Y’all, I’m deep in the heart of …semester prep and planning! I happen to be redesigning a course grading system for an introductory Linguistics course. I thought I’d share a bit about my process for one course in hopes that it’ll be helpful as you build your own grading practices this semester.

Tl;dr: Grades should indicate learning. Not every assignment students submit needs a grade; not everything you do grade needs the same level of attention; and your capacity matters when deciding how you’ll grade a course.

In the past I’ve used tests to evaluate student learning in this course. But I don’t think tests, that I inherited from a colleague, are the way I want to assess student learning for this course going forward.

Principles of Grading Systems

I’d rather them show me work and explain how they’ve worked through a linguistics “problem,” rather than just seeing if they got the right answer on the test question. Since I’m moving away from tests, that means I need to revise the entire course grading scheme. 😬

I returned to one of my favorite academics and authors Robert Tablert to give me some help this morning. I re-read chapter 3 of the book he wrote with his colleague David Clark Grading for Growth.

Another way to think about the 4 Pillars of alternative grading shown above is:

1. Student work is evaluated based on evidence of learning.
2. Students receive feedback that they can use to improve learning.
3. Not everything students do needs a grade (or mark). If you do use a mark it should be “an indicator of progress” toward the learning objective.
4. Students can revise & resubmit work without penalty (but not limitless times) until the standards are met.

Grading Individual Assignments

When I’m deciding the type of marking (or grading) I want to do for each type of assignment in my course, I ask myself a few questions:

1. Why are students completing this assignment? What’s the purpose?
2. What type of marking or feedback aligns with that purpose?
3. What type (or level) of feedback (marking, grading) feels reasonable and sustainable for me over the entire semester?
4. What are possible effective and efficient marking processes for this type of assignment given the purpose of the assignment and my capacity?

Clark & Talbert also provide a clear discussion and distinction between standards based grading and specifications grading (in chapter 3).

Standards based grading is based on a list of specific skills that students will learn to do. They say this is especially apt for courses with specific skills to practice.

Specifications grading is based on qualities and characteristics of successful submissions. This type of grading is a good match for projects, portfolios, essays where the process and communication about the process is important.

They also mention ungrading which may include criteria or a narrative description and usually students participating in determining the final grade by creating a portfolio that demonstrates learning.

I’ve used a range of these approaches in different courses before.

Applying those Grading Principles to Linguistics Course

For this linguistics course, I am leaning toward using standards and specifications. Here one tiny example of what that might look like. Please note this is a draft (That I’ve already change before I made this post live. This is real life, draft and iterate.)

In Unit 1 – Phonology, students learn to explain, give examples, and discuss the importance of minimal pairs (based on contrastive features) in the field of phonology.

In a traditional course (and on the previous test), I had students answer several questions such as:

1. Give two examples of minimal pairs in ASL
2. Give two examples of minimal pairs in English
3. Define minimal pairs
4. Explain the significance of minimal pairs in the study of phonology.

Those 4 questions were only a small part of the Unit 1 exam. So, it’s possible for students to miss all 4 of those questions, which are core knowledge in the course, and still make a good grade on the exam.

The real problem, is that the exam grade itself doesn’t tell you what students know or don’t know – unless you go looking at the test, item by item.

Here’s what I’m thinking of doing for this same core concept.

Give students a prompt with specifications for what successful submission looks like.

Standards I’m attempting to assess:

1. Can explain the importance of minimal pairs in the field of phonology.
2. Can identify and describe the features of lexical items to indicate if they are minimal pairs or not.

Possible Prompt:
Using this small data set (insert 4 – 6 pair examples), explain what you can determine about the phonology of each language.

Specifications (qualities and characteristics of successful submission):

1. Explain how you can determine if lexical items are minimal pairs
2. Use key terms from the unit in your response (i.e., contrastive features, phonemes, etc.)
3. Explain what linguists can determine about the phonology of a language based on the minimal pairs identified in the data set provided.

I’ll create multiple language same data sets (2 pairs in ASL, and 2 pairs in English, for example. So that I already have a set of items that students can use to revise and resubmit.

I’ll “grade” this on a 2 point scale – Competent or Revision needed (Satisfactory/Unsatisfactory). They either meet all of the specifications, which indicates they met the standards, or the don’t. If the student does not meet the standard I’ll provide feedback about why not, then they’ll have the chance to revise and resubmit – but not with the same data set. In this case, I guess it could be called a reassessment rather than a revision.

Notice in this case, I’m more concerned that they can explain how to determine if lexical items are minimal pairs than if they identify them correctly. The sample data sets are not designed to be difficult to determine – they’re designed to elicit an explanation of how the student make the determination – which is what the standard requires.

When students complete this assessment we will both know if they have successfully demonstrated competence with these standards. Their fluency with these standards is not hidden within a generic 89% on a unit exam.

How are Final Course Grades Determined?

Since I’m not grading based on a percentage or points you may wonder how I’ll determine the final course grade. I’ll create a specifications system similar to what I’ve talked about previously and similar to the one described by Robert Tablert here.

The final key feature I consider when determining my overall grading scheme and how I grade each type of assignment comes down to 4 factors (listed below). The first three I learned from Rebecca Campbell in a POD Conference Presentation. The last one I added myself.

1. Maintain Course Integrity – Does this grading scheme maintain the boundaries of the course? Does it assess student learning on material they’re learning and practicing in this course – not what I expect they should already know (or should learn on their own)?
2. Support Student Learning – How does this grading system support student learning? How might it hinder student learning? How does it account for student differences, preferences, and needs?
3. Mitigate Equity gaps – How does this provide structure? Who benefits most from this grading system? Who might be disadvantaged? How might I mitigate that disadvantage?
4. Account for Faculty Capacity – How does this grading scheme benefit me (the faculty member)? What are the drawbacks of this grading scheme for me? How might I mitigate those drawbacks? What is my capacity to enforce or adhere to this grading scheme (or policy)?

You can probably see how the 4th pillar of alternative grading, revise & resubmit, supports the factors Rebecca Campbell mentioned, but you may wonder how that “revise & resubmit” process works in terms of my capacity.

Just because students can revise & resubmit does not mean they can do so unlimited times. I have to establish boundaries around the number of times assessments can be resubmitted, around the number of re-dos permissible in a given week, and deadlines after which revisions are not longer accepted.

Final Course Grading System Process

Now that I’ve laid out the plan…I’m going back in to determine all of the key standards for the course, determine assessments, and then make some course activity plans.

And then, just like our students do, I’ll assess how it goes, and then make revisions the next time I teach the course based on the feedback (direct and indirect).