The research is clear: good feedback is essential to learning at high levels. Alas, too few people understand what feedback is and isn’t.
Don’t believe me? Here’s a true/false quiz: which of the following 4 statements is feedback?
“Nice job on the project, Sheshona!”
“Next time, Sam, you’ll want to make your thesis clearer to the reader”
“The lesson would be more effective, Shana, if your visuals were more polished and supportive of the teaching.”
“You taught about ants, Stefan? I LOVE ants!”
This was a bit of a trick question. None of these statements is feedback. The first and fourth merely express a personal liking for something, separate from a performance goal. They are not feedback since there is no descriptive information about the performance nor is the implied goal of pleasing the person the right goal.
The middle two are not feedback; they are advice (which is different). Yet, I am sure that many readers likely thought at least one of those two statements was true.
In writing about and running workshops on feedback for over 25 years, [I find*] this mistake comes up over and over again. And it has unfortunate consequences. It leads teachers and supervisors to be unwittingly bossy instead of truly helpful; it leads to performers being passive and dependent upon authority: “Is this right? Is this what you’re looking for?” become the constant questions, alas.
On the whole people need less advice and more feedback. Or, if advice is needed, it should be given after feedback is given and understood; rarely first.
What feedback is and isn’t. Feedback is useful information about the effects of an action in light of a goal. If I tell a joke, the goal is laughter; the “effect” is whether or not people laugh and to what extent. The specific kind and amount of laughter is the feedback. Advice might follow from a veteran comic to a joke that fell flat: “You need to work on your timing.” But unless I have clear feedback related to timing the advice is not very helpful.
Similarly in music. In playing Chopin’s Etude #4 on the piano, I can hear the mistakes as I play. I can take note of when I make those mistakes and work on those parts. I don’t need a coach to give me advice (unless there is tricky fingering or time signatures involved that I don’t understand.) I need only to attend to feedback and learn from it.
Thus, the young comic or musician may not need much advice. They almost certainly need less advice than teachers are prone to give. Lots of attention to when people laugh and when they don’t; careful note of mistakes on the piano can often lead to great improvement as the performer learns what works and what doesn’t – learns from the feedback. The advice is then (ideally) self-given.
Note that in both cases the aim is crystal-clear: make them laugh; as a relative novice, play the piece accurately. It is only when I am clear on my purpose as a performer that I can seek and use feedback effectively. Otherwise, I wait timidly to learn from an authority how I did without knowing myself.
In education, we should aim long-term for a similar system. The feedback should be as clear as the joke-telling and music, and linked to self-conscious purposes, so that the performer – be it a student or teacher – can learn from the results.
That’s the ideal; why does it rarely happen?
There are five reasons why students and teachers get too little feedback and why the feedback is often unhelpful:
- Most so-called feedback is really advice or praise (as in the four examples above)
- The feedback is not clear and descriptive enough about what did and didn’t happen as a result of some action taken to achieve a purpose. (e.g. a total score of 72 out of 100 on a math quiz is the feedback; it’s meaning for action is unclear.)
- The purpose of the task is so unclear (or non-existent) to the performer that the feedback is either random or mysterious. (Without a specific teacher goal for the observed lesson, feedback and advice are pointless.)
- The learner has not been provided with any exemplars of excellence against which to compare their work and thus obtain feedback. (Rubrics are NOT specific enough for the performer; they are inherently general. Models plus rubrics provide the basis for useful feedback).
- The feedback is too late. (Thanks to a commenter for reminding me to highlight this crucial issue, as I have done in earlier posts. It is especially noteworthy on standardized tests and final exams: there is NO feedback.)
To better understand what helpful feedback is, let’s return to the opening 4 comments and make them provide feedback:
“Nice job on the project, Sheshona! You answered the essential question in great depth, with lots of illustrative examples, and your oral presentation was polished and informative.”
“I found it very difficult to grasp your main point. At the start, it seemed that you were arguing against mining coal, but in paragraph three you focused on the need to provide healthcare to all workers. Next time, Sam, you’ll want to make your thesis clearer to the reader”
“Your spoken delivery was clear and your account of the topic was a helpful and interesting summary: most students were engaged. Alas, the supporting materials you supplied looked unfinished and rough; 5-6 students were confused by them. The lesson would be more effective, Shana, if your visuals were more polished and supportive of the teaching.”
“You taught about ants, Stefan? I LOVE ants! However, the task was not to please me; the task was to make students ignorant or afraid of an animal to become interested in them. Yet, you began as if the students already shared your interest in ants instead of helping them overcome their distaste and become more interested in them.”
Note that even with this added feedback, much of this might be less necessary if the teacher/supervisor provides a set of varied models, along a continuum of quality, for students/teachers to study and to use in self-assessment during and after the performance. Now, the system is approaching that of comedy and music: a self-correction system, with minimal input from coaches.
Imagine a world in which students and teachers had access to videos of all key performances, on a continuum of quality.
As I have noted before, the best example I have ever seen in person was a welding class where students were expected to examine welds on a table – some of excellent quality, some not – before placing their own weld down and signing off on it. I watched a boy, thinking he was done, inspect the welds and go back to his station: he realized his weld was not up to standard just by comparing his to the models. So, feedback need not be labor intensive or wildly time consuming. Video games and the use of clickers make this point clearer.
In my next post, I want to comment on an important point on feedback in the recently-released Gallup survey on student engagement and learning. HINT: “positive reinforcement.”
Want to learn more? Here are earlier posts – this one with more practical examples.
PS: John Merrow of the Merrow Report sent a humorous piece of feedback on my post by email. Note his accurate feedback (and my accurate feedback back.)
Grant:
38 Responses
Do you have any examples or appropriate feedback for high school math? I struggle with this because just identifying incorrect answers seems liketoo little information, but circling and/or pointing out an incorrect step seems to take away the students need to search for their own mistakes. I know some teachers also think writing detailed explanations is important, but with the need to give multiple formative assessments, giving extensive feedback leads to less assessments due to time restraints.
Making kids search for their own mistakes may be useful in some ways but it is the opposite of clear feedback, of course. The key is to use varied criteria, not just “right” or “wrong” – and that requires real problems not just simple right-wrong exercises. Then you can say such things as “your solution is inefficient even though it is accurate” or “your solution is accurate but unrealistic in context”. Thus, you can have rubrics for ACCURACY, EFFICIENCY OF PROCESS, REALISTIC/APPROPRIATE and give feedback against those terms. Labor-efficient, too.
However, in terms of the way you cast the question: why not circle the ones incorrect without saying why. Give students 10 minutes to self-correct. If they explain why the answer is wrong and if they self-correct, they regain some points (but the added points are noted separately in the grade book). You could also use simple code: F for incorrect “framing”, “R” for incorrect “rule” used, and “C” for “computation” error in giving initial or secondary feedback.
Thanks Grant,
I really like the coding idea. It reminds me of something a government teacher in my school was describing for a school wide editing “code”. I had forgotten about it until now. We don’t actually record all of our formative assessments, so points back on those wouldn’t always work. We are also refining a “Standards Based Grading” approach, so students have the ability to retake quizzes that are not summative. I think the coding idea would possibly be an idea that our department would be able to support. ,
Great! Let us know if and when you do something with it.
As I have a political agenda, as a parent and former educator, to see an end to high stakes testing, I’d like to support your blog and add that our current high stakes testing programs do not provide any feedback – and not even “advice”. The scores given 3 months after testings during the Summer are virtually meaningless to the student. The teacher evaluations that utilize scores that may or may not be from students the teacher teaches give zero meaningful feedback. The “grade” given to schools which show more about student SES level than the school quality is meaningless and hurtful to the school and neighborhood.
We need meaningful and timely feedback for our students, schools, and teachers. Please continue to beat the drum Grant. Thank you!
These are all apt claims, as I have discussed before. Over 20 years ago I made this very argument to the State Board in NC, the legislature in KY, and the Commissioner and Board in NJ. Nothing happened, even as people nodded and spoke in agreement about the idea. To my mind, this is the REAL scandal involving testing. The testmakers should have to provide child-specific feedback based on a fully-released test – as has long happened in other states. (Part of the MASS. miracle was that for over a decade the tests were released as soon as they were given.) That state ed. people have failed to write the RFP or demand this of all vendors – ESPECIALLY with PARCC and SB – is the key travesty in the regime of testing. Release of tests would also provide greater transparency in all testing to ensure validity.
Thank you Grant. I have to add… since tomorrow is my son’s FCAT Reading test (2 days each with 90 minutes of testing- sit and stare when done with the test) that he will have to sign a contract prior to taking the test where he agrees that he will not mention anything about the test contents to anyone. If he does, his test score will be voided. He must pass this one to graduate from high school. So, if he speaks with another students about how difficult (or easy) this or that item was and why, he puts his graduation on the line.
I mention this because it is another form of feedback that our current high stakes testing program makes impossible to achieve. Students, in sharing information about how they did on this or that test item (or in sharing with mom and dad), get feedback. I can recall many times, during my education, discussing an especially challenging test item after the test with cohorts which was extremely helpful in learning. There are many forms of feedback- not just directly from the teacher. And feedback is vital to learning.
This contract thing is very disturbing. I have been shocked about how many people cite it on our site talking tests.org as a reason they cannot be more specific. Horrible turn of events.
Best of luck to your son!
I enjoyed reading your post. I recently attended a seminar with Dylan Wiliam on assessment, which featured formative assessments and feedback prominently. As a teacher who has worked in the IB system with classes from grades 4 through to 10, I believe that no matter the grade, another important factor to add to your 4-point list of why there is too little (and often unhelpful) feedback is that said feedback is simply given TOO LATE for any meaningful retention or practical application to take place.
What is the point, for instance, in giving feedback on a persuasive piece of writing once it’s “handed in” if the student will not have the chance to adjust the current item, or produce another one of the same text type? And yet, teachers often commit this mistake, be that due to a lack of foresight and planning (failing to account for the unexpected within a unit/sequence of learning), or other reasons that shorten the length of time and/or the number of opportunities for the student to apply his or her learning in practical ways so as to yield the desired (teacher and peer) feedback in a timely manner.
Given that our inclusive classes present us with the challenges of facilitating learning for students who present a wide range of levels of understanding and ability, isn’t it the case that those who are able to produce a complete first draft earlier than other less able peers are the ones who end up benefitting most from timely feedback? Perhaps this is an unfortunate, but frequent situation in classrooms.
Excellent point! Timeliness is a key criterion, mentioned prominently in my earlier blog. Thanks for highlighting it.
Reblogged this on principalaim and commented:
Good Feedback is essential to our work with students and teachers. I truly believe this notion (and have always believed it) because I’ve seen what good feedback can do. Good feedback can inspire, motivate, support, empower, and encourage. However, like Grant, I believe good feedback should provide “insight that produces outcomes or “effects.” I appreciate Grant’s thoughtful insight because feedback matters – especially really good feedback. tlb
Hi Grant,
I really like this thoughtful post and your examples of feedback with excellent distinctions for advice and praise. That is SO important!
However, in some cases, like the welding example, I think you are talking more about students’ self-evaluation than feedback. While teaching second graders to practice cursive writing (back in Finland, about ten years ago) I used similar type of student self-evaluation method: before showing me their lines of work students went to ask for advice about their cursive writing from a lion puppet. I often overheard students talking to the lion, asking for opinion (and thus getting help from their internal talk) to their task of marking the best letters or words (i.e. the ones that were most like the example in the beginning of the line).
Feedback is so very easy to combine with self-evaluation practices, because without the mutual understanding about what the task entails, even the best intentions of providing feedback may not produce the desired results. In training and during my courses I always try to explain how “real” feedback requires a dialogue between the teacher and the student. Often this means several back-and-forth exchanges for both parties to see what the other person is seeing. Having strong self-evaluation practices cuts the amount of negotiating the meaning significantly, and thus saves time in the classroom.
I would like to add this your blogpost as a link on my website about feedback: http://ninacsmith.com/3CLearning/Nina's3CTools/CognitiveTools/FEF-Feedback.aspx
Thanks,
Nina
But it is important to see that the feedback is what permits the self-assessment. The feedback is contained in the comparison of welds. The boy looks at his weld and compares it to ‘good’ welds and sees that his is not where it needs to be. That’s no different than the joke not causing sufficient laughter. You are right, of course, that the feedback leads immediately to self-evaluation but they are notably different. He could have ‘failed’ to assess his work properly and to have left it there. To go back to his station is the ‘advice’ based on the ‘feedback’ that is visible.
Reblogged this on iAMSTEM HUB . UC DAVIS.
I would point out that just as feedback is essential to achieving a high-level performance, so is ability which you rarely seem to mention. If feedback was the primary driver of performance levels, then I could, say, learn to play baseball at the major league level given enough feedback. Clearly, that is not the case.
This is not meant to undermine your thoughts on feedback, which I think are solid. But if you present a “continuum of quality” for students to see, then it should be noted that some students may not have the ability to reach the highest level of quality. Or using one of your previous examples, it is not possible for every student to achieve the highest score on the AP Art exam.
I fail to understand what is gained by your harping on this. There is no way to know a student’s potential, so why not report and encourage?
I’m not trying to harp on this, but I think a philosophy of “standards” that ignores differences in ability has negative real-world consequences. Look at what is happening in North Carolina: http://www.charlotteobserver.com/2014/01/14/4611271/new-nc-reading-push-brings-expenses.html
Half of the 3rd graders in the state failed a new, harder reading exam, and now they want to send these thousands of kids to mandatory summer reading camps. If it was just “report and encourage”, I would support that, but that is not what is happening. It is test-and-punish with no reference to underlying ability.
Also, yes you can know a student’s potential based on simple tests:
http://www.ncbi.nlm.nih.gov/pubmed/19245744
“Data from less than an hour’s worth of testing pupils could predict school exam results 6 months later.”
“They underlie the seemingly obvious fact that intelligence tests taken many months, or indeed, years before high school examinations can predict them very well.”
And there are many more studies that say the exact same thing.
Reblogged this on teaching knowledge and creativity and commented:
This is a wonderful article about FEEDBACK. For teachers it is necessary to think again and again about their own feedback qualities. But the hard job is time-consuming and so we often need help to remind us what’s really important in our job to develop the student-teacher-relationship. And feedback, I think, is one of our best measurement to enhance the educational process.
I agree that it this is very time consuming and that student-teacher relationships are very important. By building the relationships, students are more open to the feedback and taking the time themselves to make the improvements necessary.
Great article especially the comment on rubrics. I find in my school too many teachers use these rubrics that use such words as almost, thoroughly, usually. What do they mean on the rubric and what feedback to they provide? I agree it has to be specific, meaningful and measurable. Question – what do you do when you provide feedback to a student and there reply is that is fine but I am happy with the assessment and I don’t want to improve. I will be happy with that as a final mark.
You raise a crucial issue about motivation and expectations. I would want to know the pattern, not just answer on the basis of a single event. After all, there is other work to do that doesn’t stop; one doesn’t want to fall behind and dig into a piece of work that one hated doing. We can all empathize. But if it is a pattern of low-balling one’s own expectations and thus a failure to learn about quality and one’s own abilities, then there should be feedback – and consequences. For example, passing a course requires revising at least 2 poor papers; and the marking system should reward gains.
Thanks for a very useful article. As a math teacher, I struggle with how to best provide feedback. I want students to find their own mistakes, but I know that they need guidance, encouragement and motivation to find and correct their work. If I identify the mistake, they don’t need to think. If I just mark the problem wrong, they get frustrated and don’t know where to start. I will try to give feedback by writing questions on their paper and having personal conversations. I also want to know what they were thinking and that is easier when they talk to me.
Actually, I don’t think it’s true that if you identify the mistake they don’t need to think. They could be asked to explain WHY it is incorrect, and be expected to correct it as part of the test. You can also make sure to ask questions that get at common misunderstandings or misconceptions that they have to explain…
After reading this article, I realize that feedback I have been giving is not always beneficial to students. I plan on working on using your methods to further provide my students with feedback that will benefit their success in my class. I also would like to mention that I found the comments by others about standardized testing not having feedback very interesting. These high stakes tests are very important to our students, community, and school but no feedback is given in a timely manner.
This is very true. So, regardless of the validity of the test or the merits of it as accountability, it is a terrible feedback system. Which means that we need to provide far better feedback on local assessments.
I love your commentary about student self assessment. What are some stratgies for buidling students self assessment skills
Make it part of the grade – the accuracy of self-assessment as you hand it in. Also, ask them to use their rubrics or guidelines as if they were the teacher commenting against the rubrics. Make them identify 2-3 strengths in the work and 2-3 weaknesses in the work. Ask them to compare their work to models that you supply
I think many teachers fail to give specific, useful feedback it because they feel as though students should figure it out on their own. As if giving specific detailed feedback is cheating. How to get an “A” on a performance based task shouldn’t be a mystery. As experts in our field we forget that student don’t know the steps to mastery. We should provide the map. I also find it disturbing that some think that students cannot reach the “highest level of quality” . In my experience some students may require more time, more effort, or move motivation to reach the highest level of quality. That doesn’t mean that do not have the potential. I would love advice for students who seem content with a lower grade and who do not want to follow the steps I outline for mastery.
This is a surprisingly common issue – kids who lowball expectations or who fail to take any opportunity to improve performance. I can’t say that I feel 100% confident in my advice about this but I suspect that it is a combo of low self-esteem, balancing many competing demands/assignments and some laziness, given that we accept whatever they produce. 1 approach is to say: your only options on this assignment are A, B, incomplete or failure. Another approach is to say that you need at least 4 A’s to pass the course. Another approach is to say you can choose 6 of 9 graded assignments for the final grade. In short, you should experiment with incentives that motivate kids to care about quality and/or realize that they can have a few poor efforts that don’t hurt their overall grade.
I agree that students need feedback about their performance in class. Students need positive and negative feedback. I try to give one negative feedback for every 2 positive feedback statements. The most frustrating part as a teacher is that students don’t want to read the feedback they are focused on the grade. I have found when I tell the students that if they want to redo the assignment using my feedback that I will regrade the assignment.
I understand and empathize with your frustration! But I also empathize with the kids because grades are the coin of the realm. Perhaps they need to see that taking the feedback and improving their work improves their grade – maybe that is not clear enough. Or, maybe it just seem like way too much to rewrite something if the class moves on to new content. The only way to get kids to care about feedback is to make it clear that good results happen from tasking the feedback and using it effectively.
I very much agree with the idea of providing students with models of excellence. Students need to see what excellence looks like. However, I am keeping in mind the students who have always struggled with writing. Even when given examples of a thesis statement, these students cannot bring themselves to write a thesis statement. I have one on one conversations about how to write the thesis statement. When I have these individual conversations with students I ask them if they lack confidence in writing. They will look at me very shyly and respond with yes. How do I help students build confidence in writing? When I have individual conferences with students on how to improve their thesis statements, I would like for their thesis statement for the next assignment to be clear, concise and specific because we have already discussed those ideas, we have already seen examples of thesis statements, and we have revised thesis statements together. However, by the time we have the next assignment, students still struggle with beginning their own thesis statements.
At what point do I move away from providing them the “excellent” example, so that they may begin writing their own thesis statements and have confidence that it is a well-written thesis statement? I believe they see their thesis as “well mine does not look like that, so then I still can’t write thesis statements.” I also don’t want students to copy down the “excellent” thesis statement and rearrange it so that it works for their purpose. Furthermore, what about that struggling writer who even though examples after examples are given, still can’t bring himself or herself to write at all.
Confidence in writing is very difficult to attain. Many professional writers do not have it!! I have often doubted my own competence as a writer, etc. So, it’s ok to lack confidence as a writer; it’s common and natural. That said, it is always helpful to have models and stems and prompts to get you out of a rut – even if it seems like cheating or taking away some of the responsibility. So, I would experi,net with giving them more models, prompts, stems, and checklists; and see if gutting over the hump and getting going is enough to get them to produce some good work.
Thank you for your thoughts and examples on the importance of quality feedback. I have been working with my students on having them provide peer reviews of each others work. The problem is that the level of feedback is more to praise or to point out mistakes without providing quaility information. Do you have suggestions or strategies that could be used to teach this to students. I feel if they could provide the feedback then they would understand better how to self assess their own work.
Thanks
Darren, it HAS to be modeled. Show clips from Project Runway, American Idoal, Top Chef; do modeling yourself of helpful/unhelpful feedback. Make it clear that the aim is not a summary judgment or glib statement but saying something that is genuinely helpful. And hold t5he feedback peer givers accountable for how useful their feedback is.
Mr. Wiggins,
I enjoyed your thoughts on feedback. This statement jumped out at me for some reason. “Feedback is useful information about the effects of an action in light of a goal.” I immediately thought “goal?” In order for feedback to work both the teacher and the student need to know the goal. I feel as though, too often the student is unaware of the true goal to an assignment and they are completing the assignment because they are told they have to or they will fail. Have you heard of or seen this before? I have made it a goal of mine to create a more democratic classroom, but my students don’t seem to have any drive to become involved with the classroom atmosphere. I guess what I am trying to say is that my student’s don’t seem to have goals so how can my feedback be effective.
This is a fantastic point – the goal has to be crystal clear and ‘owned’ or the feedback is likely to be misunderstood or ignored. So, some effort on getting students to design/own/accept some goals is very important. You might want to at the very least start by making clear what the goal of an assessment or project is.
Reblogged this on Kumpulan.