Enhancing writing ability of Iranian EFL learners through learning-oriented assessment: peer and teacher feedback implementation

Abstract

Despite the widespread use of Learner-Oriented Assessment (LOA) in (English as a Foreign Language EFL classes, concerns persist regarding its suitability for contexts such as Iran. Drawing on Carless’s (SHE, 31:219-233, 2006) criteria for LOA, the present study examined the effectiveness of peer and teacher feedback in an academic writing course. To this aim, a total of 124 intermediate language learners who were homogenized in terms of language proficiency, based on the Common European Framework of Reference for Languages (CEFR), were randomly assigned to an experimental and a control group. A 12-session writing instruction course with specific writing tasks was developed. One group received feedback from peers, while the other group received feedback from the teacher. Findings revealed that peer feedback resulted in better writing in terms of language, content, and organization. It was also found that the peer feedback group could retain their learning in the long run. These findings underscore the effectiveness of peer feedback within the framework of LOA, particularly in teacher-directed settings.

Introduction

Learning-Oriented Assessment (LOA) is one of the recent issues in EFL writing instruction (Ishikawa, 2018). What sets LOA apart from other assessment approaches, is that students actively engage in the learning task without primarily being assessed for their linguistic competence (Babaii, 2019; Carless, 2015; Derakhshan & Ghiasvand, 2022). According to Carless (2007), this approach to assessment underscores students’ learning process rather than the assessment of language. In LOA, feedback and learning occur simultaneously. LOA includes three elements, namely task, feedback, and evaluation, of which feedback has received more attention (Carless, 2006). Although many studies have focused on teacher, peer, and self-feedback, there is no consensus regarding the effectiveness of one type of feedback in terms of its impact on organization, language, and content of writing. This lack of consensus may stem from the diverse methodologies employed by different studies, the variability in participant characteristics, and the multifaceted nature of writing proficiency involving various dimensions beyond mere linguistic accuracy (Zaman & Azad, 2012).

A review of the existing literature reveals doubts regarding the efficacy of teacher and peer feedback. Studies show that teacher feedback neglects to take note of the learning progress (Benson & DeKeyser, 2019; Lin, 2018) and fails to support learning (Ma & Bui, 2022). Additionally, teacher feedback is consistently challenged in supporting the retention of learning (Vasu et al., 2022) and may not sufficiently empower learners to develop lifelong learning habits, such as self-assessment capacity (Lee et al., 2019).

Previous studies have also critiqued the effectiveness of peer feedback, highlighting instances where learners receive incorrect, inappropriate, or unhelpful advice (Capstick, 2004). Additionally, concerns arise regarding the timing of peer feedback, with instances where feedback is provided either too early or too late in the instructional process (Carless, 2007; Khodi et al., 2021; Kim & Kim, 2021), thus posing challenges for learners in utilizing it effectively.

Collie et al. (2019) contend that a diverse array of feedback methods has the potential to facilitate some form of learning. However, the essential question lies in identifying the feedback approach that promotes optimal learning outcomes. In response to this inquiry, different feedback types have been examined. To add to the current literature, the present study compares the effectiveness of peer feedback as a component of LOA and teacher feedback on writing quality, with a specific focus on content, language, and organization.

Review of the literature

LOA offers an organized approach to collecting and utilizing evidence from homeroom activities and tests. It includes assessment at both macro and micro levels, emphasizing the importance of integrating assessment and teaching (Lam, 2020). The term LOA suggests that every level of assessment should contribute to enhancing learning outcomes and ensuring the trustworthiness of assessment outcomes (May et al., 2020). Cambridge English considers assessment from a holistic perspective where it occurs at multiple levels and in various forms (Hontvedt et al., 2023). This approach involves both the macro level of setting educational objectives and evaluating the outcomes and the micro level of individual learning interactions within and beyond the classroom. From this perspective, all levels of assessment can and should contribute to enhancing learning efficacy, ensuring accurate outcome evaluation. The Cambridge model aims to establish a meaningful connection with education by emphasizing quantitative assessment and a personalized learning. It recognizes the importance of using quantitative measures to evaluate learning outcomes while valuing the individualized and subjective aspects of education. This holistic approach implies a significant role for teachers in creating a conducive learning environment that aligns closely with the principles of LOA. As Ma et al. (2021) suggest, LOA predicts that teachers will play a central role in establishing such an environment, akin to their role in formal assessment practices.

According to Carless (2007), LOA is guided by three main principles: assessment task, evaluation, and feedback (see Fig. 1). Based on the LOA framework (Fig. 1), learners should be actively engaged in the process of evaluating their own performance and that of their peers.

Fig. 1

Framework for learning-oriented assessment (Carless, 2007)

LOA prioritizes learners’ active engagement in the assessment process, promoting self-reflection, self-assessment, and collaboration through peer feedback. Peer feedback, rooted in Vygotsky’s sociocultural theory, emphasizes the significant impact of social interaction on L2 development (Vygotsky & Cole, 1978). According to Vygotsky, learners’ cognitive development occurs primarily through social collaboration with peers, teachers, and guardians. A rich social environment fosters learner interaction, enhancing learning and development (Li et al., 2020). Although peer feedback serves as an effective instructional approach to enhance learners’ learning processes and outcomes (Li et al., 2020; Wang and Chang, 2021), insufficient attention has been given to the learning outcomes of peer feedback, particularly in the development of writing skills.

Peer and teacher feedback play complementary roles in the language learning process. Peer feedback fosters a collaborative atmosphere where students learn from each other. The advantages of peer feedback are reflection, enhanced time on homework, more focus on work, and increased accountability and responsibility of learners (Alavi et al., 2021; Harrison et al., 2015). Teacher feedback provides personalized guidance tailored to learners’ needs; teacher feedback provides a framework through which a teacher views, interprets, and interacts with the teaching environment (Chan & Luk, 2022; Rezaei & Barkaoui, 2021; Sanger & Osguthorpe, 2013). While some studies have demonstrated that involvement in peer feedback improves L2 writing (Kaufman & Schunn, 2011; Ketabi & Torabi, 2012; Zhao, 2014), others emphasize the significance of teacher feedback (Ruegg, 2018; Yang et al., 2006; Zhao, 2010). There are also studies that have found no difference between teacher and peer feedback in language learning contexts (e.g., Ahmadian et al., 2013; Mellati et al., 2022; Mowlaie, 2014). Therefore, the present study addresses the following questions:

• 1. Is there any statistically significant difference in the content, language, and organization of Iranian learners' writing who receive regular teacher feedback and those who receive LOA-based peer feedback?

• 2. Is there any statistically significant difference between the content, language, and organization of Iranian EFL learners' writing in immediate and delayed post-tests?

Method

Participants

A sample of 160 English EFL learners was recruited from a branch of the University of Applied Sciences and Technology and Islamic Azad University. The participants were selected based on their proficiency level, determined using the Oxford Placement Test (OPT). Only those classified as B1 level according to the Common European Framework of Reference (CEFR) were included in the study to maintain a homogeneous sample. The selection process resulted in a sample of 54 males and 68 females.

Instruments

The Oxford Placement Test (OPT) was administered to check the learners’ proficiency level. OPT is a standard test for assessing the language proficiency of non-native English speakers. The Cronbach’s alpha reliability coefficient, a measure of internal consistency, was 0.86 for the present study.

The pre-test, immediate, and delayed post-test included a writing task; the participants were required to write a minimum of 200 words in 30 min on different topics such as past memories, vacation destinations, hobbies, and traditions.

The criteria for providing a Learning-Oriented Assessment (LOA) were adopted from Carless et al.’s (2006) framework (see Fig. 1): (1) approaching assessment tasks as learning tasks, (2) offering the opportunity for the student’s involvement in the assessment process, and (3). focusing on timely, performance-based, and future-directed feedback (Carless, 2007).

Procedures

Initially, based on convenience sampling, 160 EFL learners participated in the study. Based on the OPT results, those who were classified in the B1 level (N = 124) were randomly assigned to peer feedback and teacher feedback groups.

In the pre-writing phase, the teacher activated students’ background knowledge (schemata) using pictures depicting everyday activities in order to lead the learners. Then, the teacher asked the participants to discuss topics such as childhood memories, tourist attractions, influential people, ideal vacation destinations, nature’s impact, joyful hobbies, fascinating history, overcoming challenges, and cultural traditions. In both groups, the teacher presented vocabulary items and the grammatical structures related to the topic on the board; this was followed by a gap-filling exercise to help learners express their ideas more effectively.

In the teacher feedback group, the treatment lasted for 12 sessions. In session one, participants sat for their pre-test; a descriptive writing task was scored based on guidelines provided by Carless et al. (2006). Session two introduced participants to the writing task and criteria for assessment. Sessions four and five focused on training participants in feedback delivery, including how to use the grading rubric and offer constructive feedback. In sessions five to nine, participants were required to complete and submit their tasks. During this phase, participants assessed their peers’ drafts using the grading rubric and provided written feedback on content, language, and organization. Additionally, they reflected on the feedback received from peers and revised their work for final submission.

Participants in the teacher feedback group followed a similar session structure but they did not engage in peer feedback during the drafting process. Instead, they received feedback solely from the teacher after submitting their drafts.

In the post-writing phase, both groups received feedback based on Carless et al.’s (2006) framework on the content, language, and organization. Following the submissions of the drafts, both groups sat for an immediate post-test. The immediate post-test, a writing task with a 45-min time, was conducted in Session 10. Participants in the two groups received no feedback between immediate post-tests and delayed post-tests. Two weeks later, in Session 12, both groups were requested to write on a topic; this served as a delayed post-test to evaluate the long-term effectiveness of the treatment.

Results

The main goal of the present study was to explore potential variations in the content, language, and organization of the writings of EFL learners who received feedback from peers and teachers in writing classes. Pre-test scores were compared for potential differences at the outset of the study. Descriptive statistics indicated that there were no significant differences between the two groups in terms of content, language, and organization (CLO) before the treatment. This initial homogeneity between the groups establishes a baseline for evaluating the impact of the feedback interventions on writing quality. The descriptive statistics of participants’ pre-test scores appear in Table 1.

Table 1 Descriptive statistics of pre-test scores for content, language, and organization (CLO) in peer feedback and teacher feedback groups (n = 62)

Pre-test scores for content, language, and organization showed similar mean values in both groups, indicating comparable writing proficiency levels before the intervention. This ensures a fair comparison between the groups when evaluating the effectiveness of the intervention in enhancing writing quality.

The peer feedback group exhibited a mean pre-test content score of 2.23 (SD = 0.68), a mean pre-test language score of 2.15 (SD = 0.59), and a mean pre-test organization score of 2.55 (SD = 0.69). Similarly, the teacher feedback group demonstrated mean pre-test scores of 2.42 (SD = 0.64) for content, 2.31 (SD = 0.58) for language, and 2.42 (SD = 0.52) for organization. Participants’ mean scores suggested that the teacher feedback group had a better performance on the pre-test.

Levene’s test (Table 2) for the equality of variances suggested that variances were not statistically different for the content and language dimensions (p > 0.05), but statistically different for organization (p < 0.05). An independent samples t-test was conducted to examine whether the differences between the mean scores of the experimental and control groups were statistically significant (Table 2). The findings suggest that both groups had comparable performance in terms of content (t (122) =  − 1.62, p = 0.10), language (t (122) =  − 1.51, p = 0.13), and organization (t (122) = 1.16, p = 0.24) in the pre-test.

Table 2 Independent samples t-test results examining the differences in content, language, and organization in peer feedback and teacher feedback groups in the pre-test

Table 3 presents the descriptive statistics for immediate post-test scores in content, language, and organization (CLO) among participants of the two groups. As Table 3 shows, the mean post-test content score was 4.76 (SD = 0.50), the mean post-test language score was 4.50 (SD = 0.64), and the mean post-test organization score was 4.42 (SD = 0.58) for the peer feedback group, indicating a moderate to high level of intervention effect. Table 3 also demonstrates that the mean post-test content score was 2.60 (SD = 0.75), the mean post-test language score was 2.68 (SD = 0.74), and the mean post-test organization score was 2.76 (SD = 0.64) for the teacher feedback group.

Table 3 Descriptive statistics of post-test scores for content, language, and organization (CLO) in peer feedback and teacher feedback groups (n = 62)in the immediate post-test

An independent samples t-test was run to see if the differences between the mean scores of the two groups were statistically significant in content, language, and organization. Table 4 displays the results.

Table 4 Independent samples t-test results examining the differences in content, language, and organization in peer feedback and teacher feedback groups in the immediate post-test

As Table 4 indicates, Levenes’ test of equality of variances indicated the assumption of equality of variances was met for language (p = 0.49) and organization (p = 0.92), but not for content (p = 0.00). Therefore, for the dimension of content, the equal variances not assumed row was examined. Analysis of the post-test scores (Table 4) reflected statistically significant differences in writing quality across the three dimensions of content (t (105.98) = 18.74, p = 0.00), language (t (122) = 14.58, p = 0.00), and organization (t (122) = 14.98, p = 0.00). between the teacher feedback and the peer feedback group, highlighting the effectiveness of the intervention in enhancing participants' writing skills.

The descriptive statistics for the peer feedback and teacher feedback groups’ delayed post-test scores across three groups are presented in Table 5. As for the peer feedback group, the mean delayed post-test score for content was 4.77 (SD = 0.52), for language was 4.61 (SD = 0.61), and for organization was 4.40 (SD = 0.66). For the teacher  feedback group, the mean of delayed post-test score for content was 2.39 (SD = 0.61), for language it was 2.31 (SD = 0.61), and for organization it was 2.42 (SD = 0.52).

Table 5 Descriptive statistics of delayed post-test scores for content, language, and organization (CLO) in peer feedback and teacher feedback groups (n = 62) in the delayed post-test

As the two groups’ mean scores in Table 5 suggest, the learners who received peer feedback outperformed those who received teacher feedback in all three dimensions of content, language, and organization on the delayed post-test. To find out whether these differences were statistically significant in delayed post-test scores for content, language, and organization among participants who received different types of feedback, an independent samples t test was used (Table 6).

Table 6 Independent samples t-test results examining the differences in content, language, and organization in the peer feedback and teacher feedback groups in delayed post-test

According to Table 6, Levene’s test suggested unequal variances for the component of content (p = 0.01) and equal variances for language (p = 0.82) and organization (p = 0.10). Table 6 demonstrates that the type of feedback had a statistically significant effect on delayed post-test in content (t (119.36) = 23.34, p = 0.00), language (t (122) = 20.93, p = 0.00), and organization (t (122) = 18.38, p = 0.00) between the teacher feedback and the peer feedback group, indicating the superiority of peer feedback on delayed post-test. The fact that the peer feedback group outscored the teacher feedback group on both immediate and delayed post-tests underscores the superior effectiveness of LOA-based peer feedback interventions in enhancing participants’ writing skills across the three dimensions of Content Language and Organization.

The mean scores on the dependent variables were compared for the same group of students at two different points in time (pre-test vs. post-test and pre-test vs. delayed post-test) using paired samples t-test in order to gain better insights regarding the impact of treatment and see whether the performance of the participants improved from pre- to post tests.

In conclusion, the findings indicate that peer feedback plays a more significant role than teacher feedback in enhancing participants’ writing quality across the three dimensions of content, language, and organization. The results indicate that participants who received peer feedback significantly outperformed those who received teacher feedback in all aspects of writing proficiency during the immediate and delayed post-test phase.

The results of paired samples t-tests of the peer feedback group performance from pre-test to post-test to delayed post-test are reported in Table 7. As seen, there is a statistically significant difference between pre-test and post-test performance of the peer feedback group participants across the three dimensions of content (t(61) =  − 23.10, p = 0.00), language (t(61) =  − 19.31, p = 0.00) and organization (t(61) =  − 14.98, p = 0.00). Comparisons also revealed significant improvements in the performance of the peer feedback group participants from pre-test to the delayed post-test across the three dimensions of content (t(61) =  − 22.77, p = 0.00), language (t(61) =  − 21.16, p = 0.00) and organization (t(61) =  − 13.66, p = 0.00). Examining the performance of the peer feedback group from post-test to delayed post-test suggested no significant difference between the post-test and delayed post-test of the learners with regards to content (t(61) =  − 0.37, p = 0.70) and organization (t(61) = 0.22, p = 0.82). Results also demonstrated that the peer feedback group’s performance in terms of language (t(61) =  − 2.17, p = 0.03) improved significantly from post-test to delayed post-test.

Table 7 Paired samples test of peer feedback group’s test scores across the three dimensions of content, language. and organization

Table 8 displays the pairwise comparison of teacher feedback group’s performance from pre-test to the delayed post-test.

Table 8 Paired samples t-test of teacher feedback group’s test scores across the three dimensions of content, language, and organization

As Table 8 indicates, teacher feedback also resulted in a statistically significant promotion in the writing ability of the participants from pre-test to post-test. Pairwise comparison of pretest to post-test scores revealed statistically significant differences across the three dimensions of content (t(61) =  − 2.02, p = 0.00), language (t(61) =  − 4.13, p = 0.00), and organization (t(61) =  − 3.68, p = 0.00).

Results also demonstrated that the teacher feedback group’s performance significantly declined from post-test to delayed post-test across the three dimensions of content (t(61) = 2.20, p = 0.03), language (t(61) = 4.81, p = 0.00) and organization (t(61) = 3.57, p = 0.00). Results also suggested no statistically significant difference between the pretest and delayed post-test performance of the teacher feedback group across the three dimensions of content (t(61) = 0.70, p = 0.48), language (t(61) = 0.00, p = 1.00), and organization (t(61) = 0.00, p = 1.00).

Discussion

The present study explored the effect of different feedback approaches on the content, language, and organization of writing between learners who received regular teacher feedback and those who engaged in learner-oriented peer feedback practices. It also examined the differences in the content, language, and organization of writing in the immediate and delayed post-tests.

In investigating whether there is any statistically significant difference in the content, language, and organization of Iranian learners’ writing who received regular teacher feedback and those who received LOA-based peer feedback, the findings pointed to the positive effects of peer feedback on the content, language, and organization of writing. The results suggest that peer feedback can support learners’ development, particularly when personalized teacher feedback is limited. Participants who received peer feedback consistently outperformed those who received teacher feedback on immediate and delayed post-tests; this points to the longitudinal effects of peer feedback on writing development among EFL learners. The superior performance of the peer feedback group is in line with Chiu et al.’s (2021) finding which revealed the long-term impacts of peer feedback on learners’ writing performance. They argued that peer feedback can enhance learners’ motivation and self-regulated learning, and have positive impacts on long-term retention.

The results provide evidence of the positive and lasting effects of peer feedback on enhancing writing development among EFL learners. The findings highlight the importance of implementing learner-oriented approaches, such as peer feedback interventions, in writing instruction in order to foster continual improvement in language learners. The results align with prior research emphasizing the benefits of peer feedback interventions in enhancing writing proficiency (Ahmadian et al., 2013; Ganji, 2009; Ketabi & Torabi, 2012; Taghizadeh & Zafarpour, 2022). To optimize the effectiveness of peer feedback, teachers are encouraged to provide students with brief lessons on how to offer constructive feedback (Zhang, 1995). Such instructional strategies, which promote a more holistic and student-centered approach to feedback provision, align with the goals of learner-oriented assessment in writing instruction.

Despite its advantages, peer feedback presents several potential disadvantages. Firstly, concerns regarding its negative impact on interpersonal relationships may hinder the assessment process. Peer feedback inherently relies on a social and cooperative process where learners evaluate each other (Planas Lladó et al., 2014). When students are less worried about how their peers respond to their feedback, they are more likely to actively participate in the process, enhancing its effectiveness. Another limitation of peer feedback is procedural rationality, a concept from organizational behavior (March & Simon, 1958), which refers to decision-making based on procedural rules rather than individual needs or context. In peer feedback, procedural rationality may lead to rigid adherence to preset feedback formats, hindering personalized and contextually relevant feedback. In contexts where teachers are traditionally viewed as the authority and primary facilitators of the learning process, students may be reluctant to assume the responsibility of evaluation (Karami & Khodi, 2021; Sadeghi & Khezrlou, 2014). Evaluative responsibilities, such as peer assessment or self-assessment, require active participation, and reluctance can lead to a more passive learning experience. Reluctance to engage in evaluation can limit the effectiveness of peer learning environments, as constructive critiques and discussions are less likely to occur. Additionally, concerns about the validity of peer feedback may emerge, promoting doubts about its accuracy and fairness among students (Kaufman & Schunn, 2011; Mellati et al., 2022). Such uncertainties may undermine students' motivation to fully engage in the peer feedback process. Previous experiences with peer feedback can also influence students' willingness to participate in the feedback provision process. Negative past experiences may discourage students from actively engaging in peer feedback activities, posing a challenge to its implementation. Therefore, while peer feedback offers numerous benefits, addressing these potential drawbacks is essential to ensure its effectiveness in promoting learning and skill development.

Conclusion

This study underscores the effect of combining assessment practices with regular writing instruction to enhance learners' writing ability. The findings highlight the effectiveness of implementing feedback within learner-oriented approaches (LOA) in writing classes, demonstrating significant improvements in language learners’ writing quality. Specifically, our study indicates that peer feedback may be a more suitable strategy than teacher feedback for L2 writing development due to its facilitation of collaborative learning environments and reciprocal learning processes. Peer feedback can create a collaborative atmosphere where students share diverse perspectives and insights while encouraging critical self-evaluation. Furthermore, the reciprocal exchange involved in peer feedback deepens understanding of writing principles, and promotes empathy, and respect for differing viewpoints among learners, thereby fostering a supportive learning environment conducive to the development of writing proficiency. The demonstrated success of peer feedback within the LOA framework suggests a shift towards more student-centered instructional practices.

The findings reinforce the importance of incorporating diverse feedback mechanisms into educational practices to support language learning and skill development. Language teachers can consider integrating peer feedback more prominently into writing instruction based on the observed effectiveness. This might involve structured peer review activities or collaborative writing tasks that promote peer interaction and feedback exchange. Based on the findings teachers are advised to adopt a flexible approach to feedback provision, leveraging the strengths of both teacher and peer feedback to create a supportive learning environment.

Building upon these findings, future research could investigate the specific mechanisms underlying the effectiveness of peer feedback, focusing on the types of peer feedback provided and their impact on writing improvement. Additionally, exploring the influence of individual learner characteristics, such as language proficiency level and learning style preferences, on the effectiveness of peer feedback interventions could deepen our understanding of how feedback leads to writing development. Examining the interactional dynamics between teachers and students during feedback sessions, and examining the role of factors such as student engagement, could inform the design of more effective feedback practices. This can be achieved through the use of qualitative research methods, such as interviews and classroom observations which could provide valuable insights into the nuances of feedback interactions.