Teachers activating learners: The effects of a student-centred feedback approach on writing achievement

Highlights

• Examined a teacher professional learning intervention in student-centred feedback.

• Used a quasi-experimental design to examine feedback effects on student writing.

• Findings show larger achievement gains for intervention than comparison group.

• Several student-centred feedback practices were associated with achievement gains.

• Findings inform recommendations for teacher professional learning in feedback.

Abstract

This study examined the impact of a teacher and school leader professional learning intervention on student writing achievement in Australian state primary schools. The six-month intervention was underpinned by a new student-centred feedback model. The study analysed student writing assessment data from 1060 Year 4 students across 13 intervention and 9 comparison schools. Results from multilevel analyses showed substantially larger achievement gains in the intervention group compared to the comparison group across two learning periods. Increased perceived helpfulness of several feedback strategies was associated with student achievement gains in the intervention group, showing the intervention's positive impact on feedback effectiveness.

Introduction

The power of feedback as a means to reduce the gap between a student's current progress and the intended learning goal has long been recognised (Hattie & Timperley, 2007; Sadler, 1989). A widely-used definition of feedback is that by Hattie and Timperley (2007): “information provided by an agent (e.g., teacher, peer, book, parent, self, experience) regarding aspects of one's performance or understanding” (p. 81). Research overall reveals moderate to substantial effect sizes (ES) of feedback on student achievement (Hattie, 2009, ES = 0.73, p. 173; Wisniewski et al., 2020, ES = 0.48). This study is contextualised in English writing, a curriculum area where feedback from adults (ES = 0.87), peers (ES = 0.58) and students themselves (ES = 0.62) has been demonstrated to have substantial potential to positively impact student achievement (Graham et al., 2015). However, there is much variability in feedback effectiveness and, in some cases, feedback can inhibit learning (Hattie & Timperley, 2007; Kluger & DeNisi, 1996; Wisniewski et al., 2020).

Education research points to the importance of three interrelated pertinent features to realising the power of feedback: (1) the critical role of students in feedback needs to be recognised, as feedback cannot be effective unless students actively engage with it (Hattie et al., 2016, Lipnevich et al., 2016, Van der Kleij et al., 2019); (2) feedback needs to be embedded as a key feature of formative classroom practice (Assessment Reform Group [ARG], 2002; Wiliam, 2011; Wiliam & Thompson, 2008); and (3) teachers require substantial professional learning support to implement student-centred feedback practices (Black et al., 2003; Brooks et al., 2021; DeLuca et al., 2019). The subsequent sections discuss key literature in relation to these three pertinent features.

Historically, feedback has been conceptualised as a one-way transmission of information from teachers to students (Boud & Molloy, 2013; Van der Kleij et al., 2019). Within this transmissive paradigm, there is an assumption that students will act on the feedback as intended by the teacher. However, research has demonstrated that the intended effects of feedback are often not realised in classroom practice (Hattie et al., 2016), suggesting that how students receive and engage with feedback plays a major role in its effectiveness (Boud & Molloy, 2013; Hattie et al., 2016; Lipnevich et al., 2016; Van der Kleij et al., 2019; Winstone et al., 2017). For example, research in school contexts has demonstrated that there are sharp contrasts between teachers' feedback intentions and students' perceptions and use of this feedback (e.g., Dann, 2015; Van der Kleij, 2019). Characteristics of the learner such as their current ability and past achievement in the subject area, and their openness to feedback are likely to influence whether the feedback will be utilised (Lipnevich et al., 2016). In addition to concerns over lack of effectiveness of feedback, the transmissive paradigm puts the burden of assessing and progressing students' learning entirely on the teacher, which leaves the student dependent on teacher feedback that may never arrive (Boud & Molloy, 2013).

Contemporary conceptualisations of feedback position students as playing an active role in all aspects of the feedback process (Boud & Molloy, 2013; Van der Kleij et al., 2019; Winstone et al., 2017). Importantly, students are not regarded as passive recipients of feedback, but rather are required to actively seek, generate, provide, discuss and use feedback to progress learning. The student-centred paradigm places much emphasis on student self-regulation to realise the development of effective and sustainable feedback practices (Boud & Molloy, 2013; Clark, 2012; Van der Kleij et al., 2019). Namely, feedback plays a major role in self-regulatory processes such as goal setting, determination of strategies, monitoring of progress, adaptation of strategies and self-reflection, which foster student autonomy and achievement (Zimmerman, 2002).

Building off Ramaprasad's (1983) explanation of feedback attending to the gap between actual performance and a given parameter and Sadler's (1989) conditional requirement that feedback must provide a comparison between actual and expected performance, Hattie and Timperley (2007) proposed a conceptual model of feedback that can guide teachers in facilitating feedback that activates the learner. They proposed that feedback should address three questions for the student: “Where am I going?” (feed up), “How am I going?” (feed back), and “Where to next?” (feed forward). These questions highlight the need for students to have a clear understanding of the learning goal, be able to monitor their progress towards the goal, and understand how they can reach success by effectively acting on feedback (Sadler, 1989). Thus, although feedback as a stand-alone strategy is potentially powerful, it needs to be integrated into formative classroom practices for its effectiveness to be realised. As Sadler (2010) put it, “No matter how expertly and conscientiously constructed, it is difficult to comprehend how feedback, regardless of its properties, could be expected to carry the burden of being the primary instrument for improvement” (p. 541). This highlights the importance of integrating feedback within active learning contexts so students are more able to clarify and act on the feedback (Wiliam, 2011).

In an education context, feedback has come to be conceptualised as one of the five key strategies within formative assessment, or more recently, Assessment for Learning (AfL) (Assessment Reform Group [ARG], 2002; Wiliam & Thompson, 2008). The term formative assessment has been interpreted in many different ways, and its conceptualisation has evolved considerably over the past decades (Baird et al., 2014; Bennett, 2011). Broadly speaking, conceptualisations have shifted from a focus on use of assessment evidence by teachers to provide corrective feedback to students, to a focus on active student involvement in self-regulation (Baird et al., 2014) where students actively seek, generate and use feedback to improve their learning outcomes (Black & Wiliam, 2009; Clark, 2012). In this paper we focus on the latter conceptualisation, often referred to as AfL, which is underpinned by elements of cognitive, social constructivist and sociocultural learning theories (Baird et al., 2014). However, it must be noted that even the term AfL has been interpreted and operationalised in different ways (Bennett, 2011; Wiliam, 2011).

A widely cited definition of AfL is that by Klenowski (2009), which emphasised the key role of students in this process: “part of everyday practice by students, teachers and peers that seeks, reflects upon and responds to information from dialogue, demonstration and observation in ways that enhance ongoing learning” (p. 264). The five key AfL strategies include clarifying learning goals and criteria for success, gathering of a range of formal and informal assessment evidence, feedback to students, and peer and self-assessment (Wiliam & Thompson, 2008). These strategies should ultimately contribute to student self-regulation (Clark, 2012), and serve to “help students develop internal standards for their work, reflect upon it, and take ownership of learning” (Bennett, 2011, p. 9).

AfL is internationally recognised as good practice (Birenbaum et al., 2015; Torrance, 2012), but the strength and validity of evidence underpinning claims for its effectiveness have been criticised (Baird et al., 2014; Bennett, 2011; Torrance, 2012). As noted, one of the issues in relation to effectiveness relates to varying definitions and interpretations, with operationalisation of ideas underpinning AfL in practice accounting for much of the variation in effect sizes reported in research (Bennett, 2011; Wiliam, 2011). There appears to be general consensus in the literature that AfL has the potential to improve learning, yet many studies fail to operationalise AfL practices in a way that moves beyond a teacher-directed approach (DeLuca et al., 2019; Torrance, 2012).

When evaluating the translation of AfL principles to classroom practice, it is useful to make a distinction between implementation to the “letter” and in the “spirit” (James & McCormick, 2009; Marshall & Drummond, 2006). Implementation to the letter reflects the mechanistic adoption of strategies, without actively engaging students in their learning. For example, implementation of clarifying success criteria to the letter would involve teacher modelling of success criteria, with limited room for student input. In contrast, teachers who demonstrated the “spirit” of AfL used strategies in a way that encouraged student autonomy (James & McCormick, 2009; Marshall & Drummond, 2006). An example of implementation in the spirit involves co-construction of success criteria by students and the teacher, requiring students to actively create and apply notions of quality (Marshall & Drummond, 2006). Implementation of the spirit of AfL is thus consistent with student-centred feedback models (SCFM) (Boud & Molloy, 2013; Brooks et al., 2021; Van der Kleij et al., 2019; Winstone et al., 2017).

Although empirical evidence pertaining to the effectiveness of feedback is relatively stronger compared to other key AfL strategies (Hattie, 2009), the power of feedback hinges upon the successful development of students' evaluative expertise (Sadler, 2010) through the full range of AfL strategies. However, research suggests that substantial professional learning interventions may be required for teachers to realise such aspirations (DeLuca et al., 2019; James & McCormick, 2009; Marshall & Drummond, 2006).

Implementing feedback processes successfully is complex and requires teachers to develop considerable feedback literacy, which includes feedback knowledge and pedagogical skills (Carless & Winstone, 2020; Hattie et al., 2016). For example, teachers not only require deep understandings of the success criteria for assessment tasks but also understandings of the processes required for success and improvement. This knowledge then needs to be matched to the current performance levels of learners. Furthermore, consideration must be given to means of communicating and facilitating improvement-focused feedback. Various researchers have expressed concerns over levels of teacher and student assessment literacy (e.g., Carless & Winstone, 2020; Havnes et al., 2012). Teacher feedback literacy is fundamental to fostering student feedback literacy (Carless & Winstone, 2020), both of which are essential to “establishing an assessment for learning culture in the school” (Havnes et al., 2012, p. 26).

Furthermore, teacher support through generic principles and guidelines has been demonstrated to provide insufficient support to enable teachers to fully adopt AfL practices (Black et al., 2003; DeLuca et al., 2019; Van der Kleij et al., 2018). Numerous researchers have therefore emphasised the criticality of substantial professional learning support for teachers to realise the intended positive impacts on student achievement (e.g., Bennett, 2011; DeLuca et al., 2019; Heitink et al., 2016). A large number of studies have demonstrated the substantial benefits in-service teacher professional learning can have on student achievement (e.g., Lai & Schildkamp, 2016; Meissel et al., 2016; Parr et al., 2007). Yet, studies evaluating teacher professional learning interventions in AfL have demonstrated considerable variability in the quality of AfL implementation by teachers, resulting in variable student achievement gains (e.g., DeLuca et al., 2019; James & McCormick, 2009).

Research by Wiliam et al. (2004) found a substantial effect size of 0.32 in mathematics and science classrooms, following a six-month professional learning intervention in formative assessment. In their study, student achievement data were compared across intervention classes and comparison groups, where formative strategies were not an embedded feature of classroom practice. Wiliam and colleagues highlighted the variability in teacher practices as a result of the intervention, with different emphases on strategies relating to, for example, feedback, and sharing marking criteria with students. More recently, Andersson and Palm (2017) reported an even larger effect size of 0.66 in the context of mathematics following a teacher professional learning intervention, based on average student achievement gains within the participating teachers' classes compared to those in a control group. Similar to the Wiliam et al. (2004) study, Andersson and Palm (2017) reported implementation of a strategies “in different ways and to different degrees” (p. 96). Although these studies showed moderate to large effects on student achievement, it must be noted that the AfL approaches implemented by most teachers were still strongly teacher-led.

Research has highlighted that despite substantial professional learning efforts, the potential of feedback to enhance student learning is often not realised in classroom practice (Hattie et al., 2016). One possible explanation lies in difficulties experienced by teachers in shifting towards more student-centred feedback approaches as envisioned in AfL principles (DeLuca et al., 2019; Jonsson et al., 2015). Jonsson et al. (2015) called for more research to address the issue that “it is still not clear what actually separates effective AfL practices from less effective ones” (p. 107). Although there is solid theoretical support for the use of student-centred approaches to feedback, there is a lack of strong empirical evidence from classroom practice (Van der Kleij et al., 2019). This highlights the need for research into the impact of professional learning interventions to support effective feedback practices, and how such practices impact student achievement. Further, Graham et al. (2015) called for more research using quasi-experimental designs to strengthen the evidence base regarding feedback effectiveness for improving student writing achievement.

This study was part of a larger, funded project focusing on teacher professional learning in feedback in the context of English writing in Queensland, Australia. An initial study used teacher and school leader focus group interviews to investigate the extent to which the professional learning intervention using a new student-centred feedback model (SCFM) was perceived as effective in promoting classroom feedback practice that supports student self-regulation (Brooks et al., 2021). Results showed that participants perceived the intervention had caused notable changes in their thinking and feedback practices. Participants further reported substantial changes in levels of student self-regulation and positive effects on student writing achievement outcomes. However, a research design that enables analysis of student achievement data is required to empirically evaluate the impact of the professional learning intervention (described in detail in the section The Professional Learning Intervention).

The aim of this study was to examine the impact of a professional learning intervention underpinned by a new SCFM on student writing achievement using a quasi-experimental design. The research questions central to this study were:

• What was the effect of a professional learning intervention, underpinned by a new SCFM, on student writing achievement?

• Which feedback strategies from the professional learning intervention were associated with improvements in students writing achievement?

This study addresses an important gap in the literature, as it contributes insights into the effectiveness of feedback within a student-centred paradigm. By using a quasi-experimental design with pre-and post-assessment measures, the study addresses concerns over the strength of evidence in previous AfL studies (Baird et al., 2014; Bennett, 2011) and research on feedback effectiveness in writing (Graham et al., 2015). Further, this study aims to add more detailed insights into aspects of the intervention that were associated with student achievement gains, responding to Jonsson et al.’s (2015) call for more detailed studies examining AfL effectiveness.