A Quick Grading Case Study

I ran the following example with a small group of teachers evaluating standards-based grading in their PLC. The teachers are on board (and all are attaching standards to work for feedback and assessment) but they needed to see a tangible example of why standards-based grades can help all students.

I’m going to steal Dan Meyer’s Three Act structure with no shame. Each screenshot uses the same four (fictional) students.

Act 1

Questions for discussion:

  1. What can you tell (or infer) from this information?
  2. What cannot be inferred?
  3. Which student would you focus on for intervention? Why?
  4. What feedback would you give to each student?

Act 2

Questions for discussion:

  1. What do you notice?
  2. Which student would you focus on for intervention? Why?
  3. What can you infer from this information?

Act 3

Questions for discussion:

  1. What do you notice?
  2. What can you infer from this information?
  3. Where would you focus your interventions? Why?

We’re using weighted categories for our students. Looking only at classwork (Act 1) doesn’t show teachers gaps in the student learning. You can certainly target students for intervention, but it is based only on the the task completion, not necessarily the content.

In Act 2, we have a little more to go on because each assignment is aligned to a specific standard or skill. The big takeaway is that the student with the lowest assignment score (row 3) is actually learning all of the standards. The learning gaps are hidden for the "responsible" student who turns their work if we don’t take standards into account.

Act 3 brings it home for teachers. Where do the 1’s and 0’s come from? It’s from aggregated information over time. In this view, color coding (we have set up through Canvas) is a quick gauge of class comprehension on each standard. I can use this information to plan more effectively to help all students reach learning goals.

In the end, teachers wanted to know the student’s calculated score. This table shows what would be on the student report card:

Student Classwork (20%) Standards (80%) Final score (%)
1 85.2 33.3 43.4
2 70.4 50 54.1
3 62.7 100 92.5
4 81.5 66.7 69.7

Standards-based grading can help root out lack of learning by moving the focus away from compliance. Assessing learning goals and making them the focus of feedback and reporting helps make that change a reality.

What is a Grade?

an exit sign hanging on a wall.

I had the pleasure of working with about 15 people yesterday on moving to standards-based grading next year. We started off with a long discussion about what grades are and what they mean. It’s easy to get into what they should be, but I wanted to make sure we all had a solid understanding of what grades actually do in most of our classrooms.

I had a couple of guiding questions and one that generated the most interesting response was the following:

A student rarely comes to class and when they do, work isn’t turned in. At the end of the semester, that student easily passes the final exam. Does that student pass your class?

Lots of eyebrows furrowed.

There was some uneasy looking around.

About half said yes, the other half said no.

Now, there are major assumptions here. Is the test valid and reliable (standards-aligned)? How did the teacher intervene? Did a student show growth before taking the test in some other way?

All issues aside, the root of the question forces us to consider whether a grade in our class represents learning or compliance.

I also wonder why we’re more accepting of the inverse situation: a student who has not taken the class who passes the final is allowed to skip the course (or is given credit, etc).

If we’re comfortable with allowing students to skip a class (be given credit) by testing out we should be just as comfortable allowig a student who "shows no effort" to be given credit for hitting the same benchmark. The difference is our perception of that student.

Challenging our biases is important, particularly long-held assumptions that dictate our perceptions about "good" vs "bad" students. Grades are the output of those biases in many cases.

What do you think?

The featured image is Br… flickr photo by Peter Schüler shared under a Creative Commons (BY-NC-SA) license

A Critical Gradebook

I am finding the right balance of scaffolding to provide the best learning environment for my students.

Source: A Critical Gradebook

The gradebook seems like the most frustrating and under-developed part of any LMS. We use Canvas and have had our own struggles with making the gradebook helpful, not hurtful. Laura Gibbs has more thoughts on that than I do.

The Learning Mastery component of the Canvas gradebook is immensely powerful if you take time to set it up correctly. It’s a shift away from singleton points and gives students and teachers a more high-level view of what objectives/skills/standards a student has attained over time. This can be (but doesn\’t have to be) linked to the students course grade. Again, my view is to stick with Frank Noschese’s Keep It Simple SBG schema.

Translating that is a chore of it’s own, but I’m hacking away at a helper tool…more on that another time. I think this is where something like an LTI tool can help across multiple platforms, if the new gradebook (or commentbook) is flexible enough to focus on feedback rather than a specific assessment protocol.

Reassessing in Standards Based Grading

I’m helping several teachers move toward standards-based grading practices this year. We work a lot on philosophy – why they’d want to use this grading mechanism over traditional scores, how to support learning, and the language of SBG in general with students – before we get into the how-to. That helps make sure everyone is in the right frame of mind.

Once they’re ready to start, that’s where the how-to work comes in. I know what I think about how to set up a class, but there is no gold standard when it comes to actually running the class. If you’re looking to start, allow me to redirect you to Frank Noschese and his excellent blog as well as pretty much anything written by Rick Wormeli.

Today’s post started as an email asking how I handled retests in my class. The following is more or less what I wrote back, with some edits for clarity and more general application.

I’m trying to up my standards based grading game. We briefly talked about this last semester, but I’m wondering…how can I most efficiently update students’ grades to show mastery when I’m having them do test corrections? Ideas welcome!!

This came in an email

Do you do paper-and-pencil corrections? How are you building your tests? I ask because there are a few ways you could consider, but each kind of depends on your own style and class processes.

Grading paper-and-pencil corrections

When I did this, it was usually something like:

  • a) write out the wrong answer,
  • b) write the correct answer,
  • c) why is is it right,
  • d) give a reference to the right answer,
  • e) which standard/outcome does this relate to?

So, they would go through the material, evaluate their responses, and then find the right answer and justify it. I was mainly concerned with the justification of the response, not so much that they found the right answer. I would grade their mastery on that justification, bumping them up or down a little bit.

To track it, you could download the MagicMarker (iOS only) app and mark them on Outcomes as if you were talking to them in class. It aggregates those scores into the Canvas Learning Mastery grade book and then you can evaluate the overall growth rather than give credit based on that one assessment.

Question Banks

This is definitely the most time consuming to set up, but once it’s set up, you’re golden. Getting questions in standards-referenced banks allows you to build out Quizzes that pull randomly, so you can give a retake or another attempt that updates those Learning Mastery grade book results. This is what I tended to do instead of paper/pencil once I had everything going.

Students would get their results and then focus on any standards that were less than a three in their Learning Mastery grades (out of four total). There’d be some kind of work involved so they weren’t blindly guessing, but then they could take the test again because the questions were likely to be different with the bank setup.

Set up banks based on standard and then file questions in there. When you build the Quiz, you use Add new question group rather than Add question in Canvas. You can link the question group to a question bank and specify how many items to pull at X number of points.

Student defense and other evidence

This one is probably my favorite: just giving students a chance to plead their case…a verbal quiz, essentially. I’d use MagicMarker while we were talking to keep track of their demonstration. I would ask them to show me work we’d done, explain how they know what they know, and then prod them with more questions.

I typically did this if they were having trouble demonstrating understanding in other ways. I wanted to remove test anxiety or reading comprehension from the equation, but this was typically the last option for those kids. I’d then work with them to get over those test-taking humps (granted, this was more important to do in the AP class because they had to take the test and I needed them to be ready for it).

I think all of this boils down to get more data into Canvas (or your LMS if you can)…try not to rely on a single demonstration to judge understanding. My goal was to have students show mastery on standards by the end of the semester. So, if they’re not getting one of them now, it still goes in as a zero but it serves as a reminder that they still have to do that standard. I was updating grades on the last day of the semester for my students. It’s a weird way for them to think and it’ll take some prodding by you so they don’t forget that a zero can always convert to full credit. Usually what happens is a later unit will give them more context for whatever they’re struggling with and cycling back after more scaffolding is more effective than trying to drill the issue immediately, if that makes sense.

If you’re not using Canvas, there may be similar systems in your LMS that will help you track growth. I also have a Google Sheet template that you can use to track student growth. Shoot me an email if you’d like that and I’d be happy to send it along.

If We Don’t Allow A Redo, What Are We Teaching? – The Teacher And The Admins

Whatever the reason, he was afforded the opportunity to learn and apply. It hasn’t come easy, but that’s the point. Giving a chance to redo isn’t about being easy.

Source: If We Don’t Allow A Redo, What Are We Teaching? – The Teacher And The Admins

Still in my standards-based grading vein, this is inevitably the biggest sticking point for teachers I work with.

“What do you mean they can retake the test for full credit?”

Mentally, we can agree with the argument that redoing work or retaking tests makes sense in the scheme of student learning. The hold up, I find, is more with the work involved in making those opportunities reliable and valid more than the mental exercise of finding value in the habit.

There are ways to allow students to reassess work that does _not_ include sitting an exam again, which opens more possibilities for authentic learning and demonstration of mastery.

Fair Isn’t Always Equal: Three Grading Practices to Avoid

Remember, though, that grades should not be used as rewards. Nor should they be used as affirmation, compensation, or validation. Grades should represent an honest report of evidence at this moment in time, nothing more. If we make them something more than that, we undermine the student’s maturation and any useful purpose for grading.

Source: Fair Isn’t Always Equal: Three Grading Practices to Avoid

I’m working with several teachers on moving toward standards-based grading and we’re starting to have conversations about grades themselves. I ask how they feel about zeroes, extra credit, completion, and makeup work. This article is a great primer and/or followup to those initial meetings.