Every teacher now knows that feedback should be central to how we teach. And they should also know that by feedback, we do not mean simply giving grades and marking assessments.

Instead, effective feedback is timely, suited to the students’ needs and includes strategies for improvement. It should help students to stay engaged and make meaning of their learning in the classroom, so that they develop the skills to be lifelong learners.

But despite knowing that, do we do enough to put it into practice? And while schools have been closed, have we used technology to make sure that we have given the right feedback to the right students?

We recently surveyed the needs of both our teachers and students at Stamford American School Hong Kong and found that feedback in reading was something that both students and teachers needed to improve.

Thirty per cent of fourth-grade pupils (equivalent to Year 5) and 54 per cent of fifth-grade pupils (Year 6) were four or more reading levels (one full grade level) behind reading expectations, according to Fountas and Pinnell Benchmark assessments.

One explanation for this is that out of the 740 pupils enrolled, 263 are children for whom English is not their first language (based upon their incoming English proficiency test scores using World Class Instructional Design and Assessment). Nonetheless, we believed feedback was also an issue.

So, what did we do? We wanted to find out if technology could help us (this was before schools closed, but the question has taken on new importance during the past few months). So we set out on a research project to answer the following questions:

• To what extent does audio recording and dialogic feedback implemented for six weeks improve reading growth among previously low-growth fourth- and fifth-graders?
• To what degree and in what ways are previously low-growth fourth- and fifth-graders able to respond successfully to iterative audio-recorded dialogic feedback during a six-week reading unit?

Why dialogic feedback? Often studies show that feedback is irregular and fragmented. To achieve a more sustainable model, it is emphasised that students need to be able to self-regulate, and one of the most important features of effective feedback is how the student interprets and uses the feedback to increase learning. Although it was once thought that feedback had to be directed solely by the teacher, researchers are now putting emphasis on the ways that students can engage in feedback with both their teachers and their peers. Dialogic feedback allows for students to practise using their cognitive (questioning, critical evaluation, engagement, etc), social affective and structural skills.

There are a number of ways to create the conditions for dialogic feedback, but technology is one method that seems to be successful. In one study, students were recorded making presentations and teachers used this recording as a resource for dialogic feedback, with the intention that this would increase students’ self-monitoring, leading to more sustainable, effective feedback. This also led to a situation where students could engage in the dialogic feedback with each other (Carless et al, 2011).

But even though studies have looked at the outcomes of student learning, we felt there could be more consideration on how effective feedback specifically affects low-growth students. Additionally, research is frequently focused on older students - there is no research on providing these types of feedback to elementary-aged pupils. In our project, we looked at exactly these things.

Teachers first assigned a reading on the online guided reading programme Raz-Kids, which is used school-wide for reading fluency, comprehension and summaries. Teachers also assigned a benchmark assessment on Raz-Kids, which allowed the pupils to audio-record themselves reading an assigned passage.

Pupils were given feedback on their recording in a small group setting, which also allowed them to provide peer feedback. Teachers were equipped with conferring templates to assist with the types of feedback to give when assessing reading. Observers monitored the types of feedback given by the teacher on a scale of passive/moderate/strong. Pupil engagement and acceptance to feedback were scaled by somewhat resistant/apathetic/neutral/engaged.

Pupils’ ability to retain feedback given was assessed with first and second iterations of the above process. The pupils were able to express their opinions and thoughts, which highlighted how successfully they were able to respond to the feedback given.

Quantitative data was gathered in the form of monitoring correct words per minute (CWPM) across the assessments, which we compared with a test that was administered at the end of the unit. In addition, five brief comprehension questions were created for pupils to answer independently after reading out loud the first time. Post-tests with comprehension questions were then administered at the end of the study.

So what did the results show?

For the five-question pre- and post- comprehension tests, while almost all pupils performed the same or better, a few actually did worse. Regardless, the overall average comprehension increased modestly from 4.20 to 4.34 questions answered correctly. Likewise, pupils’ reading accuracy was almost unchanged, just slightly improving from 93.05 per cent to 93.87 per cent during the study.

Of note, however, was the somewhat significant increase in the number of words per minute pupils were able to read at the end of the study, as compared with the beginning (while still maintaining similar accuracy). Pupils’ CWPM increased from an average 96.49 to 106.03, a 9.89 per cent growth in a six-week period.

What about the pupils’ ability to respond to feedback? We started by looking at the change in perceived pupil engagement when feedback was being given. These results were mixed at best. Likewise, when we compared first and second iterations and asked if pupils knew how to implement the advice, again there was not a significant change. Twenty-six pupils (74 per cent) in the first iteration said they knew how to implement the strategies; that number grew to a modest 30 (86 per cent) with the second iteration.

So, what do we take from this?

We are relying on tech for feedback more than ever during the current crisis and the above method does set out one way we might conduct feedback on reading during remote learning. It was noted that several of our originally predicted outcomes were simply not borne out with this study: the study did not conclusively prove that the audio-assisted iterative feedback method has a significant impact on previously low-growth pupils.

Still, we noticed pupils applying their newly acquired learning strategies to other subjects, and this has been encouraging. At the same time, by modelling giving feedback in a small group setting, we saw students acquiring the skills and language to provide peer feedback effectively. They felt empowered to provide and receive peer feedback, which is crucial in a world of collaboration.

That said, we have observed first-hand how the overall process is a benefit to students, especially with their reading fluency. So what is our verdict? More work needed: which is feedback we can definitely work with.

Michael Galligan is an academic coordinator, Rachel Goss is a homeroom teacher and Trishna Harjani is an upper elementary teacher at Stamford American School Hong Kong

This article originally appeared in the 22 May 2020 issue under the headline “Listen up: audio isn’t a panacea for feedback’s issues”