What’s in a grade? This is a question that I ask myself almost everyday, if not actually everyday. Recently, my learning partner, Synergy co-teacher, PLC co-director, and fellow faculty member – all one person…Jill Gough – posted this important system of inquiries about grades. The original post can be found at her primary blog Experiments in Learning by Doing, but I have also pasted the post below.

At countless learning opportunities, we have listened to, read from, collaborated with, and contemplated among the “grading and assessment experts” about the issue(s) raised in Ms. Gough’s post. We need your help. Please read through the story-inquiry, which is all very real, and let the comments flow. Hopefully, we can engage this discussion and learn with each other – for the benefit of all of our learners, both young and old.

I want a culture or climate of revision and redemption for my learners. The first step is second chance testing. My learners have the opportunity to learn from their mistakes and redeem points from their struggle to learn. I am comfortable and confident in this approach to testing even when my learners’ parents struggle to understand this break from tradition. For many learners, we see immediate results in improved test scores.

I am now faced with a new struggle concerning my grading practices. After reviewing a couple of examples, I am hoping that you will comment below to give me your opinions and thoughts to help me move forward. Please? Just press post comment. It will be okay.

MC is a hard working Algebra I student. She regularly attends Office Hours with her team to work on her homework and check her understanding. She really struggled with exponential growth and decay, but she stuck with it and learned. Let’s look at her test scores:

Exponential Functions: 88

Polynomial Functions: 92

Cumulative Midterm: 96What grade or test average would you assign this learner? Why?

FH is a great learner in class. He makes great eye-contact; he listens and asks great questions. His work outside of class is average. He really struggled with the algebra of polynomials. Let’s look at his test scores:

Exponential Functions: 72

Polynomial Functions: 52

Cumulative Midterm: 83What grade or test average would you assign this learner? Why?

Is the average of these three grades an accurate reporting of what has been learned. Didn’t the struggle to learn more about polynomials cause this learner to continue to improve? Should he be held to that 52 when it may have helped him learn? It the spirit of revision and redemption, how can we accurately represent – with one number – what he has learned. If time is the variable and learning is the constant, what do I do with this data? How do I make an accurate report?

PK shows up and does the daily work. She regularly attends Office Hours with MS’s team to work on her homework and check her understanding. She really struggles to put it all together. She is great when the learning is compartmentalized, but when given choices, she struggles to know what to do when. Let’s look at her test scores:

Exponential Functions: 90

Polynomial Functions: 83

Cumulative Midterm: 80What grade or test average would you assign this learner? Why?

Is there only one algorithm for computing the summary grade? Are the conditions where the algorithm could/should change to represent what is learned?

I want a culture or climate of revision and redemption for my learners. But there are deadlines, right? Should a learner be held accountable for work in January that caused them to struggle and learn? How could/should these scores be weighted?

How can one number communicate and summarize to a learner, a parent, and a future teacher what these children know and are able to do?

These are the paradoxes that I struggle with most. Grades are an invention to help us manage a great deal of information, right? On the administration side, there’s no way to understand where your school is without some kind of summarizing, and with more than a few learners, the summaries have to be short and use comparable vocabulary – either the 0-4 GPA scale or the 0-100 scale, usually. On the student side, anyone working towards more than one goal needs summaries to keep track of the feedback he or she gets. If I’m in 5 classes which each focus on 30 different things over the course of a year, I can’t keep everything in mind – I need succinct reminders.

But, desperately, we try to avoid trivializing our students’ education, and what can a single number be but trivial in the face of a year of life spent in school?

I’m developing a software solution called ActiveGrade that I think can help by delaying the dreaded averaging moment until the last possible moment. It presents students, teachers, administrators, and parents with specific details instead of vague, averaged grades, but retains that succinct, comparable quality that we need grades for in the first place. The paradox still exists – how do you combine these scores without losing the important information they hold – but we have some exciting new ideas that go a long way towards handling it. We think averaging is inappropriate for most sets of scores, so we provide decayed averaging and a power model (http://activegrade.com/blog/?p=34). We also think trying to find a formula for your final, overall grade is too hard, so we invented what we call “Natural Grade Calculation” (http://activegrade.com/blog/?p=58).

I hate to turn this comment into a huge ActiveGrade plug, but these are exactly the questions that inspired us to start working on ActiveGrade, and we have a little community of teachers and admins that are forming around it. I think we’re doing a good job of supporting teachers who have these questions. Thanks for the post – it’s re-inspired me!

Thank you Riley. I appreciate that you took the time to post a comment here and on my blog. Two different comments. Wow.

I am very interested in ActiveGrade, but I don’t have the time before my next deadline to enter in all of my data.

I’m curious why you chose a power function to model the decay. I’ve read Dr. Marzano’s work, I know that he uses a power function too. When I’ve played around with the data, I find that a logistic function usually fits as well or better than the generated power function. I can send you some of the data that I’ve been playing with if you want to see it.

I have to make a decision about what to do, at least temporarily, in the next 24 hours. My next round of grades are due Monday at 3:00.

Thanks for thinking with me and thanks for the encouragement about SBG.

The power model has a lot of flaws, and I actually prefer a decaying average – an average that weights more recent assignments much more heavily than earlier assignments. My feelings about this are very strong and can only be expressed in a lot of words, but basically I think initial scores aren’t very important except as indicators when we’re analyzing trends. Decaying average and a power regression both use initial points as pattern setters only. I haven’t looked at a logarithmic regression, but I’m basically fine with any regression that weights the most recent scores more heavily than earlier scores.

I am absolutely *NOT* ok with specifically targeting high and low scores (e.g. dropping the lowest score or taking the top 5 scores from a student’s data), but that’s another blog post 😉

What?? You don’t have the time to make the changes in your grading system before the deadline to enter your grades??

I suggest you have a conversation with your principal and explain why you need more time. I am sure he is like many of the principals and others who post here and would love to have that conversation so you can be successful as an instructor. I am sure he won’t mind calling his central office and asking them to keep the grade entry portal open a few more days.

What nonsense.

Grades should be more subjective and represent both the tests and performance/participation

Plus the system should be standard from teacher to teacher within the subject and respective grade

I think the problem here is that you’re trying to measure an unmeasurable factor. The generally accepted solution is to introduce a subjective “class participation” grade, or some such. The real problem is that teachers can’t grade an individual student. They have to grade and teach whole classes. So that subjective grade just becomes “teacher’s choice”. Either they give it to a few teacher’s pets, average for those who kept their heads down, and low marks for the troublemakers OR they just give everyone an A on that one. The other solution I’ve seen is to calculate attendance into the grade, as if a lack of attendance wouldn’t be reflected in the rest of their marks (until you get to college).

I’m afraid this problem doesn’t have a solution. Grades can’t be subjective for a long list of reasons. This is an unfortunate byproduct of our industrialized learning system. The only real solution is to change the paradigm drastically. I’ve got my own developing idea of what that would look like and I hope to one day make it a reality.

While it’s true that we could spend a great deal of time trying to sort out these questions about the implementation of grades and homework, we might be better off asking why we are assigning grades and homework in the first place.

Second chance testing is a great way of giving the system what it wants without giving anything the kids actually need.

I’m bothered by this post in more than a few ways, but perhaps what’s most disturbing is that this post concerns itself more with grading than learning.

Well Joe, I have to sat that I’m a little surprised by your comment. I realize that it lacks context that you might need. I focus on learning with and for my kids daily, hourly, all the time. The easy way is to teach, test, average and move on; I reject the easy way.

You say “Second chance testing is a great way of giving the system what it wants without giving anything the kids actually need.”

The traditional system, in my opinion, does not understand or want second chance tests. It really bothers many of my colleagues that my learners get to try again. I believe I am giving my learners something that they actually need. We work on learning, we stop and check. If they are not where they want or need to be and if they don’t understand as well as they want, we will dig in. They can try again after more coaching and additional time. I am not holding everyone to the same pace; I am offering them time to learn. Coaching customized to their strengths and struggles.

It is about the learning, not a schedule.

However, to stay employed, to have the opportunity to help kids learn, to work on changing the system, I have to measure and assign grades a minimum of 4 times a year.

How will the system change if we don’t discuss alternate ways to measure/report learning?

Are you recommending that I not spend time grappling with how to best represent learning with the grade I am required to assign?

Wow. I’m not willing to do that.

Two things came to mind as I read this: 1) use weighted grades so that the cumulative midterm receives a weight to reflect its worth 2) the grade debate deserves more in-depth discussion from educators. Some years ago I read Robert Marzano’s “Transforming Classroom Grading.” I was skeptical as I began the book, but I was very impressed by the time I was done with it.

Jill, I think your post does a really good job of laying out some tough questions about grading and how to represent a student’s level of understanding. I disagree with Joe that your post focuses on grading instead of learning. In everything, we learn by trying, making mistakes, getting feedback (either from others or from observing the outcome of our mistakes), and trying again. Grading is ultimately feedback and therefor an essential step in the learning process. What matters is how we treat grading, how we use it to help the student reflect and keep going. Sadly, for many, it has become the focus, the thing that matters. It is clear from your post that you are focused on how best to help students learn and not punish them for that learning with a grade that does not best represent what they have accomplished.

I am trying to think about the questions you’ve raised and I have so many thoughts in my head. In the second example, of the student who got a low polynomial grade, you ask if he should be held to that 52. I think it’s helpful to think of skills on a scale of mastery. If he got a 52, what does that indicate about his level of understanding. A 52% is quite low in one regard. On the other hand, if he understood about 1/2 of the questions, he knows and understands quite a bit more than nothing. He’s clearly learned *something.* He could maybe even be considered more than a novice. Once we frame it that way, we have to decide how much a student *needs* to learn in order to be ready to move to the next class. If 100 means total mastery, could do it with my eyes closed and hidden inside a hugely complicated problem with many steps, then that’s probably more than each student needs to understand. And it’s possible that that level is different for each skill. Maybe they need to have a very firm grasp of exponents but just need some basic understanding of polynomials, in which case a 52 might be enough.

To sum up – 1) good job! – I think you are raising excellent questions and 2) I think reframing the way we think about the scale and considering what level of mastery is actually necessary can help us give more accurate feedback (and grades).

Michal, There’s a new development from the learner that you focused on. We have taken another test. On the first chance test, with no partial credit, he scored an 86, well above his first chance test average. This most recent test continues working with quadratic functions and solving quadratic equations. He has worked hard to overcome the 52. This latest test shows his effort is paying off. You should see the improvement in his written work. Finally, after hours of work and coaching, he has begun to document his work to record this thought process. His written work has improved dramatically. Is it because of the experiences? It is because he is growing up? Is it because he is tired of working with me after school? Probably some of all of the above.

I think I can more confidently report a number for this child. But what do I do for others that don’t fall into his category of “the struggle caused him to learn?” What about the child that can handle one topic on a test, but struggles to put it all together on the cumulative tests?

Again, I’m opening myself up for criticism from Joe by talking about grades. It is really about their learning. If a child’s scores are 81, 78, and 60, then I need to intervene and help. They need support. They are not retaining the information and skills. It appears to me that it was put in temporary memory rather than learned/mastered.

I agree that reframing is needed…and soon! We can complain about it forever, but when are we going to try something? I’m not really asking for a solution, though that would be great; I’m looking for a next step.

I appreciate your thoughtful response. Thanks for thinking with me, and thanks for using my learner’s data to continue the conversation.

Here’s my example of my most successful grading and evaluation process. Within it is a strong assessment process as well. for those of you still using assessment and evaluation in one sentence, please stop.

Here’s what I did a few years ago with my grade 8 math classes:

1. Lnk the curriculum items across the strands of math. For example, I taught evaluating algebraic expressions, then area of a recatangle, followed by factoring numbers (we used rectangles to help understand factors). thee concepts build on each other and teach kids that math is not compartmentalized into algebra, geometry, and numbers. Look up “Math 44” for details.

2. Stop having unit tests – I had weekly and monthly quizzes coving the topics of the week or month. No UNIT TESTS!

3. Create a portfolio-based method of assessing learning. For each learning outcome, students had to demonstrate their understanding in at least 2 ways – including one test.

4. Students assess their own portfolios and their peer’s portfolios

5. Record student progress in a spreadsheet, but do not calculate averages

6. At the end of the term, meet with each student and have them provide a rationale for the grade they think they deserve on each learning outcome (1-4 scale) and for the term (letter grade). If I idsaggree, I make them prove their understanding in a third way.

7. Finally, assign a “work habit” grade (required in my province) based on their effort in class and completion of required work. Do not confuse effort with learning.

I have examples of these forms, just ask and I’ll send them your way.

Hi Kyle…Thanks so much for your comment! My team has decided that we will not give unit tests next year; we will offer regularly scheduled assessments that spiral the learning that should be mastered.

I’ve also been working on a “negotiation” plan where my learners can build a body of evidence (portfolio) to persuade me and work to collaboratively determine their grade.

I would absolutely love to see any forms that you have used. You can send them to me at jillg@westminster.net.

My principal, who made the post, is working with us to develop a balanced assessment plan where our learners have multiple ways and multiple opportunities to demonstrate their learning. Our assessment study group and PLCs are focused on assessment for learning.

We are moving to 1:1 next year in our JH. I’m hoping that my learners can chart their own progress in a spreadsheet shared with our learning team. I’m a big believer in charting your own growth and sharing the information with your learning team.

Thanks! I can’t wait to see more of your work and thinking.

The argument about whether to grade or not and how to grade has gotten more and more interest over the last few years and I’m glad it has. As a math teacher for nearly 20 years I have struggled with examples (every year) like you have given here. It is so easy to give a grade to a hard working, always scoring nearly 100%, likable student. It is also fairly easy to give a grade to a kid who is never there, disrupts class, and scores below 50% on everything. The rest of the 80% of the class can (and do)perform all over the charts.

Our school is a New Technology Network school. We are 100% Project Based and we assess in learning outcomes based upon 21st century skills (in addition to the content grade) and, based upon weighting, a grade is calculated. Some of the learning outcomes assessed include: Written and Oral Communication, Collaboration, Work Ethic, Numeracy, and Global & Community Engagement. Each of these areas are graded upon a rubric created for each project. So, in the current project my students are working on there is no grade for Written Communication but there is a grade for all of the other learning outcomes.

The key to grading with this system is understanding the weighting. It is very easy (if you are not paying attention) to get a high (above 80%) grade in Algebra, for example, if the weighting of the learning outcomes besides the Algebra are too high. Just picture that really hard worker who is willing to help the other students but isn’t necessarily the best math student in the class. He speaks well and writes well and gets above 90% in all of the learning outcomes. But, his quizzes, tests, and the algebra part of his presentation are not sound and he has demonstrated that he is not knowledgeable with the content. It would be bad to send him on to the next teacher with a grade above 80%.

And, therein lies the biggest stumbling block to changing the current 0 – 100 scale, as I see it. Teachers, parents, higher education, and the community understand a grade of 83. Anything else starts to set in motion a sense of doubt that content has been covered and that the student has mastered (in this case) about 83% of the material. But, I think we need to continue this discussion because I think if there are enough brave schools, school districts, and states then the change could happen and I would find that exciting.