Alternative grading as a graduate student
Wins and lessons learned from a course redesign
Today’s guest post is by Carly Gray, a PhD candidate in developmental psychology at the University of Washington. Throughout graduate school, Carly has worked as a teaching assistant serving nearly 2,000 students, led a departmental graduate student working group to promote inclusive pedagogy, and dove into alternative grading when revamping her department’s undergraduate research methods course in her first quarter as instructor of record. When she’s not teaching or conducting research, she spends her time with her husband and their 70-lb mutt, Toby, exploring Seattle and its beautiful surrounding natural areas, trying new recipes, or playing board games.
I came to graduate school at the University of Washington eager to learn and conduct research, but also thrilled to teach. When it came time to teach my own class—Fundamentals of Research Methods, or PSYCH 209—I was excited to have the responsibility but anxious to do it well. I was afraid of letting students down, of not providing the education they’d come to UW for. I had a hard time feeling confident that the 3.7, 2.8, or 4.0 grades1 I would be responsible for assigning at the end of the quarter really meant anything about what they’d learned.
Desperate for some guidance, I asked a teaching mentor, Lauren Graham, the current instructor of PSYCH 209, for a book recommendation. I’d read educational philosophy in undergrad: Freire, hooks, Dewey, and others. But I found myself in search of practical tips. How do you actually run a class? How do I make sure students learn something and that their grades reflect their learning? She and I began a two-person book club on Grading for Equity by Joe Feldman.2 Following a nudge from Lauren, I committed to implementing alternative grading in PSYCH 209 in my first quarter of teaching as instructor of record.
I chose to implement alternative grading for several reasons. I use these “whys” when introducing the alternative grading system to students but also to remind myself of why my efforts are worthwhile:
To focus on learning rather than on points
To communicate clear learning goals for the course and for students’ learning and studying
To create transparency and meaning in grades
To build in opportunities for feedback and reattempts
This blog post is about how I went about the redesign and my reflections on trying alternative grading as a graduate student instructor. I’ve kept nitty gritty details to a minimum, but you can find my syllabus, interactive grading checklist, and other materials in this shared folder.
The context
UW is a large public research university in Seattle, Washington with an undergraduate population of about 33,000 and nearly as many graduate and professional students. Psychology is a “capacity-constrained” major at UW, meaning our department admits about 175 students into the major per quarter. The course I’m discussing here, PSYCH 209, is one of three courses students must take before applying to the major.
Most recent renditions of PSYCH 209 have followed a similar structure. Students meet for a 50-minute lecture on Monday, Wednesday, and Friday and 50-minute discussion sections on Tuesday and Thursday. The full class enrolls 200 students, and the eight discussion sections enroll groups of 25 students. Students complete four exams, including a final exam (sometimes comprehensive, sometimes not), four to five writing assignments including a final research proposal, weekly reading quizzes3, and some type of low-stakes weekly worksheet or discussion board assignment. All of this happens in just 10 weeks of instructional time plus a finals week since UW follows a quarter schedule. The class is typically considered challenging and a heavy workload among undergraduates, and graduate students know it to be one of the highest TA workloads in our department.
How I did it - a newbie’s first steps into alternative grading
My implementation of alternative grading for 209 involves standards-based testing, plus specifications-like determination of final grades. Students have multiple opportunities to meet learning objectives (LOs) through multiple-choice exams. Additionally, students complete four writing assignments (including a final research proposal) that are considered in their final grade, but these writing assignments don’t assess LOs. Students’ final grades are based on the number of “critical” LOs met, the total number of LOs met, their final research proposal grade, and their grades on the other three writing assignments. For example, to earn a 4.0, students must meet all LOs (including five “critical” LOs), earn Excellent on their final research proposal, and earn Excellent on the other three writing assignments. The syllabus and interactive grade checklist in this shared folder provide more details about the grading scheme.
To get to this alternative grading approach, I worked backward. I began by identifying LOs— what do we want to be sure students can do by the end of this course? Fortunately, others and I had already collected the learning goals and objectives listed in recent 209 syllabi as part of a separate effort. I added these to the detailed LOs available in the course textbook, Research Methods in Psychology by Beth Morling, and culled this long list down to LOs that we taught in most iterations of 209. After identifying possible LOs, we decided the existing assignments were generally sufficient to evaluate the LOs. We decided LOs were best assessed through standards-based testing, while writing assignments would be assessed using more traditional rubrics.
To finalize the LOs we would test through standards-based grading, we decided we wanted the final number of LOs to be divisible by three so that a new set of LOs could be tested on each of the three exams, along with previously tested LOs to honor my fourth “why” and the alternative grading pillar of “reattempts without penalty”.4 For the first summer, we settled on 15 LOs, each of which would be assessed through five multiple-choice questions. Students needed to correctly answer four of the five questions on a single exam to have “met” the LO. Five of the 15 LOs were considered “critical,” meaning they were considered differently in the final grading scheme.
In an attempt to not absolutely overwhelm myself, I made only minor adjustments to writing assignments. Each assignment was evaluated with a rubric and assigned an overall grade of Excellent, Satisfactory, Needs Improvement, or Insufficient Evidence. Students earning Needs Improvement or Insufficient Evidence could resubmit their assignment within a specified time to replace their initial grade. As described above, final grades were based on the number of LOs met and writing assignment grades.
In brief, this system worked out well! Below I share my reflections on what made this possible for me to try as a grad student, what was hard or didn’t work out well, and some reflections for grad students (or others!) exploring alternative grading strategies.
What helped me, a grad student, take on this project
Multiple previous exposures to the class. In 2022, I took a teaching assistant (TA) position that committed me to TAing PSYCH 209 all three quarters during the academic year. Rather than teaching discussion sections as TAs typically do, I was responsible for creating the materials other TAs used to teach these discussion sections, facilitating grading reliability, and creating continuity among instructors from quarter to quarter, plus it allowed me to teach 209 in the summer to a smaller group of up to 50 students.
This year-long teaching position was the perfect on-ramp to my first experience as an instructor of record for several reasons. In our department, graduate students often bounce around between different TA positions without many opportunities to develop deep familiarity with an existing course. In contrast, this role, plus my earlier TA experiences in 209, exposed me to three unique ways of teaching 209 from three different instructors over five quarters. I took notes throughout the year on tiny details and big-picture approaches that I found compelling and that were effective for students. This helped me gain the confidence and experience to take on this alternative grading redesign.
Additionally, by the time I was thinking about my summer course, I had spent two quarters creating a repository of activities, slides, and optional assignments that I could adapt relatively quickly to each week’s needs and the instructor’s preferences. I used some of the time I’d usually spend on typical TA work to redevelop this alternative grading strategy.
If a similar position exists in your department or institution, consider applying to develop your expertise in a specific class. If no similar position exists, you may consider TAing multiple times for a class you’d likely be asked to teach if you pursue an academic career (e.g., research methods for a psychology department).
An existing class and assessments that already worked fairly well. Other resources and relationships were critical for my redesign efforts. As mentioned above, all instructors for 209 were already using the same textbook and, for the most part, sequencing topics similarly and using similar assignments. The textbook provided detailed learning objectives for each chapter. My efforts were focused on rethinking the grading of what already existed rather than redesigning course objectives and assignments.
If you use a textbook for your class, or even if you refer to several textbooks to prepare your own materials, consider what existing textbook resources can help you implement alternative grading.
Teaching mentors supportive of alternative grading. Finally, mentorship and encouragement from trusted teaching mentors also made my alternative grading experience far smoother and enjoyable than I could have hoped. I met with Lauren every other week during the spring quarter to share my latest iteration of the grading scheme and hammer out details. Her wealth of experience and commitment to the principles of alternative grading provided the encouragement I needed.
Identify the people you trust and admire as you build up your teaching and TA experience and ask them for help!
What was hard as a grad student
Despite the relative success of this redesign, it was tough to be a new instructor and alternative grader.
Managing TAs who are my peers. As a predoctoral instructor, I found it somewhat uncomfortable to manage my peers as TAs. While this would have been the case regardless of my grading approaches, alternative grading added the responsibility of getting TAs to buy into alternative grading. I was uncertain about how the new grading approaches would affect their grading responsibilities (i.e., how much time grading would take), which made me anxious to ask TAs to support this approach.
To address this, I met with the teaching team before the quarter started to share my reasons for alternative grading, including being honest about my uncertainties. I also worked with undergraduate peer tutors who graciously shared their feedback throughout the quarter from an undergraduate perspective. Fortunately, I worked with great teams of TAs and peer tutors both summers who came to appreciate the system I’d developed. Without their work, this would not have been possible.
Negative student feedback. One of the hardest moments as a new instructor was when one student formally contested their grade at the end of the first quarter I taught the class. This plagued me with guilt for a while, feeling like I had taken too big a risk and had no authority as a grad student to have made choices that may have led to a lower grade than this student would have earned under a traditional grading approach.
I had to remind myself of one of my why’s: this grading scheme is meant to create interpretability and meaning in grades. Fortunately, the faculty member handling the contested grade supported my grading decisions as fair and appreciated the alternative grading approaches I’d implemented. This circumstance was bound to happen eventually in my career, so I was glad to have been in a supportive environment when it did.
Changes I’ve made
Both summers, I adjusted my grading scheme to allow up to two LOs to count as met if a student had demonstrated improvement on the LO throughout the quarter, but only if the class reached 80% or higher completion of the course evaluation. I did this based on a quick audit of student grades before the third and final exams and realized students’ grades seemed to be on track to be lower than in previous quarters. I also wanted to reinforce students’ dedication to their learning by rewarding improvements on LOs rather than just my predetermined standard of proficiency. I don’t know if I would continue making this adjustment in future iterations, but I am glad this incentivized high course evaluation response rates in both quarters, providing me with ample feedback at this stage of my career.
I also cut the number of LOs down from 15 to 12 to cut out some redundancies that became apparent the first summer. To make this work with standards-based testing, I tested four (rather than five) new LOs per exam. Students had to answer five of six (rather than four of five) questions correctly on an LO to meet it.
I added an engagement objective, so really there were 13 LOs in the course’s second iteration. I’d previously made InQuizitive assignments and worksheets completely optional but would give students feedback if they submitted worksheets by certain deadlines. This worked fine, but seemed to disincentivize student participation. For this 13th engagement objective, students had to engage with the course through at least two methods (participation in sections, lectures, InQuizitive assignments, or drop-in hours) for at least 80% of the quarter. For sections and lectures, we tracked this via their completion of in-class activities, like PollEverywhere questions and group activities and discussions. InQuizitives were easily tracked in our LMS based on whether students met a completion threshold. Drop-in hours were informally tracked by the teaching team at the end of each of our respective hours.
A case for considering alternative grading methods
Whether you’re a fellow graduate student or not, I’d like to offer my case for trying out alternative grading early in one’s teaching career.
First, the process of developing an alternative grading strategy is likely to force you to really think through what it is you want students to learn. I firmly believe this helps us become better educators.
Second, alternative grading can be a way of demonstrating care for students. I found that students generally responded well to this grading approach, in part because they could see how much I’d thought about and put effort toward their course experience and learning. I shared a video on our LMS explaining the grading system so students could refer back to it throughout the quarter when confusion arose and built multiple grade check-in activities to the class lectures to support students adjusting to this approach.
Third, it can help you and your students focus their efforts effectively (see my second “why!”). Standards-based testing meant that students knew the topics and skills they needed to study for the next exam as soon as their grades were released. My office hours visits were typically efficient: students often came to review exams with questions about specific topics, and if not, we could readily identify what they needed to work on together. These conversations were often more centered around their understanding of course content than on what they needed to do to pass the class or to earn a certain grade. The LOs also guided my preparation of lectures and discussion section activities.
Fourth, it might be a worthwhile “risk” as you establish your teaching career. Since implementing this alternative grading strategy in the summers of 2023 and 2024, two full-time teaching faculty have implemented and adapted the same approach, reaching over 600 students. It doesn’t appear that this alternative grading scheme significantly changed the grade distribution in the course compared to previous quarters taught by the same instructors. I’ve presented this approach at a poster session of a national conference and will co-facilitate a workshop on alternative grading for UW’s teaching and learning community with Lauren. I also received a university-wide teaching award, based in part on my redesign of 209 according to alternative grading principles. I expect the thought I’ve put into my teaching philosophy and approach, as well as this recognition, will be helpful in my future search for academic jobs.
So give it a shot! Know that you’ll hit setbacks and have to adjust on the fly. Most importantly, consider your whys and find mentors you trust to guide you along the way.
UW uses a numeric 4.0 scale rather than letter grades. Passing grades range from 0.7-4.0 in 0.10 increments
David and Robert’s book Grading for Growth hadn’t been released yet. Otherwise, I’m sure it would have been our pick!
We use InQuizitive (link to demo), an adaptive quizzing program that complements the textbook. Students and instructors typically really like this platform, which is fantastic!
Specifically, exam 1 covered LOs 1-5, exam 2 covered LOs 1-10, exam 3 covered LOs 6-15, and the final exam included all 15 LOs so students could reattempt any LOs they’d not yet met.
 | A guest post by
|