This is going to be quite a long post about feedback- make tea now! Prior to becoming a teacher, feedback was a very familiar term to me in the context of engineering, specifically in a systems ability to self-regulate (think of a GPS system giving feedback on where you are and guiding you to a specific destination). In such systems, the output is monitored and fed back to the input to cause desirable changes to the system.
As a teacher, I have found feedback used as a catchall for a diversity of practice (in much the same way that AfL is) and is always considered as self-evidently a ‘good thing to be doing’. This is partially true, in that meta analyses carried out by different groups (Hattie and Timperley, 2007; Kluger & DeNisi, 1996; Shute 2008) have established that the effects due to feedback are among the highest that we know of in education, but also that feedback effects are among the most variable. The Kluger review, in particular, showed that engaging in high dosages of feedback doesn’t always correlate with learning. They found that in 38% of well-designed studies, feedback actually made performance worse.
What is clear is that feedback is powerful, but effectively harnessing this power is trickier than it may appear. Side effects that have appeared in several studies have included feedback influencing how much effort students allocate to activities, that some feedback can ‘harm’ learning and that students ‘tune out’ when teachers administer group- level feedback (Hattie and Gan, 2011).
Feedback: information allowing the learner to reduce the gap between what is evident currently and what could or should be the case (Sadler, 1989).
Common sense suggest that feedback is of little use unless students are given time to act on it. Dylan Wiliam argues this point stating that ‘if it’s worth spending your time generating feedback, it’s worth taking instructional time to ensure that students respond.’ Carefully created feedback is really 1-1 tuition and as such is pure differentiation. So, if marking can instead be viewed as planning, then it follows that it is worth dedicating part or all of a lesson to what has come to be known as DIRT (Directed Improvement & Reflection Time) in which pupils act on your feedback. This one-to-one tuition is the most expensive form of education- what Benjamin Bloom called “the gold standard” of instruction (Bloom, 1984). Dylan Wiliam also argues that feedback should be more work for the recipient than the donor and by providing the time and the expectation that it will be read and acted upon, it will be. To offset workload issues, Joe Kirby suggests the following:
Don’t write out comments. You end up writing such similar comments across the class, and they won’t read them anyway. Instead, get them to write them out. Choose three to five targets or questions before you start marking, then scan their answer, choose the best fit between the student’s work and the group target, and draw an icon. One minute per book maximum. At the start of the next lesson, you write the targets on the board, students write their targets in their books. They get instant feedback and can take action on their target straight away.
And in Teach Like a Champion (P185- 6), Doug Lemov shares a strategy from a Dutch teacher he observed:
The idea is that you assign students independent work and, as they are working, circulate to observe their work. If their work is wrong, you put a dot on their paper. Very subtle, not a permanent “wrong” mark, just a reminder that there’s something that needs checking. And here’s the best part – that’s ALL you do/say. The idea is that the dot reminds students, subtly, to find their own mistake and, in time, encourages self-reflect and self-correct. You could even then ask students to discuss: who got a dot and found it? who got a dot and didn’t.
Thinking a bit about goals next- there is a lot of literature on this, such as Locke & Latham, 2013. These suggest that feedback can’t exist in isolation as it’s impact must also depend on the nature of the goals set and on a students conception of an assessment. I’m thinking about examples I’ve seen of detailed feedback, given in checklist form, setting out what to include in a new shinier version. When asked, students have had no idea that they should include these in the first place, otherwise they would have! This looks like feedback but isn’t- it’s down to success criteria not being known by the learner in advance, and so the feedback is really brand new work, new success criteria if you like, rather than developmental, and so isn’t necessarily addressing the sweet-spot of challenge that good feedback can.
In terms of how students are progressing with a particular task, Dylan Wiliam is clear that grades and feedback must be separated- in formative feedback, grades should not feature.
Letter or numerical grades on papers give students information about their current performance. Class rankings give students information about their performance compared to others in their class. But grades and rankings do not in themselves give students one iota of information of how to improve. When teachers pair grades with comments, common sense would tell us that this is a richer form of feedback. But our work in schools has shown us that most students focus entirely on the grade and fail to read or process teacher comments.(Dylan Wiliam, 2014)
Thinking about the nature of feedback in terms of the direction a student should take or their next steps, the GPS analogy is useful to consider. The GPS directs you to your destination without reminding you of your wrong turns or errors, it just looks forward. Similarly, studies find that students aren’t that interested in a postmortem but want definite guidance that they can act on NOW (Lipnevich & Smith, 2009).
The cognitive psychology page on this site (here) talks a bit about Dan Willingham’s thesis on learning, in particular that we are biologically and psychologically resource limited and so students are likely to be unconsciously economic in their cognitive efforts. Our feedback to students should acknowledge this by scaffolding and instructing to make difficult goals seem bridgeable (jumping the chasm in Willingham speak)- this will encourage students to move forward. Hattie and Yates (2014) suggest that the nature and timing of feedback should address a students current position in function. In beginning to learn the base facts- beginners need immediate feedback which is more focused on encouragement and of a right/wrong nature, for those working at an ‘application’ level, it seems less critical that feedback happens straight away and they tend to need confirmation that their inferences are right, such as ‘good use of a principal’. Those at a more advanced understanding may need (or benefit from) strategies that reduce cognitive load- these don’t seem to be time critical- apart from at the point of possible cognitive overload! There is also evidence that sometimes delaying feedback is a good thing to do, with Bjork saying that “feedback that is given too immediately and too frequently can lead learners to overly depend on it as an aid during practice, a reliance that is no longer afforded during later assessments of long-term learning when feedback is removed.”
In summary, from my reading and observations, the following principles identify effective practice in feedback:
- Be clear to students that you will expect them to act on your feedback and give them time to do this– so come to view feedback as planning and allocate lesson time to act on it (DIRT).
- To save time when providing feedback on work use codes and when circulating during a lesson use the Dot Technique.
- Feedback can work best when success criteria are known by the learner in advance, are in the sweet-spot of challenge and where the goal of achieving such success (the culture of motivation and striving) is shared by both students and teacher.
- Feedback is more effective when summative grades are not presented with formative feedback comments- students will focus entirely on the grade and neglect the comments.
- Feedback should focus on looking forward, helping students to see a structure that will make investing thinking worthwhile as a way to bridge the chasm.
- Feedback should engage a learner at, or just above, their current level of functioning with consideration given to timing and possibly delaying.
- Feedback works well when students see errors as a natural part of learning rather than a criticism.
Hattie, J.A.C., & Gan., M. (2011). Instruction based on feedback. In R. Mayer & P. Alexander (Eds.), Handbook of Research on Learning and Instruction. (pp. 249-271). New York: Routledge.
Hattie, J.A.C., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81- 112. doi: 10.3102/003465430298487
Hattie, J.A.C., & Yates, G. C. R. (2014). Visible learning and the science of how we learn. London: Routledge.
Kluger, A. N., & 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. doi: 10.1037/0033-2909.119.2.254
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57(9), 705-717. doi: 10.1037/0003- 066x.57.9.705
Locke, E. A., & Latham, G. P. (Eds.). (2013). New developments in goal setting and task performance. New York: Routledge.
Sadler, D. R. (1989). Formative assessment and the design of instructional systems. Instructional Science, 18(2), 119-144. doi: 10.1007/bf00117714
Willingham, D. T. (2009). Why don’t students like school? A cognitive scientist answers questions about how the mind works and what it means for your classroom. San Francisco: Jossey-Bass.