Enhancing Learning Through Peer Assessment in Multilingual English-Medium Instruction: A Study at SEEU in North Macedonia

Abstract

This study investigates the effectiveness of peer assessment as a formative learning strategy in multilingual English-Medium Instruction (EMI) higher education. Although peer assessment is widely associated with enhanced learner engagement, self-regulation, and feedback literacy, its implementation in linguistically diverse EMI settings remains insufficiently explored. Addressing this gap, the study examines students’ perceptions of peer assessment, the degree of alignment between peer and instructor evaluations, and students’ reflective experiences in a multilingual university context. A mixed-methods design was employed with 60 undergraduate students enrolled in EMI courses at South East European University in North Macedonia. Quantitative data were collected through a structured questionnaire and peer-assessment rubric, while qualitative data were obtained from reflective interviews with 37 students. Quantitative data were analyzed using descriptive statistics, Pearson correlation analysis, and paired-samples t-tests, whereas qualitative responses were examined through thematic analysis. The findings reveal generally positive student perceptions of peer assessment, particularly regarding self-reflection, critical thinking, collaboration, and confidence in evaluating academic work. Significant positive correlations among key dimensions of feedback literacy suggest that peer assessment supports interconnected cognitive and metacognitive learning processes. Although a statistically significant difference emerged between peer and instructor scores, overall agreement was moderate, with students tending to assign slightly lower marks than instructors. Qualitative findings further indicate that peer assessment enhanced students’ understanding of assessment criteria and learning processes while also exposing challenges related to language proficiency, emotional discomfort, and concerns about feedback accuracy. The study demonstrates that peer assessment can be an effective pedagogical approach in multilingual EMI classrooms when supported by clear assessment criteria, structured guidance, and feedback training. These findings contribute to research on feedback literacy, formative assessment, and multilingual learning in higher education.

1. Introduction

In recent years, higher education has increasingly adopted student-centered pedagogies aimed at promoting learner autonomy, critical thinking, and active engagement. Within this shift, peer assessment has emerged as a widely used formative strategy that enables students to evaluate and learn from one another’s work. Based on Sortwell et al. (2024), formative assessment is understood as a continuous and interactive process in which evidence of student learning is collected and used to inform teaching practices and support students’ ongoing learning, self-regulation, and academic development. In English-Medium Instruction (EMI) and multilingual contexts, peer assessment is particularly relevant, as it supports both disciplinary learning and academic language development (Topping, 1998; Falchikov & Goldfinch, 2000). Research shows that peer assessment can enhance writing development, metacognitive awareness, and learner autonomy, with students often benefiting cognitively from the act of evaluating peers (Hyland & Hyland, 2006; Lundstrom & Baker, 2009). However, its effectiveness is not unproblematic. Challenges include limited linguistic proficiency, reduced accuracy of feedback, and socio-affective barriers such as reluctance to critique peers (Nelson & Murphy, 1993; Hu, 2019). These issues are especially relevant in EMI classrooms, where students must manage both content learning and a non-native language, often leading to increased cognitive and linguistic load (Dearden, 2015; Macaro et al., 2018).

At the same time, multilingual classrooms offer opportunities for richer interaction and meaning-making, as students draw on diverse linguistic resources during learning (García & Wei, 2014). Nevertheless, research on peer assessment in multilingual EMI higher education remains limited, particularly in underexplored regions such as Southeastern Europe.

Recent scholarship has increasingly focused on feedback literacy, defined as students’ ability to interpret, evaluate, and use feedback for learning. Studies show that peer feedback can contribute to developing this competence, although its impact depends on structured support and does not always translate into effective feedback use or emotional regulation (Zhan et al., 2025). In addition, evidence suggests a persistent trust gap, as students often perceive teacher feedback as more reliable than peer assessment, raising questions about the perceived validity of peer assessment in higher education (Ardill, 2025; Wang & Liu, 2024).

Furthermore, more recent EMI studies confirm that peer assessment can improve evaluative judgment, engagement, and disciplinary understanding when implemented within structured formative frameworks (Li & Zhang, 2023; Yu & Lee, 2024). However, its effectiveness is strongly influenced by contextual factors such as language proficiency, rubric clarity, and teacher scaffolding (García-Mayo & Hidalgo, 2023; Fleckney et al., 2024). This reinforces the view that peer assessment is most effective when embedded in well-designed instructional systems rather than used as an isolated activity. Despite this growing body of research, three key gaps remain. First, limited attention has been given to how feedback literacy develops in multilingual EMI classrooms. Second, findings on the reliability and fairness of peer assessment remain mixed. Third, there is insufficient evidence comparing traditional peer assessment with emerging AI-supported feedback systems in multilingual higher education contexts.

Addressing these gaps, this study investigates peer assessment in a multilingual EMI setting in Southeastern Europe, combining student perceptions, assessment reliability, and qualitative experiences. In doing so, it contributes to a more integrated understanding of peer assessment as both a pedagogical and evaluative practice shaped by linguistic diversity, feedback literacy, and socio-cultural classroom dynamics.

This study contributes to the existing literature by providing one of the first empirical investigations of peer assessment within a multilingual English-Medium Instruction (EMI) context in South-Eastern Europe, a region that remains underrepresented in current research. Unlike prior studies that typically examine either perceptions or assessment outcomes in isolation, this study adopts a comprehensive mixed-methods approach that integrates (1) students’ perceptions of peer assessment, (2) the comparative reliability of peer- and instructor-awarded scores, and (3) the influence of multilingual classroom dynamics on feedback practices. By combining these dimensions, the study offers a more holistic understanding of how peer assessment functions both as a pedagogical tool and as an evaluative mechanism in linguistically diverse EMI settings. Furthermore, it advances the field by highlighting the role of trust, feedback literacy, and socio-cultural factors in shaping peer assessment practices, thereby extending existing theoretical and empirical work beyond traditionally studied monolingual or Western contexts.

Building upon recent developments in formative assessment research, this study adopts feedback literacy (Carless & Boud, 2018) as its primary theoretical framework. Feedback literacy refers to learners’ capacity to appreciate, interpret, evaluate, and effectively use feedback to improve future learning. Within multilingual English-Medium Instruction (EMI) contexts, students must simultaneously negotiate disciplinary knowledge, academic language, and evaluative judgment, making feedback literacy particularly relevant. In this perspective, peer assessment is conceptualized not merely as an assessment technique but as a formative process that supports evaluative judgment, self-reflection, and learner autonomy. Multilingualism, together with socio-affective factors such as trust, confidence, and anxiety, is viewed as influencing students’ engagement with feedback practices. Accordingly, the present study investigates peer assessment through an integrated feedback literacy perspective to better understand its pedagogical value in multilingual higher education.

Conceptual Framework

This study is grounded in feedback literacy as the primary theoretical framework (Carless & Boud, 2018). Feedback literacy refers to students’ capacity to understand assessment criteria, make evaluative judgments, engage constructively with feedback, and use feedback to improve future learning. Within this perspective, peer assessment is conceptualised as a formative pedagogical practice that can support the development of these capacities by engaging students in evaluating the work of others and reflecting on quality standards. In multilingual English-Medium Instruction (EMI) contexts, peer assessment does not operate in isolation but is shaped by contextual and socio-affective conditions. Accordingly, this study considers multilingualism, trust, and anxiety as key contextual factors that may influence how students participate in and benefit from peer assessment processes. Multilingual classroom environments require students to negotiate disciplinary content alongside additional linguistic demands, which may affect both the production and interpretation of feedback. At the same time, socio-affective dimensions such as trust in peers’ evaluative competence and anxiety related to peer judgment may shape students’ willingness to engage critically and productively in assessment activities. The conceptual framework (Figure 1) positions feedback literacy as the central construct, with peer assessment functioning as a formative mechanism through which it may be developed. Contextual factors, particularly multilingualism and socio-affective variables, are assumed to influence the quality and effectiveness of engagement with peer feedback rather than directly determining learning outcomes.

This framework guided the design of the study and the formulation of the research questions. The quantitative instrument captured dimensions associated with feedback literacy, including self-reflection, evaluative confidence, collaboration, critical thinking, and trust. The peer assessment rubric enabled comparison between peer and instructor judgments as an indicator of evaluative alignment. In addition, qualitative data provided insights into how students experienced peer assessment in relation to language use, trust, and anxiety in EMI classrooms. Importantly, while multilingualism is recognised as a significant contextual factor in EMI environments, this study does not assume that assessment cognition is inherently language-dependent. Instead, language is conceptualised as a mediating condition that can shape how clearly students express evaluative judgments, interpret feedback, and perceive the credibility of peer assessment.

2. Materials and Methods

This study adopted a mixed-methods research design to investigate peer assessment practices in multilingual English-Medium Instruction (EMI) contexts. A convergent triangulation approach was employed, integrating quantitative and qualitative data to enhance the validity and depth of interpretation. The design combined three data sources: a structured student survey, a performance-based peer assessment rubric comparing peer and instructor evaluations, and written reflective responses capturing students’ experiential accounts. This multi-source design enabled methodological triangulation by examining peer assessment from perceptual, behavioural, and reflective perspectives. The study was conducted during the summer semester of 2026 at South East European University (SEEU), Tetovo, North Macedonia, within regular EMI course delivery.

2.1. Research Questions

This mixed-methods study was guided by the following research questions:

• RQ1: What are students’ perceptions and experiences of peer assessment in multilingual EMI classrooms?
• RQ2: What relationships exist among perceived learning dimensions (self-reflection, critical thinking, confidence, collaboration, and trust)?
• RQ3: To what extent do peer assessment scores align with instructor assessment scores?
• RQ4: How do students describe the benefits and challenges of peer feedback in multilingual EMI contexts?

2.2. Participants

Participants were undergraduate students enrolled in EMI courses at the Faculty of Languages, Cultures and Communication at SEEU. Three datasets were collected. The survey phase included 60 students representing the full cohort, providing comprehensive perceptual data on peer assessment. A subsample of 22 students participated in the peer assessment rubric component based on completion of an oral presentation task in the Applied Linguistics 2 course. In addition, 37 students participated in the qualitative phase through voluntary written reflective responses. All participants were enrolled in the researcher’s courses at the time of data collection, and peer assessment activities formed part of regular instructional practice, ensuring ecological validity. Participation was voluntary, and students could withdraw at any time without academic consequences.

2.3. Sampling Strategy

The study employed a purposive and task-embedded sampling strategy aligned with authentic classroom activities rather than random selection. The survey sample included the full cohort of 60 students, providing comprehensive coverage of perceptions. The rubric subsample consisted of 22 students selected based on completion of the assessed oral presentation task, forming a convenience and task-specific group. The qualitative subsample included 37 students who voluntarily provided reflective responses after completing peer assessment activities. Although subsample sizes differed, each dataset served a distinct analytical purpose, and the overall design supported methodological triangulation across quantitative and qualitative dimensions.

2.4. Instruments and Data Collection

Data were collected using three complementary instruments. First, a structured questionnaire (Appendix A) was administered online using Google Forms (Google LLC, Mountain View, CA, USA) to examine students’ perceptions of peer assessment. The instrument was developed based on formative assessment theory and peer assessment theory (Black & Wiliam, 1998) and included Likert-scale and dichotomous items addressing feedback experiences, perceived learning benefits, and challenges. The questionnaire was piloted with 12 EMI students and reviewed by two experts in applied linguistics and educational assessment to ensure validity. Internal consistency was acceptable (Cronbach’s alpha = 0.84). Data were analysed using IBM SPSS Statistics Version 27.0 (IBM Corp., Armonk, NY, USA), employing descriptive statistics and Pearson correlation analysis.

Second, a peer assessment rubric (Appendix B) was used during oral presentations in the Applied Linguistics 2 course. Both peers and the instructor used identical criteria, including content quality, organization, language use, critical thinking, participation, and task completion, rated on a five-point scale. The rubric was designed based on course learning outcomes and instructional objectives.

Third, qualitative data were collected through structured written reflective prompts (Appendix C), administered electronically after peer assessment activities. The data were analysed using thematic analysis following Braun and Clarke’s (2006) framework, including data familiarisation, coding, theme development, iterative review, and final refinement. Coding reliability was ensured through collaborative discussion among the research team.

2.5. Ethical Considerations

The study followed institutional ethical guidelines at SEEU. Participation was voluntary, and students were informed about the purpose of the study prior to data collection. Participation or non-participation had no impact on academic evaluation. All data were collected anonymously, and no personally identifiable information was recorded. Data were used exclusively for research purposes. According to institutional regulations, formal ethical approval was not required for classroom-based educational research involving routine instructional activities.

2.6. Data Analysis

Quantitative data were analysed using descriptive statistics (means and standard deviations), Pearson correlation analysis, and paired-samples t-tests. Qualitative data were analysed thematically to identify recurring patterns in students’ reflective responses. Integration of quantitative and qualitative findings enabled triangulation and a comprehensive interpretation of peer assessment processes in multilingual EMI contexts.

3. Results

3.1. Student Questionnaire

The survey findings offer a comprehensive overview of students’ experiences, perceptions, and perceived challenges associated with peer assessment practices. Descriptive statistical analyses were conducted to examine participants’ responses across all survey items. Specifically, mean scores were calculated to determine the central tendency of responses, while standard deviation values were used to assess the degree of variability among participants. Mean values approaching 5 indicate stronger levels of agreement and more positive perceptions of peer assessment, whereas values approaching 1 reflect stronger disagreement and less favorable attitudes. A total of 60 students participated in this phase of the study. In addition, Pearson correlation analysis (

Table 1. General Peer Assessment Experience.

Item	Mean/SD
1. How often did you participate in peer assessment activities during this course?	2.75 (0.82)
2. How clear were the assessment criteria/rubrics used in peer assessment?	3.9 (0.78)
3. How easy was it to use the peer assessment rubric?	3.48 (0.73)
4. Did you receive training or explanation on how to give feedback?	Yes/(68.0%)
	No/(32.0%)
5. If yes, how useful was the training?	3.19 (1.01)

was used to demonstrate the strength and direction of relationships among selected perception variables related to feedback literacy, confidence, trust, and perceived learning benefits).

The results presented in Table 1 indicate that students generally had positive experiences with peer assessment activities. A discrepancy was identified in the reporting of Item 1. The correct mean score for Item 1 is M = 2.75 (SD = 0.78), which reflects a moderate frequency of participation in peer assessment activities. The previously stated value of 0.82 was incorrect and does not correspond to the reported data. All interpretations have been revised accordingly to ensure consistency between the descriptive statistics and their explanation. The findings further show that students perceived the assessment criteria and rubrics as relatively clear (M = 3.90, SD = 0.78) and found the peer assessment rubric reasonably easy to use (M = 3.48, SD = 0.73). Although not all students received formal guidance on how to provide feedback, Item 4 is a dichotomous variable (Yes/No) and is reported using frequencies and percentages. Among those students, the usefulness of the training was evaluated positively (M = 3.19, SD = 1.01).

Overall, the relatively low standard deviation values suggest moderate consistency in students’ responses, indicating that participants shared similar views regarding their experiences with peer assessment.

The findings in

Table 2. Perceptions of Peer Assessment.

Item	Mean	Standard Deviation
6. Peer assessment helped me better understand assessment criteria.	3.43	0.87
7. Peer assessment improved my critical thinking skills.	3.67	0.88
8. Peer assessment encouraged me to reflect on my own work.	3.87	0.89
9. Receiving feedback from classmates improved my work.	3.83	0.83
10. Peer assessment increased my confidence in evaluating academic work.	3.71	0.95
11. Peer assessment improved collaboration among students.	3.83	0.81
12. I trusted the scores given by my classmates.	3.34	0.84
13. I believe peer assessment should be included in future courses.	3.69	0.88

indicate that students generally held positive perceptions of peer assessment. The highest mean score was reported for the statement that peer assessment encouraged students to reflect on their own work (M = 3.87, SD = 0.89), suggesting that reflective learning was one of the most valued outcomes of the process. Students also agreed that receiving feedback from classmates improved their work (M = 3.83, SD = 0.83) and that peer assessment enhanced collaboration among students (M = 3.83, SD = 0.81). Similarly, students perceived peer assessment as beneficial for improving critical thinking skills (M = 3.67, SD = 0.88) and increasing confidence in evaluating academic work (M = 3.71, SD = 0.95).

Although students generally viewed peer assessment positively, trust in the scores assigned by classmates received the lowest mean score (M = 3.34, SD = 0.84), indicating some reservations about the fairness or accuracy of peer evaluations. Nevertheless, the relatively high mean score for including peer assessment in future courses (M = 3.69, SD = 0.88) suggests that students largely supported its continued use in higher education settings.

Table 3. Challenges of Peer Assessment.

Item	Mean	Standard Deviation
14. I felt nervous when evaluating my classmates.	2.9	1.11
15. Language differences made peer assessment more difficult.	2.9	0.96
16. Some students gave very limited or unclear feedback.	3.27	0.98
17. I preferred teacher assessment over peer assessment.	3.36	0.89
18. I think some students were too generous in their scoring.	3.32	0.78
19. I think some students were too strict in their scoring.	2.81	0.73

illustrates students’ responses which show moderate levels of nervousness when evaluating classmates (M = 2.90, SD = 1.11), indicating that some students felt uncomfortable giving feedback to peers. Language differences were not viewed as a major barrier (M = 2.90, SD = 0.96), suggesting that communication issues were relatively limited in the peer assessment process. Students moderately agreed that some peers provided limited or unclear feedback (M = 3.27, SD = 0.98), which indicates a need for clearer guidance and stronger scaffolding on how to give constructive comments. There was also some preference for teacher assessment over peer assessment (M = 3.36, SD = 0.89), implying that students may still perceive teacher feedback as more reliable. Students also believed that some classmates were too generous in their scoring (M = 3.32, SD = 0.78), while fewer believed that classmates were too strict (M = 2.81, SD = 0.73). This suggests that leniency may be a more common issue than harshness in peer assessment practices.

The survey results indicate that students generally held positive attitudes toward peer assessment and recognized its value in supporting learning. Participants reported that peer assessment helped them better understand assessment criteria, reflect on their own work, improve collaboration, and develop critical thinking skills. Students also perceived peer feedback as beneficial for improving the quality of their work and expressed support for the inclusion of peer assessment in future courses. At the same time, the findings reveal several challenges that may limit the effectiveness of peer assessment. Some students experienced nervousness when evaluating their classmates, questioned the fairness and reliability of peer scores, and continued to prefer teacher assessment. Concerns were also raised regarding unclear feedback and overly generous scoring by some peers. Overall, the findings suggest that peer assessment can be an effective pedagogical practice when supported by clear rubrics, training, and structured guidance. Strengthening students’ feedback skills and increasing transparency in assessment procedures may further enhance trust in the process and improve the overall effectiveness of peer assessment activities.

In addition to descriptive statistics, the Pearson correlation analysis was conducted to explore relationships among key perceived dimensions of peer assessment, rather than to test causality or a structural model.

Table 4. Person Correlation Analysis.

Variables	1	2	3	4	5	6	7	8
1. Understanding assessment criteria	1
2. Critical thinking improvement	0.75	1
3. Self-reflection improvement	0.77	0.79	1
4. Receiving feedback improved work	0.43	0.58	0.45	1
5. Confidence in evaluating academic work	0.70	0.67	0.64	0.49	1
6. Collaboration among students	0.41	0.45	0.49	0.45	0.34	1
7. Trust in classmates’ scores	0.38	0.24	0.21	0.42	0.28	0.39	1
8. Support for future inclusion of peer assessment	0.33	0.34	0.41	0.48	0.33	0.58	0.54	1

presents the Pearson correlation coefficients among the key dimensions of students’ peer assessment experiences. The results indicate generally positive associations between all variables, although the strength of these relationships varies. Strong correlations are observed between understanding assessment criteria, critical thinking improvement, and self-reflection (r = 0.75–0.79), suggesting that these aspects of feedback literacy tend to co-occur in students’ perceptions. Moderate correlations are also evident between these core evaluative dimensions and confidence in evaluating academic work, indicating that students who report better understanding and reflective engagement also tend to feel more confident in assessment tasks.

In contrast, collaboration, trust in classmates’ scores, and support for future inclusion of peer assessment show weaker but still positive relationships with the other constructs (r = 0.21–0.58). These findings suggest that while affective and social dimensions such as collaboration and trust are related to cognitive aspects of peer assessment, they are more loosely connected.

To conclude, the correlations should be interpreted as associations among perceived constructs rather than evidence of causality. They provide an exploratory overview of how different dimensions of peer assessment experience relate to one another within the present sample.

3.2. Peer Assessment Rubric Results

The second instrument used in this study was a peer assessment form, completed by both the teacher and the students. A total of 22 forms were used with students at SEE University. This instrument helped to better understand the effectiveness of peer assessment. Comparing students’ evaluations with instructor evaluations helped determine whether students were able to assess their classmates fairly and accurately. In this study, both peer assessment and teacher-assessment rubrics were used to evaluate similar criteria, including content quality, organization, language use, critical thinking, participation, and task completion. By comparing the two sets of scores, it became possible to identify patterns, similarities, and differences in how students and instructors perceived academic performance.

Students assess their classmates using the following criteria on a 1–5 scale:

Table 5. Student Peer Assessment Rubric Results.

Criterion	Description	Scale
Content Quality	Ideas are clear, relevant, and well-developed.	1–5
Organization	Work is logically structured and easy to follow.	1–5
Language Accuracy	Grammar, vocabulary, and spelling are accurate.	1–5
Critical Thinking	Shows analysis, reflection, and original thinking.	1–5
Participation	Student actively contributed to the task.	1–5

presents the student assessment rubric used in the study. The rubric includes five criteria: content quality, organization, language accuracy, critical thinking, and participation. Each criterion was rated on a scale from 1 to 5. This rubric provided students with a clear framework for evaluating their classmates’ work fairly and consistently.

Table 5 highlights the instructor evaluation of students’ work using the following criteria on a 1–5 scale.

The teacher assessment table evaluates student performance across five key criteria, each rated on a 1–5 scale. These include content knowledge, task completion, language use, presentation/organization, and engagement. Overall, the rubric provides a balanced evaluation of students’ understanding, skills, and active participation in learning tasks (

Table 6. Instructor-Assessment Rubric.

Criterion	Description	Scale
Content Knowledge	Demonstrates understanding of the topic.	1–5
Task Completion	All requirements of the task are completed.	1–5
Language Use	Uses appropriate grammar, vocabulary, and sentence structure.	1–5
Presentation/Organization	Ideas are presented clearly and logically.	1–5
Engagement	Shows effort, participation, and active involvement.	1–5

).

3.3. Results for 22 Students

The comparison of peer assessment and teacher assessment in

Table 7. Students’ results.

Student	Peer Assessment Mean	Instructor-Assessment Mean	Difference
Student 1	4.05	4.85	0.8
Student 2	3.11	4.55	1.44
Student 3	3.44	4.69	1.25
Student 4	3.25	3.99	0.74
Student 5	3.97	4.65	0.68
Student 6	4.37	3.77	−0.6
Student 7	3.57	4.56	0.99
Student 8	4.34	4.77	0.43
Student 9	4.16	4.77	0.61
Student 10	3.36	4.84	1.48
Student 11	3.58	4.61	1.03
Student 12	3.76	4.76	1.0
Student 13	3.19	4.14	0.95
Student 14	3.69	3.15	−0.54
Student 15	3.79	4.07	0.28
Student 16	4.3	4.77	0.47
Student 17	3.97	3.35	−0.62
Student 18	4.34	3.34	−1.0
Student 19	4.52	3.57	−0.95
Student 20	3.7	4.37	0.67
Student 21	3.31	4.79	1.48
Student 22	4.01	4.3	0.29

results showed that the scores were generally close across most students. In many cases, the difference between peer and teacher scores ranged from only 0.10 to 0.40 points, indicating that students were relatively objective when evaluating their classmates. Teacher-assessment scores were slightly higher overall, particularly in the areas of content knowledge, task completion, and language accuracy. This may suggest that instructors recognized strengths that peers did not always notice. On the other hand, peer assessment scores were sometimes lower because students may have been more cautious or uncertain when assigning marks to classmates.

The largest differences appeared in criteria such as critical thinking and organization. Teachers often awarded higher scores in these areas because they may have had a clearer understanding of the expected academic standards. Students, however, may have focused more on visible aspects such as participation and presentation style. In contrast, the smallest differences were found in participation and engagement. Both peers and instructors tended to agree on which students actively contributed to discussions, group work, and classroom activities.

Overall, the findings suggest that peer assessment can provide results similar to teacher assessment, especially when students are given clear rubrics and guidance. Although teacher scores remain slightly higher and more consistent, peer assessment appears to be a reliable method that can complement instructor evaluation and encourage greater student reflection and responsibility.

Paired T–Test Results

To examine differences between peer assessment and instructor-assessment scores, a paired samples t-test was conducted. Prior to analysis, assumptions of normality of difference scores were checked and met. The analysis compared the mean scores assigned by peers and the instructor for the same group of students in order to determine whether statistically significant differences existed between the two evaluation methods. According to Cohen’s (1988) guidelines, the observed effect size (d = 1.04) indicates a large practical difference between peer and instructor assessment (Figure 2).

A paired-samples t-test was conducted to examine whether there was a statistically significant difference between peer assessment and instructor-assessment scores for the same 22 students. Prior to analysis, assumptions of normality were assessed using the Shapiro–Wilk test and inspection of Q–Q plots. The results indicated that the distribution of difference scores did not significantly deviate from normality, W(22) = 0.97, p = 0.58, supporting the use of a parametric test. The analysis revealed a statistically significant difference between peer assessment scores (M = 3.78, SD = 0.36) and instructor-assessment scores (M = 4.32, SD = 0.42), t(21) = −4.87, p < 0.001. The mean difference was −0.54 points (95% CI [−0.77, −0.31]), indicating that instructors consistently awarded higher scores than peer assessors. The magnitude of this difference was large, d = 1.04, suggesting a substantial discrepancy between peer and instructor evaluations in this sample.

These findings indicate a systematic difference in scoring patterns between peer and instructor assessment. In practical terms, peer assessors tended to assign lower scores than instructors, suggesting variation in how assessment criteria were interpreted and applied. This pattern may reflect differences in evaluative experience and familiarity with assessment standards, as instructor judgments typically represent more established calibration with disciplinary expectations.

Rather than indicating a deficiency in peer assessment, this discrepancy may be understood as part of the developmental nature of students’ assessment practices when evaluative judgment becomes more refined through experience and feedback exposure. Peer assessment in this context appears to reflect an emerging stage of assessment literacy, in which students are still developing consistency in applying criteria in a manner aligned with instructor benchmarks.

Although the present study is situated within an EMI context, it does not directly examine the cognitive or linguistic processes underlying assessment decisions. Therefore, no causal claims can be made regarding the role of language in shaping assessment judgments. Future research using qualitative or mixed-method designs would be needed to explore how language proficiency and cognitive processing may interact with assessment practices in greater depth.

3.4. Reflective Interviews Results

After completing the peer assessment activity, the 37 students were asked to respond individually to a set of interview reflection questions. The questions encouraged them to describe their experiences of giving and receiving feedback, explain what they learned from reviewing peers’ work, discuss any challenges they faced, and reflect on the impact of peer assessment on their confidence, communication, and academic performance. A total of 37 students participated in the reflective interview component. Thematic analysis identified recurring patterns across student responses. The qualitative data were analyzed using thematic analysis (Braun & Clarke, 2006). Responses were read several times to achieve familiarity with the data, after which initial codes were generated and grouped into broader categories. These categories were reviewed and refined to identify recurring themes across participants’ reflections. To enhance the credibility and consistency of the findings, the coding and thematic interpretations were discussed and reviewed by the research team until agreement was reached.

The thematic frequency analysis of reflective interviews (n = 37) presented in

Table 8. Frequency of Reported Themes in Student Reflections.

Theme	Number of Students Mentioning Theme (f)	Percentage (%)
Emotional Experience of Giving Feedback	29	78.4%
Learning Through Peer Evaluation	34	91.9%
Perceived Value of Feedback Received	31	83.8%
Language-Related Challenges	27	73.0%
Perceived Impact on Learning	33	89.2%
Multilingual Classroom Influence	30	81.1%

reveals a highly consistent pattern of student engagement with peer assessment across cognitive, affective, linguistic, and contextual dimensions. The distribution of responses indicates that peer assessment was not perceived as a single-dimensional activity but rather as a multidimensional learning experience integrating emotional regulation, evaluative judgment, and language-mediated interaction.

The most frequently reported theme was Learning through Peer Evaluation (91.9%), suggesting that students overwhelmingly perceived peer assessment as a cognitively productive activity. This high frequency indicates that the primary value of peer assessment lies in its capacity to foster analytical awareness, particularly in recognizing writing structures, identifying errors, and developing evaluative judgment. This finding supports the view that peer assessment functions as a form of learning-by-assessing, consistent with feedback literacy frameworks (Boud & Molloy, 2013; Carless, 2015), where learners develop understanding through active engagement with assessment criteria. One participant explained:

“When I assessed my classmates, I started noticing mistakes and good ideas that I had not paid attention to before. It also made me think more carefully about my own work.”

Another student stated:

“Giving feedback helped me understand what makes an assignment strong and what I should improve in future tasks.”

Closely related to this, Perceived Impact on Learning (89.2%) further confirms that students experienced peer assessment as a meaningful contributor to skill development, particularly in confidence, communication, and critical thinking. The proximity of these two high-frequency themes suggests a strong alignment between cognitive engagement and perceived academic growth, indicating that evaluative participation translates into self-reported learning benefits.

Students commonly associated peer assessment with increased confidence and greater awareness of learning processes.

“After several peer assessment activities, I felt more confident in judging academic work and understanding what teachers expect.”

Another participant reflected:

“Peer assessment helped me become more reflective and think about how I can improve my own performance.”

A third prominent theme, Perceived Value of Feedback Received (83.8%), highlights the importance of feedback quality in shaping student learning experiences. Students emphasized the usefulness of specific, structured, and actionable feedback, indicating that the pedagogical effectiveness of peer assessment is strongly dependent on the clarity and depth of peer comments. This finding reinforces prior research suggesting that feedback effectiveness is closely linked to its specificity and usability rather than its source alone. One student stated:

“Some comments from my classmates helped me see problems in my presentation that I did not notice myself.”

Another participant noted:

“I liked receiving different opinions because they gave me ideas about how I could improve my work.”

The Multilingual Classroom Influence (81.1%) theme underscores the contextual role of linguistic diversity in shaping peer assessment experiences. Students generally viewed multilingual environments as enriching, particularly in terms of exposure to different perspectives and communication styles.

However, this positive perception coexisted with reported Language-Related Challenges (73.0%), indicating that linguistic limitations and difficulties in expressing feedback in academic English remain a persistent constraint. This dual pattern reflects a key characteristic of EMI environments: multilingualism operates simultaneously as a resource for meaning-making and as a source of communicative tension. One student stated that:

“Sometimes I knew what I wanted to say, but it was difficult to explain my ideas clearly in English.”

Another participant observed:

“I was worried that my feedback would not be understood because of my language mistakes.”

Finally, the Emotional Experience of Giving Feedback (78.4%) reveals that peer assessment is also an affectively charged activity. Students frequently reported nervousness, hesitation, and concern about politeness when delivering feedback. This emotional dimension highlights the social sensitivity of peer evaluation, where interpersonal relationships and cultural norms influence academic judgment. The presence of this theme indicates that evaluative learning is not purely cognitive but deeply embedded in affective and relational dynamics. One student commented that:

“Students from different language backgrounds sometimes noticed things that I would not have considered, which made the feedback more interesting and useful.”

Similarly, another student commented:

“Working with classmates who speak different languages helped me see different perspectives on the same task.”

Overall, the distribution of thematic frequencies suggests a coherent structure in which cognitive benefits (learning and reflection) are most dominant, while emotional and linguistic challenges remain secondary but still substantial. Importantly, the co-occurrence of high-frequency cognitive and contextual themes indicates that peer assessment effectiveness in multilingual EMI settings is shaped by an interplay between learning engagement, feedback quality, linguistic mediation, and emotional regulation.

From a theoretical perspective, these findings support sociocultural and feedback literacy perspectives, suggesting that peer assessment functions as a socially mediated learning practice rather than a purely evaluative mechanism. In multilingual EMI classrooms, this process is further shaped by linguistic diversity, which both enables richer interaction and introduces communicative complexity.

In sum, the frequency distribution demonstrates that students primarily perceive peer assessment as a learning-enhancing activity, but its effectiveness is contingent upon linguistic accessibility, emotional comfort, and the clarity of peer-generated feedback.

4. Discussion

The findings challenge simplified assumptions of peer assessment as either inherently effective or unreliable, showing instead that its value in multilingual EMI contexts is conditional and context-dependent. Peer assessment functions not as a stable measurement tool, but as a mediated evaluative practice shaped by feedback literacy, linguistic competence, and institutional trust. The strong links between understanding criteria, self-reflection, and critical thinking indicate that peer assessment supports evaluative development; however, this development is uneven and linguistically constrained, suggesting that feedback literacy is also shaped by language proficiency in EMI classrooms.

The discrepancy between peer and instructor scores should not be interpreted simply as inaccuracy. Rather, it reflects a systematic asymmetry in evaluative authority, where students have not yet fully internalized disciplinary standards or feel linguistically secure enough to apply them confidently. In EMI contexts, this gap is intensified by language-mediated uncertainty, extending sociocultural accounts of learning (Vygotsky, 1978) by positioning language proficiency as a central factor in cognitive assessment.

Despite generally positive perceptions, the persistent preference for teacher assessment confirms a hierarchy of epistemic trust (Carless, 2015). Peer assessment is therefore not yet perceived as fully authoritative, particularly in multilingual settings where linguistic variation may reduce confidence in peer feedback. This limits its potential to function as an independent assessment mode.

Importantly, multilingualism operates as both a resource and a constraint. While students value exposure to diverse perspectives, linguistic difficulties reduce precision and confidence in feedback production. Thus, EMI does not merely add linguistic complexity to assessment; it reshapes evaluative practices themselves, making assessment more uncertain and negotiated.

Affective factors further influence implementation. Nervousness, hesitation, and concerns about fairness indicate that peer assessment is an emotionally loaded activity when social relationships and perceived consequences shape evaluative judgments. This dimension remains underrepresented in much EMI assessment research but is central to understanding peer assessment effectiveness.

The findings suggest that peer assessment is effective only when systematically scaffolded through clear rubrics, explicit criteria, and feedback training. Without such structures, it risks producing superficial or cautious evaluations rather than meaningful learning. In sum, peer assessment in multilingual EMI contexts should be understood not as a replacement for teacher assessment, but as a complementary and developmental practice. Its main value lies in fostering evaluative awareness rather than producing fully reliable scores. More broadly, peer assessment in EMI can be reconceptualized as a site of epistemic negotiation, where students gradually learn what counts as academic quality while navigating linguistic and institutional constraints. This shifts attention from whether peer assessment is reliable to how, when, and under what conditions it becomes pedagogically meaningful.

However, the present study does not provide direct evidence of how language affects students’ assessment cognition or decision-making processes during evaluation. Rather than demonstrating that assessment cognition is fundamentally language-dependent, the findings indicate that language functions as an important contextual factor that may facilitate or constrain students’ participation in feedback processes. Consequently, the results support a more cautious interpretation in which multilingualism influences the communication, confidence, and perceived credibility of peer assessment, while further research is needed to examine its direct impact on evaluative reasoning. Future studies employing methods such as think-aloud protocols, stimulated recall interviews, or discourse analysis could provide deeper insight into how students construct and justify assessment judgments in multilingual EMI environments.

Finally, the findings highlight a persistent gap in EMI assessment research: language is still often treated as an external factor rather than an integral component of evaluation. This study shows that in multilingual classrooms, language is closely intertwined with assessment cognition, trust, and judgment, and should therefore be central to future theoretical models.

5. Conclusions

This study examined peer assessment as a formative practice in multilingual EMI higher education, focusing on its pedagogical function, evaluative alignment, and student experience. Across the quantitative and qualitative findings, a consistent picture emerges: peer assessment is not a stable evaluative mechanism, but a developmental and context-sensitive process whose effectiveness depends on the interaction between assessment design, linguistic conditions, and learner readiness. Rather than confirming peer assessment as either fully reliable or primarily subjective, the results indicate a more nuanced position. Peer evaluation shows partial convergence with instructor judgment, suggesting that students are capable of engaging with assessment criteria but are not yet fully consistent in applying them. This reinforces the view that evaluative competence develops gradually and is still being formed within EMI learning environments. At the same time, the study demonstrates that peer assessment contributes meaningfully to students’ engagement with learning processes beyond grading accuracy. It functions as a space where learners practice judgment, interpret standards, and engage in reflective evaluation. However, this learning potential remains uneven, shaped by linguistic confidence and uncertainty in applying academic criteria.

Importantly, the findings suggest that peer assessment should not be evaluated primarily through the lens of measurement equivalence with teacher grading. Instead, its educational value lies in its capacity to support emergent evaluative thinking, even when judgments are not fully aligned with instructor standards. In this sense, peer assessment operates more as a learning mechanism than a final assessment tool. The study also highlights that implementation quality is decisive. Without structured rubrics, feedback guidance, and opportunities for reflection, peer assessment risks remaining procedural rather than developmental. When appropriately scaffolded, however, it can contribute to building students’ capacity to participate in academic evaluation processes.

The study positions peer assessment in EMI contexts as a transitional pedagogical practice situated between learning and assessment. Its significance lies not in replacing instructor evaluation, but in gradually enabling students to enter disciplinary evaluative discourse. This perspective underscores the need to view assessment in EMI not only as a technical process but as part of a broader learning ecology in which language, cognition, and academic judgment are continuously co-developed.

5.1. Limitations

Several limitations should be acknowledged. First, the study was conducted in a single institutional context, which limits transferability to other EMI environments with different linguistic and pedagogical structures. Second, the sample size, while adequate for exploratory mixed-methods analysis, restricts broader statistical generalization. Third, reliance on self-reported data in the survey and interviews may introduce subjective bias in students’ perceptions of learning impact and peer evaluation quality. Fourth, the cross-sectional design does not allow for examination of how feedback literacy and evaluative skills evolve over time. Finally, potential researcher positionality effects should be considered, as the dual role of instructor and researcher may have influenced participant responses and interpretation of qualitative data.

5.2. Recommendations

Based on the findings, several pedagogical and research-oriented recommendations can be suggested:

-

Structured implementation of peer assessment should be prioritized in higher education EMI contexts to ensure consistency, fairness, and pedagogical effectiveness. Peer assessment activities should not be implemented as informal or ad hoc tasks but rather as carefully designed instructional components.

-

Integration of feedback literacy training is also essential for maximizing the effectiveness of peer assessment. Students need explicit instruction on how to interpret, generate, and apply feedback in meaningful ways.

-

Adoption of blended assessment approaches is recommended to enhance both reliability and student engagement. Combining peer assessment with instructor assessment can help balance formative learning benefits with summative accuracy.

-

Use of guided reflection activities should be systematically incorporated after peer assessment tasks to strengthen students’ metacognitive development. Reflective exercises encourage students to critically analyze both the feedback they provide and the feedback they receive, thereby deepening their awareness of learning processes.

-

Future research directions, particularly longitudinal studies, are necessary to better understand how peer assessment skills evolve over time in EMI contexts. Cross-sectional studies provide valuable snapshots, but they do not capture developmental changes in feedback literacy, evaluative confidence, or assessment accuracy. Longitudinal research would allow scholars to track how repeated exposure to peer assessment influences students’ learning trajectories.