Planning for Grading for Growth: The Final Countdown
With this post, we are concluding our ongoing series on how to build an alternative grading system into one of your courses. To recap:
Part 1 gave a framework of five questions for getting clear on the context of your course.
Part 3 looked at finding the primary focus of your course and using it to draft a list of assessments you'll use.
Part 4 focused on choosing the marking methods (or lack thereof) and crafting a narrative description of the grades of "C" and "A".
Part 5 dove into designing the feedback loops in the course and prototyping a single assessment from your list from Part 3.
In the forthcoming Grading For Growth book, this series of posts will be in a chapter that's intended as a workbook that leads you through a ten-step process for building an alternative grading system. The five parts I've blogged about make up the first six of those steps. This post goes through the last four steps.
If doing four entire steps in one blog post sounds overly ambitious…. you may be right. On the other hand, there’s not much left to do at this point. Much of the creative work of building a grading system was done in the earlier steps. To finish up, we need to finalize the system, including figuring out how course grades will be assigned; and then think carefully about the whole thing before we launch.
Finalizing the components
In several places in the earlier steps of this process, we postponed actions on certain decisions, so that we avoided getting lost in the weeds at the wrong moments. But now it’s time to look back over all those postponed items and ask questions about them, including:
Your standards/learning objectives. Are there any that don’t make sense to include any more, now that you’ve built more of the system? Are there any you can eliminate or consolidate? Do you need to add any more? (We highly recommend not adding things, but if it’s critical then go ahead.) How will they be posted and communicated to students?
Your overall assessment plan and workload. Is there anything in your list of assessments that no longer makes sense to include? Estimate how much time your students will need to complete assessments, both in- and out-of-class, and make sure it fits within the course expectations. Include reassessments; we advise estimating that they take half as long as the original assessment. Is your plan overly full? Can you cut back on the number of assessments, or eliminate some entirely?
Your marking system. Are you happy with how you plan on marking those assessments, including any plans to use ungrading? How do you feel about the names and meanings of your marks, assuming you are giving any? How will you record these marks (if any) in your LMS? How will you help students keep track of their own progress?
The feedback loops. Are you satisfied with your plans for feedback and reassessment? Are the feedback loops simple? Can you make them simpler?
The last two sentences are well worth repeating: Keep it simple. Is your prototype as simple as possible? If not, what can you do about it? This is not your last chance to simplify your system, but simplification will be harder to do from this point on.
Note that while these items will now be “finalized” (although you can keep editing if needed), we are not finalizing the assessments themselves. Unless you intend to post assessments quite far in advance, generally speaking your assessments won’t be in their final form until they are assigned. However, you can continue to work with additional assessments, getting closer to final versions in each. When you do begin to finalize each assessment, think about:
Along with finalizing an assessment, how will you provide examples of different levels of student work?
If offering reassessments through new attempts, how will you plan to pre-write new attempts to save yourself time in the future?
Assigning course grades
We are now at the point where we can think about how all the feedback and marks in your prototype system will come together to determine a course grade. Finally, you might say, because sometimes when we think about grade systems, the first thing that comes to mind are the course grades (A/B/C/D/F) and what comprises them. But as we’ve seen, a lot of groundwork has to be laid before we can think about assigning course grades in a way that represents real learning and growth, and not just as the outcome of an arbitrary statistical process (using fake data).
Start with the narrative description, developed in Part 4, of “C” (or whatever letter is “minimal passing” at your institution). Now that you have more definite plans, how might you encode this description in concrete terms? Begin to write a grade table or bulleted list of requirements that will eventually become part of your syllabus. Be specific, and write these requirements in terms of the items you’ve developed in the previous steps (standards, assignments, marks, completed assignments, etc.).
There is no formula for how to create this table or list, but it should focus on two things: being minimal, and representing passing. By this we mean that you should include items that any student wanting to pass the course must do, and only those items. Ask yourself if each item is absolutely critical to passing, and if not, leave it for a higher grade (or remove it entirely). You can use the case studies for inspiration.
Once you’ve made a table or list of accomplishments that would indicate a “C”, follow the same process for determining an “A”. Generally speaking, an “A” requires everything a “C” requires along with either more hurdles or higher hurdles, selected from items that aren’t essential for passing. Again, use the case studies for inspiration.
For both “C” and “A” grades, the requirements for the grades should faithfully reflect your narrative description of what those grades mean. For example, if you cannot honestly say that the list of requirements for an “A” represents truly excellent work -- or if excellent work could be indicated with fewer requirements -- then adjust your requirements.
Once you have determined the requirements for a “C” and for an “A”, the remaining letters grades can be formulated quickly:
Requirements for a “B” should reside somewhere between those for a “C” and for an “A”. Not necessarily halfway between, although that is one simple way to do it.
Requirements for a “D” should be fewer than for a “C” but should still indicate the student has given a good-faith effort to engage with the class. Anything less should indicate nearly complete disengagement from the course.
There are no “requirements” for an “F” grade; this grade is given if the requirements for a “D” aren’t met.
The suggestions above apply to ungrading too, with modifications. Students in a partly or wholly ungraded course still benefit from concrete descriptions of work that constitutes different letter grades. But ungraded courses are typically less prescriptive. You could create a list of qualities describing work at the level of a C, B, and A, and then students would be responsible for making the case, using a portfolio of work, that they have met those requirements. These qualities are your standards, and you should be as clear as possible about them.
If your institution does not give plus/minus grades, then you’re done! Otherwise, you’ll need to determine how those modifiers are assigned. You can go several different directions with plus/minus grades. Josh Bowman’s case study shows how to use “in-between” results to assign plus/minus grades, for example. (I’ll take up different approaches to plus/minus grades more fully in a future post.)
Finally, think about how you will help students understand partial progress in the course, for example midterm grades, periodic progress checks, or if a student-athlete needs a weekly checkup.
Details and unintended consequences
At this point, all the pieces are coming together and it’s looking like a system. Congratulations! But, there’s a crucial step to be done next: Find as many places as possible where the system could fail.
A particular focus in this step is the course grade assignment method. It’s the newest addition to your system; and without examining it now, you run the risk of having a first encounter with potentially serious problems at the very end of the term when it’s too late. Some questions you might ask about your method include:
Does it handle edge cases and false negatives? What happens, for example, if a student meets all the requirements for a grade except for one? For example, consider a student who has met all the requirements for an A except for one, but the one they missed doesn’t even rise to a “D” level. Do they really get an “F” in the class? Work out your plan for handling such cases now, not later.
How likely are false positives? This is when a student earns a grade that is higher than their evidence of learning would suggest. False positives plague traditional grading, but alternative methods aren’t immune. Have you set the bar high enough for a grade of “A” (or “B”, etc.) that only truly excellent work in a course meets it?
What is your plan to handle disagreements about course grades? This is especially critical if using ungrading, but it applies to all grading systems. If a student argues that they deserve an A because of how hard they worked, but the actual results of their work are not really “A” level, how will you respond? Conversely, what happens if a student has done work that merits an “A” but they argue for a “B”?
There are other potential pain points in your system:
How likely is “grading jail”? Could student submissions, particularly at the end of the semester, pile up so badly that the grading load becomes unmanageable? Do you need to limit the frequency of reassessments or initial submissions to mitigate this?
How will you communicate and build trust with students? Think about how you will introduce the new grading ideas to students. Will this be too much for students to handle in terms of workload and attention? Is the system so complex that it takes lots of effort just to understand what it says?
Do students have enough opportunities to demonstrate evidence of learning, especially at the end of the term? If there is a standard or assignment that doesn’t appear until near the end of the course, do students have enough chances to demonstrate skill with it? Could there be alternative methods of demonstrating skill?
You’ll likely think of more questions as you do this step; make the time to address them. Ask yourself: What could go wrong?
Sleep on it and simplify
After a lot of hard work building a prototype of your grading system, you need a break. Save your files, close the computer, erase the whiteboard, and go home and do something else, including sleep so you can have a clear head the next day.
A decent night’s sleep has a way of clearing your head and debugging your thought processes. So get some sleep. You will likely see things in your system — good and bad — the following morning that you didn’t see before. And you’ll be better equipped to deal intelligently with issues if you are rested. So seriously — log off and get some sleep.
Having slept on it, there is one more task to complete before the prototype is done: Read through everything and find ways to simplify.
We have stressed simplicity throughout this series, and in many other places on the blog. Simplicity is perhaps the most important part of any grading system, especially one that is different from what students are used to. Grades, for better or worse, are important to students, and making major modifications to how they are earned might be viewed as a threat. Your enthusiasm about alternative grading and the clarity of your explanations are important. But it’s even more important that the system is easy to understand and easy to live with. Leonardo da Vinci said that “Simplicity is the ultimate sophistication”; it is also the ultimate broker of trust.
You may come to your prototype after a night of sleep and find that some parts of it don’t make sense. You may find that the parts make sense but the system as a whole feels unclear or overcomplicated. After sleep, your brain is ready to find holes, glitches, and complexities that you couldn’t see yesterday. Look with a critical eye over everything you’ve done and ask, over and over: Can this be simplified?
It’s one thing to read impassioned moral arguments in favor of reforming grading. Anybody can do that. But it’s quite another thing to put these ideas into practice by building a system that instantiates alternative grading practices. That’s what this series has hopefully equipped you to do. If you have followed along and built yourself a system — even if it’s just a prototype and even if it has some issues that need to be ironed out — then you’ve taken the next step.
David and I firmly believe that alternative grading is not just a good idea but one that can be done on a practical level by anybody inclined to commit the time and effort. We hope this series of posts has given you the tools you need to bring alternative grading to life in your own classes.
The framework we’ve presented may not fit everyone’s situation and there may be placed where it can be improved. If that’s the case for you, please comment!
Thanks for reading Grading for Growth! Subscribe for free to receive new posts in your inbox every Monday.