someone tell that to the people in charge at ETHZ. :D
    – Nobody
    Commented Aug 23, 2021 at 10:24

There are various exam systems in the world. In our system it has traditionally been the case that you may repeat the exam within the designated exam period if you fail your first try, up to two times (if there are free spots exam terms scheduled. Only if you fail those, you have to retake the class. This should cover those cases that "students that I thought could have passed if they had had two or three more weeks to study.".

If your systems allows only a single try and you are doomed, then credit by exam sounds like a good alternative as the student indeed may have just needed a more time. If you are honestly convinced about that, I see no pedagogical reasons in allowing such an exams and require all the coursework again.

Or the student may have needed a just quick lesson in how to study at a university. True story: I have never needed to systematically study at the high school and even my first uni exam (physics I) went easily. But then the calculus I came and what I thouth was being reasonably prepared was actually a disaster (O was also a bit ill). On Monday I got my quick leson and failed miserably my first attempt. I started to really study the proofs hard and to really do the exercises from the book. On Friday I passed with an A. I later gradueted with distinctions, later got a PhD, later got an academic job. In the single attempt only syatem I might have easily been marked as a unperspective black sheep with a transcript with no chance of a grad school.

  • 2
    Did the exam on the Friday have the same question as on the Monday?
    – Ian
    Commented Aug 24, 2021 at 9:26
  • 5
    @Ian Is that even a serious question? What sane person would make it that way? Commented Aug 24, 2021 at 9:27

A test-out is similar to transfer credit. It's a way for new students who clearly know the material to skip to the next course. They may be missing a few things and need to work to catch up, but that's understood; and generally only the better students test-out anyway, so can handle it. In a sense, a test-out is saying this person is not our typical incoming fresh-person. If they took this course than our best guess is they'd be bored senseless 95% of the time.

But if a student has taken the class and failed we know that's not true. We have a pretty good idea they aren't ready to move on. Even if they somehow passed the test-out; we have much, much more evidence to the contrary. It seems I'm overly relying on the potential test-out class being a prereq for another, but we purposely make it that way. Our real test of whether you learned Calc-I on your own isn't the test-out, it's whether you pass our Calc-II. In fact, our most popular test-out was only gave you credit after you passed the next class in the sequence.

It works from the student's point of view, for the same reasons. A well-run class is easy to pass. Sure, it takes a semester and hours of homework and tests, but given that, it's not too tough to scrape out a C-. And with a low passing grade you should be a little scared of the next class. If you actually failed the class -- got a F -- you didn't learn it. And if you didn't learn it after hearing the lecture, talking about it, doing the homework, and taking the test directly based on what you just did; over and over; then you're not going to learn it on your own in 2 weeks. The only way not to horribly fail the next course is by retaking this one.

Terrible analogy incoming: it's like speed-dates, coffee-dates and the goal: a night-time date. You might have a speed-date go really well and agree on a night-time date. More likely you'll go on a coffee date. If that's not great, you might get another one. But there's no way you'll have a bad coffee date and argue "I messed that up, but let's have a speed-date that might lead to a proper night-time one". They've already met you.

Then an administrative reason: test-outs and transfer credit are indicated on your transcript. It's a way for the school to tell employers "if this guy sucks, you can see they only took 3 classes from us and transferred in the rest" (true story). If someone "passed" a test-out after failing the class then how would we mark it? I suppose as test-out credit, but also keeping the F?

  • Could someone test out Calc-I then sit in on a few Calc-I classes whiel taking Calc-2?
    – Ian
    Commented Aug 24, 2021 at 9:13
  • 1
    @Ian I don't see how that would help. Calc-II's first week is going to use things from the last few weeks of Calc-I. Better to use the time rereading that chapter (it's probably the same book) and going to office hours. As I wrote, passing a test-out means "don't waste you time taking this class, you'd be bored. You know enough to pick up the few things you may be missing". Commented Aug 24, 2021 at 15:46

A school may give a student who has never taken a course the chance to get credit for it by passing a special exam because the school has no evidence that the student is not competent.

A school may refuse a student who has failed a course the chance to get credit for it by passing a special exam because the school has evidence that the student is not competent.

  • 2
    very simple/elegant/succinct and very good statistical view! but perhaps others may consider it a little too simple given the camp stuff i edited to emphasise. but for me personally i upvoted your answer
    – BCLC
    Commented Aug 24, 2021 at 8:27

I don't think there are pedagogical reasons operating against your proposal, so much as there are accreditation reasons. One major problem with the practice you propose is that it essentially gives students in courses a second opportunity (and possibly more) to attempt their "final exam" to get credit for the course. This essentially means that all courses then involve two attempts at the final examination. Moreover, if there is no limitation on repeated use of "credit by exam" then a student could ---in principle--- continue to attempt examinations until they pass and are credited for the course. This has negative implications for the reputation of the university as an institution that assesses and accredits academic knowledge.

As others have pointed out, universities will generally impose quite strict requirements on giving credit for their courses. In most cases, universities don't credit a student with completion of their own course at all; they just waive the corresponding degree requirement and allow the student to do an alternative set of courses for the degree (e.g., substitute a course that is usually compulsory with an elective). This is usually reserved for cases where a student has already acquired the requisite knowledge from another university (through a course with similar coverage) or through demonstrable professional practice.

  • 8
    I do not any reason for the number of tries to be unlimited, this could be very well limited by the rules (or even laws). Commented Aug 23, 2021 at 4:59
  • 1
    Opinion: if someone studies and then tries to retake a test, and manages to pass after their 10th try, i might be willing to believe they actually learned the content by that point and probably deserve the credit. If the exams are so easy that repeatedly taking the exam guarantees passing without understanding, then the exams need to be made more difficult/comprehensive. do accreditation committees disagree with this? Commented Aug 23, 2021 at 20:24
  • 2
    @frogeyedpeas: Evidently we all do, or students would (already) be given unlimited attempts at university exams.
    – Ben
    Commented Aug 23, 2021 at 20:35

This is probably a good policy to implement.

There are very often cases of students, without any formally equivalent course, using credit by exam to move up (ex: scoring a 5 on an AP Exam without having taken the corresponding AP class (i did this personally at least a dozen times)), in principle we should expect a higher level of competency of someone who failed a course and then scored high on a test, than someone who never took the course and passed the test. The latter might have just gotten lucky, the former at least was DEFINITELY exposed to the entire curriculum (although they failed to display sufficient amounts of it the first time they got tested). In reality, of course most institutions like to categorize people prematurely and so assume the first person is still incompetent while giving benefit of the doubt that the latter might be a prodigy/wunderkind.

  • 1
    You took a dozen AP tests? Commented Aug 23, 2021 at 3:27
  • 1
    Yes, I can recall at least 14 that I took, the university I went to unfortunately did not allow me to transfer credit for all my exams so i don't have the list on hand. I essentially viewed this as an arbitrage opportunity that i would lose access to after high school. Commented Aug 23, 2021 at 3:54
  • Dang, I didn't even have that many near me. Commented Aug 23, 2021 at 14:35
  • 2
    Please rephrase the first sentence to replace "this" by what you believe is a good policy to implement. The question discusses opposing views so "this" is not sufficiently obvious.
    – einpoklum
    Commented Aug 24, 2021 at 20:44

A university degree is not simply a certification that the student has learned the content of some courses. It should also indicate that the student is a responsible person and a good citizen. A student who takes a course and chooses to wait until after they failed the course to learn the content appears to be an irresponsible person. Therefore, students who have failed courses should not receive credit by exam for the same course.

One could reasonably argue that there should be some exceptions when the student has failed for a reason beyond their control.

Students who have failed a course often have some other procedure available to them besides credit by exam; for instance, they may opt to receive an incomplete grade which can be corrected later. This might be better than credit by exam.

  • 10
    "A university degree is not simply a certification that the student has learned the content of some courses. It should also indicate that the student is a responsible person and a good citizen." I strongly disagree.
    – nick012000
    Commented Aug 23, 2021 at 3:33
  • 2
    @nick012000 Are you claiming irresponsible people should receive university degrees? Commented Aug 23, 2021 at 11:12
  • 3
    Anonymous Physicist: In order to strongly disagree with your claim it is sufficient to observe that it's simply not a university's business to assess whether a peron is a "good citizen". Commented Aug 23, 2021 at 13:43
  • 2
    I think this answer deserves better. First, it is an answer about the broad expectations of universities in society. Second, some universities have denied degrees for reasons of character, eg, committing a felony between finishing final exams and awarding the degree at commencement. Third, as I noted above, the requirement to repeat a class might be a requirement of accrediting bodies, which shows there is broad social agreement on this standard. Commented Aug 23, 2021 at 15:24
  • 2
    @AnonymousPhysicist: Your reply is actually a good demonstration of my point: While you consider being a good citizen and being well educated as almost identical, I consider them to be almost independent. This is an excellent illustration of a major problem that arises from requiring that a university degree should indicate that its holder is a "good citizen": "good citizen" is a highly subjective notion, and different people will associate completely different types of character traits and behaviour with it. [...] Commented Aug 24, 2021 at 10:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .