Your grade will depend on all aspects of class.
There are several pieces to grading: (1) Graded Assignments; (2) Ungraded Assignments; (3) Lecture Attendance; and (4) Discussion Participation. All will be combined in the end. Being a consistent contributor is at least as important as doing an exceptional project.
Canvas will not show your final grade. It will not even estimate it.
We will consider all aspects of your work in determining your final grade.
Note there are two aspects to “grading”: assessment (so we can give you a grade in the class), and feedback on your work. We will try to do both.
Grading in this class is challenging. We do not want to assess your artistic skill or your programming abilities (from before the class).
This class is designed with a large number of small activities that are designed such that if you really do them, and take them seriously, you will meet the learning objectives for the class. If you actively participate consistently, I believe you will learn the lessons. Conversely, the quality of a project may depend on factors beyond class topics (e.g., a student’s implementation skills), and project evaluation can be quite subjective.
Therefore, grading emphasizes the consistent completion of small things, with less emphasis on the quality of a small number of projects.
We measure how well you participate in class by (1) attending class; (2) consistent performance on the “ungraded” aspects of class (online discussion postings, seek and finds, design handins, in class exercises, etc.); and (3) discussion grading.
For attendance: if you do not attend class “enough” (you miss more than 2-3 classes, unless there is some extenuating circumstance), we will deduct 1/2 letter grade from your final class grade.
Otherwise, we will determine a baseline grade from the “ungraded” handins and discussion (generally, 75/25). Consistent Gs will get an AB.
Design Challenge Grading
For “graded” elements (design challenges and projects)…
A great project can pull your baseline grade up by a half-letter grade. A terrible project can pull it down.
Ungraded Grading Scale
For many things, there isn’t enough objective difference for us to determine a grade. And sometimes, it’s just not worth it for us to look at everything in detail - we can trust that everyone in the class is a responsible graduate students and will do a “good enough” job.
Note: ungraded is a bit of a misnomer. It’s more “loosely graded.” And there’s a lot of noise - how we decide between G/E (see below) can be subjective and arbitrary. Please don’t argue about individual marks - hopefully the noise averages out in the end.
So, we will use the following scale:
- N - Nothing turned in
- C - Check - something turned in, but we didn’t
actually look at itlook at it closely enough to decide the grade (we might later determine it to be B/G/E)
- B - Below Expectations - something turned in, but of poor quality
- G - Good - meets our (generally high) expectations of good work in this class
- E - Exceeds Expectations - something that is truly notable
Note that while N/B/G/E is an ordinal scale it is not interval. If you don’t understand that, we’ll learn about scales when we get to data abstraction and measurement. But basically, the steps aren’t necessarily meaningful, so it doesn’t make sense to take about averages. And C is simply different than B/G/E - it’s not better or worse, it just means we didn’t look closely enough to tell.
If a quiz/survey is anonymous, we won’t be able to connect answers to students, so we can only give N/C grades.
Given that we have limited resources for grading, we might just use N/C for an assignment, but later go back to check individual assignments where it might make a difference.
A problem: Canvas really dislikes anything that isn’t an interval (or ratio) scale that can be mapped to a percentage. We can’t use letters as scores. And, when we assign numbers to the letters (so we can enter them in Canvas), Canvas will try to average them and do other things which will make you think they directly map to a grade. Ignore it - the averages aren’t meaningful.
The coding we will use:
- N = 0
- B = 2
- C = 4
- G = 6
- E = 8
Again, this is just a code - you can’t average them. A C isn’t twice as good as a B.
Canvas might give you a “100” for a survey - this is a sign that Canvas gave you a score,and we did not. This translates to a C (4) - you did the survey, but we did not grade it.
Grading discussion participation is hard and subjective.
For the online discussions: there is a notion of quantity and quality. It is useful to say “I agree” to a point made by a classmate, or “nice example” to an interesting posting by a classmate. These are different than longer, more substantive postings, for example to make an insightful observation, or a careful critique of a point. The best online participants have both: quantity and quality, with a mix of posting types. (it would probably be inappropriate to always have a long and insightful posting).
Individual assignment grading doesn’t make sense. For any particular assignment, a student might not have anything new to contribute, or might be in a less conversational group, or might simply be in a quiet mood. Some assignments just seem to lead to less discussion. Therefore, we will only assess discussion in aggregate - over the course of the semester.
Last year, we had a complicated scoring system, and tried to give people feedback throughout the course of the semester. This semester, we will assess your online discussion quality at the end, on the following loose scale:
- Unacceptable - contributes inconsistently, (almost) never provides an insightful posting
- Good - consistently contributes, occassionally provides insightful postings
- Exceptional - consistently contributes, often provides insightful postings
Note that this rewards both consistency and having specific good contributions. It is not about quantity (you can’t just post a lot of things the last week), it’s not just about quality (writing a totally amazing thing once or twice in a semester). We want students to be active contributors (to help each other learn as a community) and also think deeply.
We will group all online discussions together in terms of evaluation: the weekly Online Discussions, the weekly Seek and Finds, and any other discussions that come up as part of Other Handins and Surveys and Graded Assignments (Design Challenges). For all of these, the “initial posting” (the first response to the prompt) is distinct from the postings made later in response to others.
For in-class participation, we can’t assess all the small groups. For “large group” discussion, it is hard to keep track, and the class is big enough that not everyone can have the opportunity to participate. Therefore, we will not try to “grade” in-class discussion. We will assume that if you are there, you are engaged and participating at a “good” level. We reserve the right to penalize or reward outliers.
If you believe there was an error in grading, send a private message to course staff via Piazza. Please tag it with “regrade”, and be clear if you are looking for an explanation (for feedback) or are just looking to recover points.
There undoubtedly errors in grading, and students only point out the ones likely to be in there favor. Statistically, it is implausible that all of the grading errors are against the students. It is unlikely that we made exactly one error. Therefore, if you point out a grading error, we may look at the entire assignment to make sure there aren’t other grading errors, in fact we might need to look at other assignments to check for errors as well.
You must request a regrade within one week of receiving the grade. Don’t wait until the end of the semester.