Assessment for Learning

I just finished Jo Boaler’s What’s Math Got To Do With It?, and I am particularly intrigued by “assessment for learning.” Boaler says that assessment for learning doubles the speed at which students learn.

I have been trying to learn more about this, and here is what I have so far. It appears to be a lot of formative assessment, but in a very particular way:

  1. You tell students exactly what they will be learning.
  2. You give the students examples of strong and weak work.
  3. You and the students determine how the students can self-assess their work.
  4. You design lessons that focus on one learning goal.
  5. You give the students a ton of descriptive feedback (no scores or grades), and the students do a lot of self-assessing.
  6. Students revise their work.
  7. Students track what they know and what they have yet to learn.

I got this off of an internet search. I also ordered Assessment for Learning: Putting it into Practice through interlibrary loan. I hope that this book will put me in a better position to understand what assessment for learning is, because it is all a bit hazy now.

I am hoping that this will still be as great as it sounds (doubles!) after I know more about it. If so, I am hoping to integrate it heavily into my real analysis courses in the fall.

Does anyone know more about assessment for learning?


17 Responses to “Assessment for Learning”

  1. Patrick Bahls Says:

    Although I’ve never called it by this name, what you describe is pretty much the way I run almost all of my courses these days. I provide frequent models of students’ work in all of my classes; in most classes they serve on “homework committees” that are tasked with responding to peers’ work on various problems; in most classes students have virtually unlimited opportunity to revise their work. They are more than anything else meaning-makers and co-discoverers. They’re not so much students as they are self-correcting interpreters of disciplinary ideas. It’s a rich experience for me, as well as (I hope) for them!

    • bretbenesh Says:

      Hi Patrick,

      I was thinking of you when I was finding out about this. The fact that this particular method has a lot of research behind it may have removed my last barrier to trying it. I just need to figure out if I can already start incorporating it this semester (the answer is “yes”).

      I also thought that Assessment for Learning could work really well with portfolio grading. I haven’t worked out the details yet (since I don’t fully know what Assessment for Learning is yet), but it seems doable.

      Question: How would you allocate your classroom time during the week? How much new instruction/lecture, how much “assessment,” how much problem solving?

  2. Andy Rundquist Says:

    I’m very interested in this too. I have some experience with points 1, 3, sort-of 5, and 6. I don’t really keep the ultimate grade out of it, though, so that’s a big difference.

    I think point 2 is really interesting. In my class-wide oral assessments we all discuss what rubric score someone deserves. In a sense they see examples of good and bad work.

    I wonder what my friend Joss (@jossives) thinks about point 4 (single outcome lesson). He’s been pushing me to make sure I think about synthesis with my students and my various learning outcomes. It has caused me to really think about the role that single-outcome lessons play.

    Thanks for the great post! -@arundquist

    • bretbenesh Says:

      Hi Andy,

      This is one more thing that I am going to have to pick from your brain later this month. I have done 6 well, 3 and 4 not at all, and the rest in a mediocre fashion. I have not tied it all together, though.

      One of my main hesitancies about SBG was that it might be a barrier to synthesis. I only decided to try the idea when I figured out that I could have my “standards” be “topics” rather than “skills.” This gives more room for synthesis, although I could certainly stand to do this better.

      I just subscribed to Joss’s weblog—I was wondering if you knew him.

      • Joss Ives Says:

        Gents, you raised the bat-signal and now I’m here.

        It’s true that I’ve been bugging Andy to think about synthesis. But the rub is that, beyond giving them meaty exam problems that draw from a bunch of different parts of the course, I don’t think that I am doing even a half-decent job of assessing or bringing focus to synthesis. Many of the SBG implementations that I’ve been looking through focus on narrow skills and not so much on the interplay between the skills or (as far as my novice eyes could tell) the problem-solving/decision-making skills where the student has to choose between a bunch of tools in their toolbox and decide which ones are most appropriate for the task at hand. Frank Noschese helped me see how to assess the problem-solving skills in the comments on one of my posts. Jason Buell’s post that is linked to below suggested another way to assess synthesis and that is to make the standards much more coarsely grained so that higher proficiency on that standard actually assesses synthesis directly. I agree with Bret that I would like to see some concrete examples from Jason.

        In terms of #4 “You design lessons that focus on one learning goal”, I think that is usually the most common place to start (assuming narrow learning goals/standards). But I certainly wouldn’t want to restrict myself to focusing on only one learning goal in that lesson. My standard lecture consists mostly of clicker-questions and whiteboarded problems with some scaffolding. The clicker-questions are more often than not in sequences where questions based on the same concept are asked with increasing difficulty. Often the most challenging ones bring in previous concepts (learning goals), so this is one of the places where some sort of course synthesis occurs and for that synthesis to happen, these previous learning goals need to be brought back in. Similar stuff happens with the whiteboarding where the more challenging problems often ask for lots of concepts to be brought together, but the concept that is front-and-center is the one that is the main learning goal for that lesson (lesson=some portion of 80 minute lecture). Bret, once you get a firmer grasp on what “Assessment for Learning” is, you can share exactly what is meant by the one learning goal per lesson in terms of calling on previous content.

        Bret, I also have previously subscribed to your blog, but my RSS reader is so overloaded I just finally got around to catching up with your recent posts.

      • bretbenesh Says:

        Hi Joss,

        Thanks for commenting! And don’t worry about a loaded RSS reader—that is the way things go sometimes.

        It is interesting that you mention Buell’s post. I only learned about SBG late last summer, and I shared the same concerns about synthesis that you did at the time. But Jason’s convinced me that it is possible to pull off.

        Finally, having learned more about Assessment FOR Learning, it seems like “single topic lesson” is not a major point of it. In fact, I don’t remember reading anything about single-topic lessons in the book I read over spring break. Let’s not consider that to be a part of Assessment FOR Learning right now.

        Thanks for posting!

      • Joss Ives Says:

        I would love to see synthesis assessments in the wild so I have something concrete to hang my hat on when it comes to trying to do that same sort of thing in my own eventual implementation. “Hey everybody, do all the hard work for me, let me know how it turns out, and then I might try it out eventually with lots of help from you” 🙂

        Interesting that the single learning goal per lesson didn’t turn up in the book.

      • bretbenesh Says:

        Hi Joss,

        When I come up with good synthesis assessments, I will try to post them. I think that I have some, but I will have to blog about it on a different day.

        “Hey everybody, do all the hard work for me, let me know how it turns out, and then I might try it out eventually with lots of help from you”

        Isn’t that quote basically a summary of the purpose of reading blogs? That is what you are SUPPOSED to be doing (and I will leach off of your hard work on different ideas).

  3. Siouxgeonz Says:

    This is hazy because it’s missing really important critical ingredients.

    This *could* be like saying “you’re going to learn this gymnastics move.” Then you show ’em somebody doing it well, and somebody doing it poorly. Then you say “how do you think you should be graded?”
    Pardon me, but if you let me figure out how I’m going to be graded, I’m going to do … oh, about twice as well. Or better. *I* think I should be graded on how well I land. Yea, assess me on that. Guess what? I’m going to look pretty good.
    Now, the giving the student some control over what they’re learnign is excellent… but the teacher had better have a good handle on what kinds of skills are important and do some steering along the way.
    OF course, if you dress this up a little differently, then what this is, in disguise, is that strictly controlled, isolated skills drilling — except the sequence of skills is determined by the students, who *don’t* really know the big picture. I *do* think that could be a powerful reality check for the teacher, who may be surprised at the skills the students want to be assessed on, and I *do* think student input is invaluable… but this is hazy because… it’s hazy.

    • bretbenesh Says:

      Do you have any idea on which ingredients it is missing?

      I definitely share your concerns about this being too reductionist/skills drilling. However, Jo Boaler is the opposite of a reductionist, and she thinks that it is pretty cool. This suggests to me that it is worth looking into. At the very least, this seems like it would be extremely useful in teaching students to write (which would not be simply teaching the students to mimic skills). Since I am trying to turn all of my classes into writing classes (since mathematicians are actually just a certain type of writer about certain types of ideas), this could be very useful.

      I also agree that the teacher has to play a strong role in shaping how the students should be assessed—I don’t think that anyone is advocating for the students to completely run the classroom. However, I have had terrific experiences letting students decide these types of things in the past. I am always nervous that the students will take the easy way out/not take things seriously/think the exercise is stupid, but I am almost always end up being pleased at how maturely the students handle the responsibility.

      So I am working on turning my “hazy” into “less hazy.” One way in which I am doing this is to reach out on this weblog to my PLN for help.

      • Erica Says:

        I too would think this would work pretty well with writing, but I’ve done it only a limited amount, and had a rather vivid example last semester of how bad students can be at assessing examples. I gave students examples of good, mediocre, and bad writing, and they assessed them all as bad until I pointed things out. I hope they learned from this, but I didn’t follow up systematically. In the future I’ll think about using this method more intentionally.

        On another note, I don’t think that self-assessment is the same as self-defining the learning goals. The point of using the examples is to show them that there’s a better and worse that can be defined, and getting them to internalize that so that they can self-monitor. That anchor prevents it from becoming students grading themselves according to their own criteria.

  4. bretbenesh Says:

    Hi Erica,

    I am hoping that mathematical writing will be easier than other writing because there is more of a “correctness” about the content. However, this is probably just a hope—there are still all sorts of issues about the appropriate amount of justification for any idea.

    And I like your point about “self-assessment” versus “self-defining” learning goals—so much more that I have nothing to add.

  5. Jason Buell Says:

    re: reductionist.

    Using the gymnastic example, think of it more like going through the whole routine, going to the video tape, and then practicing the part you need to work on the most (the single lesson/1 learning goal).

    The problem comes in when you try to do each individual move in sequential order instead of letting them do the whole thing and then assessing what to work on next.

    The semantic difference (of learning vs. for learning) is supposed to indicate that the assessment is used to drive future learning rather than just an evaluation of what has been learned.

    This requires that at some point, your students aren’t all working on the same thing at the same time.

  6. bretbenesh Says:

    Hi Jason,

    Your comment is a good reminder that we should be giving students meaty problems from the start (“having them work on a whole routine”), rather than just giving students a bunch of tiny exercises.

    Question: is this roughly what you do? Give them meaty problems and assess them according to which of the standards they demonstrate on the meaty problems? Or do you give them bite-sized exercises, too?

  7. Jason Buell Says:


    Ideally, yeah. I can’t always come up with anything really meaty for everything but when I can, that’s the goal. Then just have them do bite-size exercise on whatever they need help with to accomplish the meaty goal. Rinse, repeat. The idea is here:

    For some things it’s definitely separate skills though. At the MS level, we’re not doing anything with the periodic table beyond basic identification (where are the metals, etc) and protons/neutrons/electrons. That’s definitely pretty skill based. I’m sure there’s a way to turn that into something higher level but I haven’t figured it out yet. Even for that though, they have a Periodic Table grade and individual standards within that they they receive feedback on. So about once a week or every two weeks they’ll be working on whatever individual standards within that topic that they need to improve (some kids will be working on finding the number of protons, others on classification, etc).

  8. bretbenesh Says:

    Hi Jason,

    The “whole game” post you linked to was the last piece I needed to read before I could confidently go ahead with SBG.

    Question: which “whole game” are you working on now with your classes? I think that a concrete example could be helpful.

  9. SBF « Solvable by Radicals Says:

    […] fact, once we do this, we are awfully close to Assessment FOR Learning…stuff (I needed a fourth word for my fourth link. The proper term is Assessment […]

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: