The Process Journal: My Attempt to Get My College Students to Not Entirely Farm Out Their Thinking to Chatbots.
I teach lots of sections of general education research writing courses at a medium-sized private college. It’s a great job, in a lot of ways. But I, like many other writing teachers, had a bit of a crisis of faith this year as I grappled with how to address generative AI. Chatbots are revolutionizing the writing process for my students. I had seen some of my students in the spring ‘23 semester in near-constant conversation with ChatGPT: it just sat open on their laptops, and they would use it to help them take notes, to brainstorm interview questions, and, of course, to draft papers.
I spent the weeks leading up to the fall semester agonizing over what policy to put in place around the use of generative AI and also dreading a year full of hand wringing in department and writing program meetings. I am in an English department with a strong contingent of faculty who are very opposed to any use of generative AI technology in our classes. There are plenty of good reasons for that. I feel pretty ambivalent about letting students use it, partly because it is a lot of work to police it, so an outright ban isn’t practical. I also worry that a really strict policy would encourage deception, and I want my students to think through the ethics of the technology and make their own decisions about how to use it. I am an Enneagram 6 and really like to be in alignment with the institutions I work with, and I carried a lot of anxiety about feeling like I didn’t know what was going to be the best approach for my program, my students, or myself.
But I was really pleasantly surprised in conversations with my colleagues to see a shift in our collective thinking away from the problems of the technology and toward our students: what is it that we really want students to learn? And do our assessments actually reflect the things that we value?
After a couple of inspiring pre-semester meetings, I got the green light from our director to play around a bit with assessment, and I immediately got to work redesigning my research writing course to value process over product. At the University of Texas, I had briefly used the Learning Record, which is a fantastic, complex assessment tool geared toward having students track and reflect on their own learning in relation to course goals. Because the semester was going to start in less than two weeks, I didn’t want to go all-in on the portfolio work the Learning Record requires. I did, though, want to emphasize learning over the end product of a final research essay. And so the Process Journal was born.
Piloting the Process Journal
For each unit, I set up template Process Journal Google docs that students copied and filled in on their own with work samples and observations about their writing process. The journals included everything from notes on readings to peer review feedback to a repository of sources they might use in their projects. Their task was to use the material from each unit’s journal to write short reflection showing what they learned in relation to specific course learning outcomes (CLOs) by going through the steps to complete the assignment.
I made the Process Journal worth 50% of the final grade, and I evaluated it in two different ways:
Daily(ish) class points. There was generally a pretty low bar for this. So long as students weren’t sleeping or detracting from others’ experiences of the class, they’d probably get the point. They usually did some sort of in-class writing in their journal, which I occasionally checked as an indicator of whether or not they were mentally present.
Overall unit assessments of the Process Journal. Students added work samples and/or observations about their writing process to the Process Journal pretty much every class period, and I encouraged them to add anything from outside of class that was significant to their process–phone calls with parents, conversations with roommates, and emails or office hours with me all featured. They were tasked with tracking how they were practicing the (CLOs) in our daily work. I didn’t want to evaluate it super regularly on top of the writing assignments they were submitting. At the end of the unit, they wrote a 3-to-5-page reflection essay to make a case for their learning in regard to CLOs.
The major assignments–a Research Proposal, Annotated Bibliography, and Final Research Essay–were collectively worth 20% of the final grade. I’ve taken a line from a colleague who tells her students at the beginning of every semester that she doesn’t personally feel a need for a big stack of research papers at the end of the course. There are plenty of research papers out there in the world, and it’s the students who are the end product—it’s their ability to find good information and organize it to answer a meaningful question that we care about. In that spirit, the Process Journal would require that students complete the major assignments, but if they didn’t show their work along the way, they would fail the course. They needed to demonstrate some metacognitive awareness of their learning and not just churn out different components of the assignments like little grade-earning automatons.
Polishing the Process
During the semester, I made a couple of changes based on student feedback to try to make the Process Journal a little more purposeful and user-friendly.
I had these beautifully formatted template Process Journals, but after I got a bunch of bizarrely formatted submissions after the first unit, I taught them how to use heading styles in word processors.
I added the course learning outcomes at the top of the template with some sample color coding. I found that students had a very difficult time matching CLOs to the tasks they completed to offer a coherent account of their learning, even with them having an example reflection essay. A lot of students really seemed to like color coding class activities as contributing to different CLOs, and I think the visual organization made writing the reflection essays a lot easier.
Feedback from Students
In our class postmortem discussions, I gathered four really significant pieces of feedback on the Process Journal:
That I should make the Process Journal even more central to my course organization. Students did have to juggle a lot of websites and documents: some things were in the Learning Management System (LMS–my institution uses Canvas), or linked to through the LMS (like slides), some things were in the Process Journals, and other things were in their working documents or on other websites. They recommended updating the template with more detail about what we worked on each class day, encouraging students more emphatically to use it to take notes, and to link to the slides there (rather than in the LMS calendar, which is what I’ve been doing).
That I should make the Process Journal more of an ongoing repository for feedback. They suggested that I leave my own comments on their writing in the docs rather than in our LMS (though of course I couldn’t post anything about grades), and to have students share their Process Journals with other students so that they can leave each other peer review feedback more easily.
That I could be clearer about assessment. A shocking amount of students didn’t realize until the last week of class that I used a rubric to evaluate the Reflection Essays and Process Journals, even though I had showed them the rubric at the beginning of the semester, used it in the first two units, and posted completed rubrics along with my comments to the LMS when I graded the assignments. One student suggested I link to the rubric at the top of the templates, which I thought was a great idea.
That I should keep using it. I didn’t have a single student out of sixty say they thought it was a waste of time or that I shouldn’t use it again. They were helpfully critical of other things about the course, so I think they would have told me if they weren’t getting something meaningful from it. They liked how the Process Journal kept them organized and gave them a tangible sense of the progress they made and the sheer amount of work that they’d done. They also did almost all of the work for it in class, so it didn’t feel like much of a lift for them.
I’m sure that some of my students still use Generative AI in ways I’ve deemed inappropriate for my class, but I suspect that they are not doing as good of a job of showing their work. There are some final essays I’ve received with suspiciously well-written paragraphs that are vapid and lack any references, for example. If they don’t show me that they have had thoughts about sources they’ve found (and documented how) and read (and documented how), their process grade will suffer. But it isn’t (just) about punishing students for using AI–it’s not rewarding students for a kind of work that is meaningless within the framework of my class.
And conversely, through the Process Journal, I think I’m making the experience of the class more meaningful for the students who want to challenge themselves and learn something. They have *so* much pressure on them to get good grades, and although I don’t think that they should have to feel that pressure, that isn’t something I can fix. I *can*, though, rework my assessment so that students get rewarded for really pressing into their learning. And they do have to convince me that they learned something. If their research essay shows me that they don’t know what a peer-reviewed source is, then that lack of learning is going to be reflected in the Process Journal grade. And of course, the students who put in a lot of work and come up with an amazing final product that reflects a bunch of knowledge and skills that they put into practice will do really well, so long as they can explain how their work got them to that point.
So far, I’m a fan of the Process Journal, and my students seem to be, too. I am pumped to try the new and improved version next semester.
