Thursday, October 20, 2011

Helping Students Give Feedback

Recently, Frank posted a picture of how his students correct their own tests and John posted a sample of the feedback his own students give themselves. This provoked a lot of discussion among my online colleagues and one thing that came up was how it is difficult to get students to leave good feedback.

I noted in the comments of John's post that I found it really interesting how his students were mainly leaving reminder notes and questions for him to answer. In this case, I'm talking about feedback in a traditional teacher-directed sense.

This is something I've really been focusing on this year. It's definitely been rough. I attribute this to the lack of quality feedback students normally get from us but that's another story. I wish I was friends with more English teachers because this is their bread and butter.

The whole thing is pretty standard but as usual, I buried my big insight. It's Key Point #2 if you want to skip ahead.

First, I narrowed the scope. I tried to teach them to leave feedback for only one type of question. In the topic on Atoms, we focused on the explanation questions—how you can explain different phenomena through the motion of atoms.

Second, I went through the process of how I'd evaluate those question types and wrote out a flow chart. It looks like this but you can certainly just do it as a series of questions or a checklist.


I drew it up on the board but you get it nice and typed. For the science folks, at the time we used the words atoms, molecules, and particles interchangeably.

Students are supposed to go through the flow chart and then write their feedback based on the chart. I gave them a sentence frame where each step they pass is a positive comment and when they hit a "no" they step back and write the improvement step.

For example, if you got to the second "no" you'd write something like, "You wrote an explanation for why water boils when heated. Next time your answer needs to mention atoms, molecules, or particles." or "You wrote an explanation for why water boils when heated and included atoms. Next time your answer should include how atoms move."

After that it was pretty standard. We used some generic sample answers to try as a class. Then they wrote peer/self feedback on some of their previous answers. I had them predict some things they hadn't seen yet, like sticking a balloon in a freezer and they traded and wrote peer feedback as well. Each time we wrote a set of explanation questions, we'd do this for at least one of the questions.

After a few tries at this I added another step in the flow chart at the end, "Does the explanation connect the motion of the molecules back to the observation or prediction?" This was much more difficult for kids to get than the first three but it is also a much harder skill.

Two key points:

  1. My real goal here is that they do this enough times and when they're writing their own explanations they have the flow chart running through their head. Is this an explanation? Did I talk about atoms? Did I talk about how atoms move? Did I explain how that's related to what I observed/predicted?
  2. What's really important is what's missing. I didn't have them leave feedback for correctness. This is where kids usually get hung up on feedback. If I don't know the answer myself, how can I give that kind of feedback? I can't tell where you went wrong if I don't know what the right move is. What I want them to look at is the quality of the explanation itself. Every kid can look at a written explanation and decide if it has certain qualities. It is important for a student to understand that he or she can be factually incorrect but can still give a quality explanation and vice versa. These are two different skills and we are going to improve both of these. This is the part I feel like I got right. Student feedback can't depend on the level of content knowledge. 

I'm giving this a tentative endorsement. The higher level feedback is still up to me but this has definitely helped move the lower and middle level responses up a notch.

One teacher-bonus I noticed from doing this is how dumb it is when I ask a kid to, "Check your answers." Yes, sometimes there are careless errors they can catch. Mostly, if they didn't know it when they answered the first time, they still don't know it. (Actual quote: "I checked it. I still don't get it.") I really need to do a better job of teaching my students different methods for verifying an answer.


  1. Thanks for this -- I agree about helping students evaluate the quality, rather than the correctness, of an explanation. This semester I've really focused on this, using "clarity, precision, lack of contradictions, and connections to other things we know." Those are getting a lot better. But students are still having a very hard time with cause and effect. They either don't know where to start, or beg the question, or talk themselves in circles, seeming to conclude that everything causes everything else. I'm at a loss as to how to help them evaluate their own answers. What criteria do you use to distinguish explanation from observation?

    Incidentally, Joss wrote about feedback here, giving strong and weak examples that I found helpful.

    And I'm guessing you've already seen Brian F.'s comment about "clarity, consistency, seamlessness, and coherence" (but if you haven't, check out the last comment on this page).

  2. Here you go questioning assumptions about what students do and don't know again.

    I've been chatting with an elementary education professor about reflection a lot lately (we're presenting at ASCD together on it). Teachers who don't reflect on what is happening in their classrooms, on their students, on the choices they make, etc. don't improve as teachers. They never question assumptions and therefore continue to do things the same way again and again. How do we help teachers to become more reflective? Do you think you are just naturally reflective or did you develop this in some way?

  3. @shiftingphases - I don't have good advice here other than what I said before. If you're teaching a difficult skill and combining it with difficult content, that's too much. If you need a buzzword, I'm going to go with cognitive overload. What I do struggle with though is easing it in. So we'll start w/ something familiar where students have a lot of background knowledge and practice the skill.....then I jump to all new knowledge. There has to be a transition period but I start to feel the pressures of the school year flying by.

    For explanation vs. observation we usually go with key words, like "because". For this specific topic its easier because I can stress that if they're only writing about stuff they can see directly, it's probably an explanation. Any explanation involving atoms is going to necessarily involve something not directly observable. I have them draw a picture as well "using the model." A lot of times its the language itself that trips up my students and drawing it out helps them solidify their thoughts.

    @Jenny - Thanks. Speaking for only myself, I know I've never felt like I was a natural teacher. A lot of stuff that I see new teachers pick up on instinctively goes over my head. To compensate, I need to be very intentional in the things I do. As for others? I figure everyone is different. I do think most of us want to work to improve but either get bogged down trying to stay afloat or don't know how to improve or don't know that improving is even possible.

    I'm still in love with that Positive Deviance idea which basically comes down to peer observation/learning. It's inspiring to see people in your same situation who are really excelling.

  4. I'll just add my name to the "thanks for this" camp. I've been looking for ways to make my students more independent in SBG, but some still can't figure out where they went wrong with anything. using a decision tree like this would add a lot more transparency to the process without much work from me, and this transparency I think will enable them for more. So thanks!

  5. When teachers only ask students to "check their answers," the focus seems to be on grading, as opposed to reflection and feedback for improvement. Your flowchart is a superb example of an opportunity for the latter!

  6. Your post helps illustrate the importance of academic vocabulary, too - e.g., the difference between observing and explaining. I haven't done that as much as I should with my tenth graders, but when I taught a lower level ninth grade class our whole first semester of vocabulary was academic language rather than random words plucked from short stories or novels.