Feedback is one of the most powerful influences on learning and achievement, but this impact can be either positive or negative. Hattie & Timperley (2007)
Feedback is the engine that powers learning. But from the quote above, it’s clear that not all feedback is effective. This presents a bit of a quandary. How do you know if the feedback you’re providing learners is the right kind? For answers, we turn to the research. And believe me, there is no lack of research on this topic. In fact, trying to understand it is like trying to sip from a fire hose! In this post, we’ll try to clarify the complexity of this foundational practice of assessment for and as learning.
Here’s what we can learn about feedback from the research:
Descriptive feedback is the most powerful tool for improving student learning.
Black, Harrison, Lee & Wiliam (2003) made this statement based on their review of 250 studies or books that contained “evidence that innovations in formative assessment can lead to improvement in the learning of students” (p. 7). And note this: the studies were diverse – students aged 5 to university students, across several subjects and several countries. The effect sizes of the studies were between 0.4 and 0.7, which range from impactful to highly effective.
In 2007, John Hattie’s and Helen Timperley’s systematic meta-analysis of 12 previous meta-analyses incorporated 196 studies and nearly 7000 effect sizes, and concluded that feedback had a powerful effect on learning outcomes.
The nature of the feedback – its content, the way it’s delivered, the timing, the focus — affects whether it will have the desired positive outcome on learning.
Kluger and DeNisi (1996) found that one-third of the studies showed negative effects, where feedback actually harmed learning.
When I reflect back on my own personal experiences as a student, the feedback I remember receiving was in the form of
- teacher approval or disapproval e.g. “you’re a good student” or “I don’t like the way you did that”;
- right/wrong marking e.g. 14/20 correct answers;
- written comments on work e.g. “start new paragraph here”, “unclear?”
As it turns out, the most effective of these three examples is the third: written comments on work move learning forward if they provide to the learner a description of what is being done well, and what needs to be done to improve. Even that has nuances. The comment, “start new paragraph here” is helpful in improving the work, but doesn’t give the learner a clue as to why starting a new paragraph improves the piece. Further, effective feedback doesn’t need to be written (we’ll look at that in another post!).
A Model for Analysing Feedback
Hattie and Timperley propose a model of feedback which helps us to identify what kind of feedback is helpful to learners. They suggest that feedback can be focused at four ‘levels’:
At the level of task performance
This feedback focuses on the correctness or sufficiency of the student’s response or performance. Is the answer provided correct? Has the learner provided enough information? Is the information relevant? For Ontario educators, the success criteria used to generate this feedback relate to the knowledge and understanding category identified in the achievement chart, an assessment framework that sets out high-level criteria by category of learning.
At the level of process
This feedback focuses on the various processes and skills students must master in order to think deeply about the subject matter. Does the student demonstrate an understanding of and a mastery of the processes related to the discipline? In Ontario, the curriculum policy documents identify various processes and skills that students are required to demonstrate. These can be found in the introduction to the document, in the achievement chart (often the Thinking category), and in the expectations. Here are some examples:
|Curriculum Policy Document||Introduction (example)||Expectations (example)|
|French as a Second Language||Critical thinking and critical literacy (p. 46)||Gr. 4 A1.1: identify a range of listening comprehension strategies and use them
appropriately before, during, and after listening to understand oral French texts
|Health and Physical Education||Living skills – the personal, interpersonal, and critical and creative thinking skills p. 18||Gr. 1 Living Skills 1.5 use a range of critical and creative thinking skills and processes to assist them in making connections, planning and setting goals, analysing and solving problems, making decisions,and evaluating their choices in connection with learning in health and physical education|
|Language||Writing process (p. 12)||Gr. 6 Writing: 1.6 determine whether the ideas and information they have gathered are relevant, appropriate, and adequate for the purpose, and do more research if necessary|
|Mathematics||The mathematical processes (p. 11)||Gr. 8 develop and apply reasoning skills (e.g., recognition of relationships, generalization through inductive reasoning, use of counter-examples) to make mathematical conjectures, assess conjectures and justify conclusions, and plan and construct organized mathematical arguments|
|Science & Technology||Scientific inquiry; experimentation skills; research skills (p. 12)||Gr. 6 Understanding Life Systems 2.3 use scientific inquiry/research skills (see
page 15) to compare the characteristics of
organisms within the plant or animal kingdoms
|Social Studies, History, Geography||The inquiry process; spatial skills (p. 22)||Gr. 3 People and Environments B2.4 interpret and analyse information and data
relevant to their investigations, using a variety
|The Arts||The critical analysis process (p. 23)||Gr. 7 C2.2 analyse, using musical terminology, ways in which the elements are used in the music that they perform, listen to, and create|
For both the task level and the process level, the feedback must relate to the knowledge and skills identified in the curriculum documents. This is where the learning goals and success criteria play such an important role. Once identified and shared with the learner, these two tools form the basis for feedback.
At the level of regulation (the metacognitive process)
This feedback relates to the learner’s ability to monitor and assess progress toward achieving the learning goals. Is the learner aware of the success criteria? Does the learner understand the criteria? Is the learner able to provide evidence of meeting the criteria in their response or performance? Can the learner identify next steps in learning?
Some of the curriculum expectations actually relate to these skills. For example, the Language curriculum has a metacognitive expectation in every strand in every grade. Here’s an example:
- reflect on and identify their strengths as writers, areas for improvement, and the strategies they found most helpful at different stages in the writing process.
At the self or personal level
This feedback focuses personally on the student, and is not connected to the specifics of the learning goals and success criteria. It contains no information about the learning, but rather, focuses on general observations of success or failure, and attributes the success or failure to personal aspects of the learner: “You are a great student.” “That’s an intelligent response, well done.”
Which kinds of feedback are helpful to learners?
It turns out that feedback focused at the task level, the process level and the self-regulatory/metacognitive level are all helpful to the learner, whereas feedback as the self level is not. As described above, each of these three serve a specific purpose – to provide detailed, descriptive information to the learner identifying strengths, areas for improvement, and suggestions for how to make progress.
Further, Hattie and Timperley (2007) suggest that the most effective feedback aims to “move students from task to processing and then from processing to regulation…” In fact, “too much feedback only at the task level may encourage students to focus on the immediate goal and not the strategies to attain the goal.”
“… the most effective feedback aims to “move students from task to processing and then from processing to regulation…” Hattie and Timperley (2007)
So what would this look like? Here’s an example from the Ontario curriculum, Grade 3 Mathematics, Number Sense and Numeration Strand
The educator is focussing on the following curriculum expectations:
- apply developing reasoning skills (e.g., pattern recognition, classification) to make and investigate conjectures (e.g., through discussion with others);
- communicate mathematical thinking orally, visually, and in writing, using everyday language, a developing mathematical vocabulary, and a variety of representations.
Overall Expectation (highlighted portion)
- solve problems involving the addition and subtraction of single- and multi-digit whole numbers, using a variety of strategies, and demonstrate an understanding of multiplication and division.
– divide whole objects and sets of objects into equal parts, and identify the parts using fractional names (e.g., one half; three thirds; two fourths or two quarters), without using numbers in standard fractional notation;
Here are the learning goals and success criteria:
|Learning Goals||Success Criteria|
|We are learning to think about math, and to explain our thinking to others. (based on Process Expectations)||I know what a conjecture is.
When working on a problem, I can make conjectures about the problem.
I can investigate my conjecture to see if I can prove it.
I use mathematical words to describe my thinking.
I tell step-by-step what I was thinking, what strategies I used, and how I reached my conclusion.
|We are learning to divide things and groups of things into equal parts, so that we can share. (based on Overall and Specific Expectations)||I can divide a whole thing into equal parts.
I can divide a set of things into equal parts.
We are learning to think about a group of objects in two ways – as a whole group, or as a group of individual parts.
I know fractional names and what they mean.
Samples of Feedback
At the level of task performance: “I noticed that you were able to divide the whole chocolate bar into equal parts. Have a look at the set of blocks and show me that the set is divided into equal parts as well.”
At the level of process: “You gave the correct answer at the beginning of your explanation, and then went back and explained how you came to it. I can clearly follow your thinking, because you told us step by step how you used strategies to prove your conjecture. I suggest that next time you share your conclusion at the end of your explanation.”
At the level of self-regulation/metacognition: “Your feedback to Jacinta about her explanation is very helpful. You used the criteria to tell her one thing that she has done well, and to make a suggestion for what she might work on next.”
At the self or personal level: “You did a really good job finding the answer to the problem.”
What does the research say about the distinction between feedback as grades vs comments?
It is not unusual for educators to use some form of grade or mark as feedback.
As far back as 1958, Page found that feedback in the form of short written comments rather than grades alone significantly improved the test performance of students in 74 classrooms.
Butler (1988) found that feedback through comments alone led to learning gains, whereas marks alone or comments accompanied by marks or giving praise did not.
These findings have been confirmed, repeatedly. The value of feedback lies in its ability to inform the learner with specifics about what has been achieved, what still needs to be achieved, and how to go about it. The learning goals and success criteria provide those specifics, which become the foundation of descriptive (think information) rather than evaluative (think grades) feedback.
One last thought, and a call to action.
Use the model of feedback described in this post to analyse the feedback that your students typically receive.
Is it focussed at the task, process, metacognitive, or self level?
Depending on what you find, you may wish to:
- Stop providing feedback at the self level;
- Start providing feedback at the level of task, process, and metacognition;
- Ensure that the feedback goes beyond the level of task. This may mean restructuring the learning goals, criteria and tasks, so that there is a focus on process (see the chart above!), and on having learners think metacognitively about their learning by identifying strengths, areas needing progress, and strategies to make progress.
Start the conversation – share your thinking, questions, experiences in the comments below.
Black, P., Harrison, C., Lee, C., Wiliam, D. (2003). Assessment for Learning: Putting It Into Practice. New York, NY: Open University Press.
Butler, R. (1988). Enhancing and undermining intrinsic motivation: The effects of task-involving and ego-involving evaluation on interest and performance. British Journal of Educational Psychology, 58(1), 1 – 14.
Hattie, J. and Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81 – 112.
Kluger, A. N., and DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119(2), 254 – 284.
Page, E.B. (1958). Teacher comments and student performance: A seventy-four classroom experiment in school motivation. Journal of Educational Psychology, 49, 173-181.
Stiggins, R. J., Arter, J. A., Chappuis, J. and Chappuis, S. (2004, 2006). Classroom Assessment for Student Learning:Doing It Right – Using It Well. Princeton, NJ: Educational Testing Service.