Grading Policies
If you have any questions or concerns, please ask in lecture, during office hours, or on Piazza.
Graded work
-
Guided problem sets are auto-graded on PrairieLearn.
-
Homeworks are graded by the entire course staff, directly within Gradescope. To keep grading consistent, each numbered problem is graded by two undergraduate CAs, under the supervision of a graduate TAs, using detailed and standardized rubrics. Under normal circumstances, homework should be graded within one week of submission.
-
Exams are graded by Jeff and the graduate TAs, also within Gradescope, using nearly identical rubrics as the homeworks. Under normal circumstances, exams should be graded within two weeks.
Extension Requests
We are experimenting with a new automatic extension policy this semester. (In past semesters, we didn't accept late homework for any reason.)
Short version: Students can
request a 24-hour extension for each week's coursework. For each student, three no-fault extensions will be approved automatically; additional extensions require extraordinary circumstances. Coursework can still be submitted up to 24 hours late without an extension, for 50% partial credit.
Post mortem evaluation: Students who respondd to our survey were overhwlemingly supportive of the extension policy. However, technical limitations of Gradescope and PrairieLearn made actual enforcement of the policy more challenging than expected, especially since we allowed for extension requests after the regular deadline. (We erred on the side of generosity.)
-
Both written homework and PrairieLearn guided problem sets can be submitted up to 24 hours late for 50% partial credit, or for full credit with an approved extension. Students can request a 24-hour extension for each week's coursework by filling out this form before the extended homework deadline (usually Wednesday at 9pm).
-
Gradescope will stop accepting submissions 24 hours after the official deadline.
-
PrairieLearn will continue to accept and score submissions through finals week, but scores for submissions more than 24 hours after the deadline will not count toward course grades.
-
Three no-fault extension requests will be approved automatically for each student. (You must actually request these extensions by filling out the form.)
-
Justified extensions will be granted only in cases of illness, injury, accommodation for a documented disability, or similar extenuating circumstances. (Planned events like deadlines in other courses, conference travel, job interviews, or concerts do not qualify as "extenuating circumstances".)
-
Each student must request their own extensions. Extension requests submitted by one member of a homework group do not automatically include other members of that homework group. Extensions apply to both guided problem sets (done individually) and written homework (done in groups), and homework groups can change during the semester.
-
Extensions cannot be broken into smaller pieces. Each approved extension is for a full 24 hours, and each extension applies to both the guided problem set and the written homework due in the same week. You cannot use part of an extension in one week and save the rest for future weeks.
- It is your responsibility to keep track of your approved extensions. If you do not hear back from us after submitting an extension request, you should assume that the extension request was approved. We will contact you only if your request is not approved, or if we need more information. We do not plan to send confirmation emails.
-
⚠️ Technical limitations ⚠️
-
Gradescope cannot show penalties for late homework. Gradescope records a single score to each submission, which applies to the entire homework group, but extensions and late penalties apply to students individually. We will grade submitted homework as though it was submitted on time; we will apply late penalties (for late homework without approved extensions) at the end of the semester when we compute course grades.
-
It may take several days for scores on PrairieLearn to accurately reflect approved extensions. There is no good way to make PrairieLearn report scores that accurately reflect extensions, either. By design, the only time PrairieLearn changes an assessment grade is when the student submits part of that assessment. Fortunately, PrairieLearn also records raw assessment scores, without the 50% cap for late submissions. We will remove late-submission caps (for students with approved extensions) at the end of the semester when we compute course grades.
-
Gradescope allows multiple submissions for the same problem; however, only the most recent submission is visible to the graders. In particular, without an approved extension, late submissions are worth half credit, even if an earlier submission for the same problem would be worth more than 50%. (PrairieLearn's grading algorithm is a bit more nuanced.)
-
We will not accept any coursework more than 24 hours late for any reason. (Longer extensions would delay both grading and posting homework solutions.) However, we will offer to forgive coursework under extreme circumstances where a 24-hour extension would be insufficient, such as a long-term illness, serious injury, disability accommodation, or other similar emergency. We will compute your final course grade as if the forgiven coursework simply did not exist; your other homeworks and guided problem sets will have more weight. Students requiring accommodation for a disability should first contact DRES. Please ask Jeff for further details.
Regrade requests
If you believe that your score for any homework or exam problem is inconsistent with the published grading rubric, or that you were graded more harshly than other students for similar work, you can request a regrade.
-
Submit regrade requests for homework and exams on Gradescope. If you have questions or concerns about any grade, we strongly encourage you talk with the course staff before submitting a regrade request. However, no grades will be changed in any student's presence; you must submit an official request through Gradescope.
-
Regrade requests can be submitted up to two weeks after the graded work is released on Gradescope. Regrade requests for the final exam can be submitted up to three weeks after graded exams are released. However, final-exam regrade requests (and any other requests still outstanding after the final exam) will only be considered if a successful regrade would change the student's course grade.
-
All regrade requests must include a brief written justification for the request. (Fill in the appropriate textbox on Gradescope.) Good justifications include the following:
- My answer agrees with the posted solution, but I still lost points.
- I lost 4 points for an incorrect time analysis, but the rubric says that's only worth 2 points.
- You took off points for missing the base case, but it's right here.
- My answer is correct, even though it does not match the posted solution.
- There is no explanation for my grade.
- I got a perfect score, but my solution has an error. (Extra credit!)
- There is a major error in the official solution; here is an explanation. (Full credit for everyone!)
Regrade requests with poor or missing justifications will be denied.
- We can only grade what you actually submitted. You cannot get a higher grade by explaining what you meant, either in person or in writing; your original submission must stand on its own.
-
The first regrade request for any homework submission will be handled by the TA who subpervised the grading of that homework problem. The first regrade request for any exam submission will be handled by whoever graded that exam problem. Additional regrade requests for the same submission will be handled directly by the instructor. Once Jeff issues a final response to a regrade request, further requests for that submission will be ignored.
-
Followup regrade requests can be submitted after the two-week Gradescope window by submitting a separate reregrade rerequest form.
-
🔥 If you submit a regrade request correctly reporting that a problem was graded too leniently—that is, your score is higher than the published rubric indicates—your score will be increased by the difference. For example, if your original score was 8/10, and you successfully argue that your score should have been 3/10, your new score will be 13/10. Yes, really!
-
🔥 If a significant error is discovered in a posted homework or exam solution, everyone in the class will receive full credit for that (sub)problem. Yes, really! (Jeff gets to decide what counts as "significant".)
Overall course grades
We will determine final course grades as follows. (What do you expect from an algorithms course?)
- Compute total scores from guided problem sets, homeworks, and exams.
-
Guided Problem Sets and Homework = 35%
-
Each weekly guided problem set and each numbered homework problem is worth 1.4% of your overall grade.
-
We will count
a maximum of 9 guided problem sets and a maximum of 16 numbered homework problems, dropping lowest scores if you submit more.
-
Exams = 65%
-
Each midterm has five problems, and the final exam has seven problems.
All 17 exam problems have equal weight.
-
Each exam problem is worth approximately 3.8% of your overall grade.
-
Exceptions:
- Forgiven homework will be treated as if it did not exist; submitted homeworks will have more weight in the overall grade computation. In exceptional cases, we may compute course grades based entirely on exams.
- Forgiven midterms will be treated as if they did not exist; their other exams will have more weight in the overall grade computation.
- We will not drop zero grades that result from cheating offenses.
- Exceptional cases.
- We reserve the right to give any student meeting at least one of the following conditions an automatic F:
- Overall exam average below 25%
- Submitted less than half of the assigned homework problems
- Otherwise does not appear to making a good-faith effort
This rule rarely applies to more than one student out of 400.
- Anyone who misses both the regular final exam and the conflict final exam will be given an ABS (“absent from final”), which is equivalent to an F, unless they get an Incomplete from their college.
- Determine fixed letter grades, according to the following cutoffs. Each possible letter grade above F covers an interval of length 5%. We reserve the right to lower the cutoffs.
- 95% ≤ A+
- 90% ≤ A < 95%
- 85% ≤ A– < 90%
- 80% ≤ B+ < 85%
- 75% ≤ B < 80%
- 70% ≤ B– < 75%
- 65% ≤ C+ < 70%
- 60% ≤ C < 65%
- 55% ≤ C– < 60%
- 50% ≤ D+ < 55%
- 45% ≤ D < 50%
- 40% ≤ D– < 45%
- 0% ≤ F < 40%
- As a backup, we will also compute letter grades according to the following curve.
- The mean is a borderline B–/C+.
- Each standard deviation is worth one full letter grade.
For example, the B+/B cutoff is 2/3 standard deviations above the mean, and the D/D– cutoff is 5/3 standard deviations below the mean. The fixed cutoffs are consistent with a mean of 70% and a standard deviation of 15%. The actual mean has been higher than 70% for several years.
-
Each student's actual letter grade is the maximum of their fixed letter grade and their curved letter grade. In past semesters, 100% of students had higher fixed letter grades than curved letter grades.
Historically, even when students were graded exclusively on a curve, grades in CS 374 have been almost entirely determined by exam scores. In a typical semester:
- About 95% of students have guided-problem-set averages over 95%.
- About 95% of students submit 100% of the homework problems.
- About two-thirds of students have homework averages over 90%.
- About 90% of students have homework averages over 80%.
The following scatterplot shows the distribution of homework averages (x-coordinate) versus total exam scores (y-coordinate) for Fall 2019. (We are well aware that because of COVID, Fall 2021 is
not a typical semester.) Notice especially the outliers: One student had a homework average below 70% but an exam average above 80%; several students had homework average over 90% but exam averages below 40%.
Assuming a total GPS average of 95% and a homework average of 90%, the fixed grade cutoff translate to the following approximate exam averages:
- 96.7% ≤ A+
- 89.0% ≤ A < 96.7%
- 81.3% ≤ A– < 89.0%
- 73.6% ≤ B+ < 81.3%
- 66.0% ≤ B < 73.6%
- 58.3% ≤ B– < 66.0%
- 50.6% ≤ C+ < 58.3%
- 42.9% ≤ C < 50.6%
- 35.2% ≤ C– < 42.9%
- 27.5% ≤ D+ < 35.2%
- 20.0% ≤ D < 27.5%
- 12.1% ≤ D– < 20.0%
- 0.0% ≤ F < 12.1%