Standards-Based Grading Traps

I’ve become pretty well sold on standards-based grading. This image from Frank Noscheese is a good summary of why I believe it’s the best assessment system for me:


While I have no plans to change my approach, I’ve felt more and more like SBG is really just the least bad out of my options for assessing — that I’m, as Shawn Cornally describes it, putting lipstick on a pig.

If you’d like to check out what I do, I wrote about it here. Short version: units are broken down by skills. Students get assessment scores from 0-4 based on understanding of individual skills. Unlimited reassessments (until the end of the semester), on student initiative, new grade replaces the old grade. There’s another 25% that tries to grade on transfer of skills, but most of my system is based on these assessments.

My particular issue recently is that I think standards-based grading can be deceptive — it can create certain traps that, if I’m not thinking carefully, perpetuate counterproductive views of student learning.

Formative Assessment 

Standards-based grading is not effective formative assessment. It attaches a grade to everything, washing out many (or, depending on your reading of the research, all) of the benefits of any feedback. While it does give me better information about what students know and don’t know, assessments don’t come until the end of that self-terminating skill. Standards-based grading doesn’t preclude the use of systematic formative assessment early on, but assessments take a great deal of time, and can make me think that they are serving as effective formative assessment when they’re not.

Understanding as a Binary Operation 

This happens all the time: student gets a 2 on a skill. I say to myself, “ok, student doesn’t understand X.” Student reassesses, gets a 4. I say to myself, “ok, now student understands X!” I feel great about myself.

Then I throw in a reassessment on that skill in two weeks. Student gets a 2 again.

I’ve been thinking of understanding less and less as a “they know it or they don’t” and more and more as a continuum of skill. Some students have developed more sophisticated skills in (arithmetic with rational expressions/solving related rates problems/graphing trig functions). Those students are likely to get 4s. Some students have much less sophisticated skills, and are unlikely to get 4s — unless they happened upon a key idea relevant to the assessment question recently (because they reviewed for the assessment) and managed to piece together enough to get it right. That student is more likely to forget, and forget quickly. A lack of longer-term, summative assessments means I can delude myself into thinking that all those 4s in the gradebook from two months ago represent my student’s current level of understanding, but learning is far more complex than that.

My Most Struggling Students

There’s an illusion that, because students can reassess, standards-based grading is particularly good for students who don’t do well on an assessment the first time. And it has the potential to do exactly that. But my experience is that, for a significant majority of students who really need it, I need to badger them to get them to reassess. And if I’m doing all of that legwork, what’s the point of a more informative grade anyway?

Instead, I see grade inflation for my students whose understanding probably deserves a high B or low A, but put in the effort to reassess until they get everything perfect. It creates a bimodal distribution, between students who insist on reassessing everything, and students who settle for lower grades.

Why Stick With It?

Well, standards-based grades more accurately reflect student learning, which helps me sleep at night. It also creates an incentive, whether students heed it or not, to get more practice in areas where students have a relative weakness. That’s not a bad thing.

That said, standards-based grading has the potential to give me excuses for a lack of sound pedagogy. I don’t give students an exit ticket, many of them do poorly on an assessment, and I say, “well, they should figure out what they don’t know and reassess.” I get positive results on an assessment and move on. Then when that skill is necessary to move on to a harder topic, students have forgotten, and I say, “well, I’ve already taught that, they should have remembered it.” I have a student who is failing, and I say, “well, they can reassess and improve their grade if they put the effort in.” In reality, I need to be deliberate about continual formative assessment, ongoing cumulative review, and targeted interventions for students who need extra help. Standards-based grading isn’t a substitute for this pedagogy. It’s just an assessment system, an assessment system that does a slightly better job than my current alternatives.

11 thoughts on “Standards-Based Grading Traps

  1. Carolina Vila

    Hi there, thanks for writing this. I’m curious if you create new assessments for each “redo” and how many you create per skill. I’m finding it difficult to keep up with the time for this in my own classroom. Thanks for all you do!

    1. dkane47 Post author

      I do create new assessments each time. It’s work intensive, at least the first time through, but I try and do them in bulk, which I find more efficient, and save them to make printing and copying more streamlined. I ended up with 7 different assessments for inverse trig functions this semester, which was a challenge, but it’s usually 3-4.

      I think there’s no way around the work necessary to make reassessments work, but I also find it one of the most important parts of the system, as long as I follow through and get students to reassess.

  2. Andy "SuperFly" Rundquist

    Wow, does this post hit me in my stride right now. I’m coming up on the end of a semester of SBG (been doing it for 4 years) and all of these points are just what I’m thinking about right now. While reading I had a crazy thought for a future experiment and I wonder what you think: Have all assessments during the semester be trumped by a final video portfolio where the student has to work some problems and explain the connections among all the standards (this semester there were 20 the way I do things). What I envision is the day-to-day would be the same (assessments initiated by both students and by me), but the hope would be they’d collect that feedback to help them with their final project (the crazy idea in the last sentence). It’s hare-brained for now, and hence destined for my blog in a while, but I’d love your thoughts. Thanks for this great post!

    1. dkane47 Post author

      That would be a fascinating experiment. I’ve considered a portfolio before based around the standards for mathematical practice, but decided it was too fuzzy to be practicable. I like the idea of connecting end-of-course reflection to specific standards. It combines spaced, interleaved practice with reflective elaboration, which are both things I think are really valuable to learning.

      I wonder what challenges this would face. I feel like if I did something like it, I would learn some unpleasant things about how little my students actually know. I am always a bit lost for what to do at the end of the year. Things to keep in mind!

    2. Matt Engle

      I just started grading by standard this year (in just one class to work out the kinks first) and Dylan hit on all the current feelings I have about it so far in his post. I’m quite interested in your video portfolio idea! On top of the change in assessment system, I incorporated a project very related to this but more just a written version, one page per standard. It is still very isolated by standard, however, and I really like your idea of having students explain connections among all the standards! Pulling that off would really drive home the connection-making in mathematics that I so often feel my students don’t get enough of on a day-to-day basis. Have you had the chance to put any more thought into this idea? I’m intrigued…

  3. Brett Gilland

    One small change that I have found addresses a few of your concerns is to include old material into new tests. I tend to aim for 50-50 splits of old versus new, but typically am closer to 2/3 new and 1/3 old. Still, that keeps older stuff relevant and helps reiterate that one shot quizzes are not enough to establish mastery. It doesn’t solve all your concerns, but definitely helps a few.

    1. dkane47 Post author

      I like that strategy, though I’ve found it tough to implement into a grading system that is transparent to students. I’ve thought about a separate category for spaced practice of old material, in place of all the work I’ve found is necessary to continually revisit old content and try to grade based on the original standards. That approach doesn’t value much of what SBG stands for, but is well supported by the relevant cognitive science research and easier to implement.

      I think that’s a good example of pedagogy that transcends assessment philosophy — it’s a good idea no matter what your assessment approach is, and assessment is not a replacement for sound pedagogy.

  4. Matt Engle

    Dylan, have you read Henri Picciotto’s system that he’s put up on his website? I am seriously considering incorporating what he calls “lagging” into homework and tests, as well as what Brett mentioned above with incorporating old material onto new tests. One of his mantras that I like is “constant forward motion and eternal review.” Knowing my students, it seems that if they knew each test would also contain material from old skills they would be more inclined to stay with it and reassess earlier. But I agree that scores for those problems shouldn’t spill over and affect the score of the current skill. I wonder if maybe it could just allow students to improve previous skills while not affecting new ones. And if they don’t get it correct they are still where they were before. Whatever it is, I’m still toying around in my mind about how best to grade by standard and still accomplish some of those goals.

    1. dkane47 Post author

      I’ve experimented with some of those models, including one from Dan Meyer where students need to get to consecutive perfect scores in order to get the highest grade possible. His ideas are here:

      At the same time, I want to maintain some of my skepticism that Standards Based Grading can be the solution for all of the challenges in teaching math. Much of my lagging happens in homework assignments, and I like that approach because it doesn’t require as much work in terms of grading, lowers the stakes, and allows me to act immediately to respond to misconceptions when we go over homework.

      I want my SBG system to accurately reflect what students know, and point usefully to what they don’t. At the same time, I would rather have an imperfect SBG system that doesn’t take much class time so I can focus on purposeful practice and formative assessment, and use those tools to address some of these challenges, than try to build a perfect SBG system that will inevitably suck up more time and energy from my class.

      1. Matt Engle

        Your last point totally echoes my current sentiment, especially with the large class load I’ve got. I was just thinking today about the right balance between a “perfect” standards based system (will it ever be?) and actually formulating purposeful practice and such.

        By the way, I finally got my own blog up and am trying to figure out how to get going on Twitter with the MTBoS. Your 10 minute short sticks in my mind about that, it’s what kicked it off! Check it out if you’ve got time; only two posts up yet.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s