Reflecting on 3 Unique Grading Strategies in Intro Bio

Note: This was originally written on December 16, 2022 and posted on my personal website.

When I redesigned the curriculum for my Introductory Biology course, I wanted my grading policy to improve equity in learning and equity in grading. To do this, I implemented three unique grading strategies – here’s how it went.

Multiple Grading Schemes

I’ve used Multiple Grading Schemes in previous semesters, but this time I implemented it in a slightly different way. If you’re new to Multiple Grading Schemes, the brief version is that instead of having one grading scheme where, say, homework is worth 15% of the course grade, quizzes are worth 25%, exams are 40%, etc, I have multiple grading schemes. Each scheme weights different components of the course differently – one scheme might heavily weight summative assessments and not include formative assessments at all, whereas another scheme includes homework and weights the summative assessments less heavily as a result. At the end of the semester, I calculate each student’s grade using each scheme and give them the best grade. That way, a student who performs well on assessments isn’t penalized for not doing the homework, whereas a student who consistently does the homework can be rewarded with a slight grade boost if their assessment scores are lower.

I’ve used Multiple Grading Schemes before, but here’s how it was different this semester: I had two “assessment-heavy” grading schemes (schemes 2 and 3), which weighted different kinds of summative assessments differently. I had two kinds of summative assessments: Essay Exams, which included a few higher Bloom’s level essay questions, and the Final Cumulative Exam, which included mostly lower-level Bloom’s questions in multiple choice format. Both types of assessments had revision baked in (see below). When I designed my grading schemes, I reasoned that demonstrating understanding on both types of assessment was important, but I was willing to let one type of assessment do the “heavy lifting” in the grading. Hence, one scheme that heavily weights the higher-level Bloom’s Essay Exams and one scheme that heavily weights the lower-level Bloom’s Final Cumulative Exam.

Each dot pair represents one student’s grades in the highest-scoring and lowest-scoring grading schemes.

Here are my students’ grades, anonymized and sorted by final course grade. A few things are obvious from this graph:

1. Almost all students who received an A had very little difference between their grade in different grading schemes. This means that they performed well on all types of assessment and they consistently completed the homework.

2. The highest-performing students got their highest score using Scheme 3, which weights the Essay Exams the heaviest. These students demonstrate exceptional understanding at a high Bloom’s level; it feels appropriate that they should earn an A.

3. The biggest spread between grading schemes occurred for students with the lowest grades. This means that the students with the lowest grades exhibited the most amount of variability in their performance – either not completing all of the homework (which was graded on completion only), or performing poorly on some but not all assessments.

4. Interestingly, students receiving the lowest grades were most likely to get the highest grade using Scheme 2, which weights the Final Cumulative Exam most heavily. I interpret this to mean that a) they learned from earlier quiz attempts, which covered the same material as the Final Cumulative Exam, and b) their understanding of the content was shallower, since the Final Cumulative Exam included lower-level Bloom’s questions compared to the higher-level Bloom’s Essay Exam questions. It feels appropriate that students who demonstrate passing-level understanding on the lower-level Bloom’s assessments but not on higher-level Bloom’s Essay Exams should get a C grade.

Overall, I appreciated having two types of assessments that assessed different levels of understanding, and I will continue to use this grading structure for future iterations of Bio111.

Late Work Tokens

This semester, for the first time, I implemented “Late Work Tokens.” Briefly, each student started with 3 Late Work Tokens that they could use to turn in (almost) any assignment late with no grade penalty. It was a good idea in principle – how did it work out in practice?

I have never had students meet deadlines as consistently as my students did this semester!

Not only did they meet the time and date deadlines, there was an obvious system for them to turn in work late, which they used. I provided a Google Form where they could submit a request to turn in an assignment late – this reduced friction for them, since they didn’t have to email the professor (email seems to feel onerous for most of my students), and it made tracking late work easy for me, since it was all captured in the Late Work form on Google.

I really like using Late Work Tokens because:

1. It provides artificial scarcity, which motivates students to limit the amount of assignments they turn in late without having to reduce their final grade for turning in late work

2. It provides deadline flexibility for students who need it, but still promotes deadline structure for students who need it

A note about the number of tokens: My course had 39 graded items that could be submitted late. Twenty-four of those were homework assignments that were graded on completion only, and which were each worth 0.4% of the final grade (or 0%, depending on the grading scheme). In this environment, 3 Late Work Tokens felt like the right number; most students prioritized using Late Work Tokens on assignments that had a significant impact on their grade, like Essay Exam Revisions or Scientist Spotlight homework assignments.

Grading Essay Exams using the EMRN Framework

This semester was the first time I assigned “Essay Exams,” which each consisted of 4 higher-level Bloom’s questions. These took time to grade, since each student submitted 1-3 paragraphs in response to each question (for 40 students, that’s 160 multi-paragraph answers to grade three times during the semester). To grade these, I used the EMRN framework, which I learned about from Robert Talbert’s blog.

Essentially, the EMRN framework is a 4-bin grading scale. This made grading essay exam responses pretty easy, since I only had to decide which of 4 (pretty different) bins each student’s response fit into (following the flow chart helped make the grading process easier!).

Because I use a points-based grading system, I needed to convert the EMRN framework into a points-based system (some of my Alt Grading colleagues will probably cringe at this…). I assigned the following points to each grade:

E = 2.5 points

M = 2 points

R = 1 point

N = 0 points

Since each Essay Exam had 4 questions, a student who received 4 M’s on their Essay Exam would receive 8/10 points or the equivalent of a B, which felt appropriate for “Understanding of the concepts is evident… some revision or expansion is needed.” Students who received 4 R’s would receive a 4/10, which would be failing – appropriately so, if the work did not demonstrate thorough understanding of the concepts. Note that students had the opportunity to submit revisions for each Essay Exam! So, receiving many or all R’s on an Essay Exam wasn’t an automatic grade-killer – they could use the grade information to submit a revision.

One Final Thought about Revisions and Equity

Following the Grading for Growth 4 pillars model of grading, building in “reattempts without penalty” is an important part of using grading to help students learn. For my two types of assessments, revision looked like this:

For the Final Cumulative Exam, throughout the semester, students took 6 online quizzes that covered the same content in the same format as the Final Cumulative Exam. After each online quiz, students took a (ungraded) team quiz using IF-AT scratch off cards to review the most commonly missed questions from the quiz. Additionally, we spent one full week reviewing learning objectives in preparation for the Final Cumulative Exam. All students were required to take the Final Cumulative Exam, even if their online quiz scores showed high levels of understanding.

Each dot represents one student’s grade on Essay Exam 1

For the Essay Exams, students had the option to submit a revision. For Essay Exam 1, 10 students submitted revisions while 24 students did not. Of the 24 students who did not submit a revision, 10 of them received a 5 or lower on the Essay Exam. Essay Exam 3 was a little better in that only 7 students received a grade lower than 5, and of those, 4 submitted revisions. Still I wonder about those students who received a low grade and didn’t submit a revision. Why not? I did build time into the homework schedule for completing revisions, but the revisions were optional. In their book Inclusive Teaching, Kelly Hogan and Viji Sathy argue that an equitable classroom is one where the activities that help students learn are not optional, because some students don’t know how important those activities are to their success, or they don’t prioritize them – and to create an equitable classroom, you must require that students do those activities (by including it in the course grade) to motivate all students to do them.

I suspect that the “optional” aspect of the Essay Exam revisions contributes to some students not taking advantage of the revision opportunity to learn the content better and to earn a higher grade. Should I chock this up to “student autonomy” and say, “well, they had the chance!” or do I incorporate required revisions for students receiving a low grade on the Essay Exam? I’m not sure what is most equitable - or feasible! – for future iterations of this course. If you have ideas, I’m all ears.

I have one more blog post planned in this series of reflecting back on my redesigned Intro Bio: building community in the classroom and how that impacts learning.