When I was a freshman in college, 1972-73, MIT didn't give grades. I believe that was to take the stress off of students, who were prone to obsess about these things. MIT had a high suicide rate at the time and this was one of the counter measures. So, instead, students would get a written evaluation from the instructor. This was done at mid semester and then again at the end of the term. (Ironically, they gave "hidden grades," which were needed for Med school and perhaps some other professional schools.) Somewhere in our house, I've still got my evaluations from back then. If I recall correctly, the mid semester evaluations encouraged a written response by the students.
All these years later, I think this is a preferred way of assessing student work. It is more personal and communicates in a way that an abstract letter or numerical grade cannot. It also offers the chance to suggest paths for improvement, something other grading schemes don't really do. (Office hours after a test might work for that but many students don't avail themselves of that opportunity.) And by giving the student the opportunity to respond, it encourages a learning as dialog approach, which I favor. Students don't get this in their other courses, and I've noticed that much of their behavior arises from conforming to norms that I assume are determined during their first year in college. They have a tough time undoing those behaviors, even when the situation no longer calls for them. I say this because I don't know how much actual dialog my approach will generate. (I posted the first assessments yesterday evening.) Earlier in the course I did suggest that each student meet with me outside of class to discuss their work. Only half did so. This written approach to evaluation offers a different possibility for dialog.
In the rest of this post I want to talk about the technology that supports this activity, what I need to do to make it work, and how that might be improved.
First, this is the part of course where I need an LMS. For the rest of my course I use a blog, and that works pretty well. Indeed, I prefer to use the blog for that part of the course because the software is better done, it is easy to use, and because it is open. But for the evaluation communication, that must be private and it must be secure. So the LMS is the right tool for that.
Second, I'm using Moodle this semester. Once I learned that I'd have a small class and I didn't need a dropbox for student assignments, I decided not to use the Campus Supported Blackboard service. So specific comments are about that particular LMS. Precisely how this translates to other LMS, I don't know.
Third, the built in grade book is not the right tool for this communication. I learned a year ago that the grade book does support columns with text entries in a cell, so conceivably that would be a way to get the message out. But it doesn't enable student response. I should add here that I have both individual work and group work of students to evaluate. I'd like to use the same sort of mechanism for both. On the group work, if one member of the team responds, the other members of the team should see that response. The conclusion is that there needs to be a separate discussion forum for each team and also a separate discussion forum for each student.
This requires making a lot of groups. Each student becomes their own group. Because my class is small the first name of the student is a unique identifier and I use that for the group name. (I've got one student from China and use her American name for this, though that info is not on the official roster from Campus.) Were I to do this in a larger class I believe I'd adopt group names of the form: First_Name (NetID). Then I make a Grouping called Individual Students, or something like that.
I repeat the process for Teams. I've opted for letting the students themselves come up with a team name, but requiring them to have it start with a specific letter, so I have an A Team, a B Team, etc. I again make groups, one for each team, and another Grouping, this one called Teams.
Then I make two discussion forums, one for the individual students, another for the teams. I make these by Grouping, and each group gets a separate forum. It is critical that one group can't see the forum posts for another group. I don't trust my own knowledge of the software to achieve this end. So I need to test this after I've set it up. Fortunately, the people who support Moodle have made some dummy accounts for me for that purpose. The built in - view as student - tool is not sufficient for this purpose, as far as I'm concerned. I typically use a different browser for the dummy student access. That way I can be logged in twice at the same time, one as me, the other as the dummy student.
This setup takes a bit of time. I recognize that for the time being my needs are idiosyncratic and since I'm not particularly time constrained I'm willing to put in the effort. If in the future the approach becomes more standard, it would be good to automate much of this process.
For the numerical part of the evaluation, I give both a number and a written explanation, I need to record that in a separate spreadsheet, which serves as the course grade book. There is no need for that to be in the LMS. My preference is that it is not. Last Tuesday in class, we showed this video with the voice over by Daniel Pink. Elsewhere, I've written that the economics part of this video is somewhat wrong. But on the intrinsic motivation part, and that income rewards distract from that, I believe it is correct. For students, grades serve a similar function as wages serve for employees in the workforce. They can be a distraction from the real learning. We don't talk about this enough. Students fixate about the online grade book. So I don't want one. (In a very large class, the grade book is a crucial management tool, but in smaller classes I don't believe it is.)
I hope from all of this the reader can see some movement away from automation and toward a more human form of evaluating student work. The principle behind this is that teaching and learning needs to take the form of ongoing dialog. This seems straightforward enough to me that it doesn't require an argument to support the point. But it does seem outside much of our current practice. I wonder whether we can change that to be more in accord with the principle.