Young learners’ synchronous online peer interaction: teachers’ beliefs of its benefits and implementation

Abstract

This study explored teachers’ perceptions of synchronous online peer interaction (SOPI) among young learners (YLs), aged 7 to 15. Thirty-eight teachers of diverse L1 backgrounds and teaching experience completed a survey and attended an in-depth interview that explored multiple aspects of SOPI (e.g. benefits, issues, frequency of use, strategies to promote SOPI’s effectiveness, and mediating factors). Survey data were analysed using exploratory factor analyses to identify themes/constructs related to areas of SOPI under investigation, and inferential statistics were reported to examine teachers’ self-ratings of different aspects of SOPI. Interviews were analysed following a content-based approach to supplement the quantitative data. The results show that teachers believed SOPI promoted YLs’ attention to language aspects and created a peer-support learning environment. SOPI’s effectiveness was affected by multiple factors (e.g. learners’ individual differences, teachers’ guidance, grouping, and learners’ financial and family backgrounds). Teachers expressed the importance of using SOPI for YLs in online classes and suggested instruction/guidance to maximize its’ benefits. However, the frequency of use of SOPI was reported to be dependent on schools’ regulations and resources, learners’ affective aspects, and parental support. The results also indicate the impact of teachers’ online teaching experience on their differential perceptions of online SOPI.

ABSTRACT IN ANOTHER LANGUAGE (i.e., VIETNAMESE)

Nghiên cứu này tìm hiểu quan điểm của giáo viên về hoạt động tương tác trực tuyến giữa các học sinh ở độ tuổi từ 7 đến 15 tuổi trong các lớp học tiếng Anh trực tuyến. Đối tượng nghiên cứu bao gồm 38 giáo viên có kinh nghiệm giảng dạy khác nhau và nói nhiều tiếng mẹ đẻ khác nhau. Các giáo viên này tham gia khảo sát và phỏng vấn chuyên sâu về nhiều khía cạnh của hoạt động tương tác trực tuyến với bạn đồng trang lứa. Các khía cạnh này bao gồm lợi ích, vấn đề, tần suất sử dụng, chiến lược để thúc đẩy hiệu quả của hoạt động tương tác trực tuyến với bạn đồng trang lứa, cũng như những yếu tố ảnh hưởng đến tính hiệu quả của nó. Dữ liệu khảo sát được phân tích bằng phương pháp phân tích nhân tố khám phá để xác định các nhóm/yếu tố liên quan đến các khía cạnh của hoạt động tương tác trực tuyến với bạn đồng lứa, và sử dụng thống kê suy luận để xem xét quan điểm của giáo viên về các mặt của hoạt động tương tác trực tuyến với bạn đồng lứa. Các cuộc phỏng vấn được phân tích theo cách tiếp cận dựa trên nội dung để bổ sung dữ liệu định lượng. Kết quả cho thấy các giáo viên tin rằng hoạt động tương tác trực tuyến với bạn đồng lứa đã thúc đẩy sự chú ý của học viên nhỏ tuổi đối với các khía cạnh ngôn ngữ, và tạo ra một môi trường học tập tích cực. Tuy nhiên, tính hiệu quả của hoạt động tương tác trực tuyến với bạn đồng lứa bị ảnh hưởng bởi nhiều yếu tố, ví dụ, sự khác biệt cá nhân của người học, hướng dẫn của giáo viên, người làm việc chung, và hoàn cảnh gia đình. Các giáo viên tham gia nghiên cứu nhấn mạnh tầm quan trọng của việc sử dụng hoạt động tương tác trực tuyến với bạn đồng lứa khi dạy các lớp học trực tuyến cho học sinh nhỏ tuổi, và đề xuất nhiều phương pháp để tối đa hóa lợi ích của nó. Về tần suất sử dụng hoạt động tương tác trực tuyến với bạn đồng lứa, giáo viên cho rằng các quy định và nguồn tài nguyên của trường, các khía cạnh cảm xúc của người học, và sự hỗ trợ của phụ huynh ảnh hưởng đến tần suất sử dụng. Kết quả nghiên cứu này cũng chỉ ra rằng kinh nghiệm giảng dạy trực tuyến của giáo viên có ảnh hưởng đến cách họ nhìn nhận các vấn đề liên quan đến hoạt động tương tác trực tuyến với bạn đồng lứa.

PLAIN LANGUAGE SUMMARY

To obtain a comprehensive understanding of language teachers’ practices in the classroom, it is necessary to know how teachers think about these practices. In this research, we attempted to explore how language teachers perceived young learners (YLs)’ synchronous online interaction. We interviewed and surveyed language teachers from different language and teaching backgrounds about their perceptions of the benefits, issues, and frequency of using learner-learner (peer) interaction in online English classes for YLs. We also asked the teachers about factors affecting the effectiveness of online peer interaction, and what strategies they used to promote learner engagement in online language classes. The results from the survey and interviews revealed that teachers believed synchronous online peer interaction promoted young learners’ attention to language issues as well as created a sense of peer-support in the online learning environment. However, the teachers cited many factors affecting the effectiveness of the use of online peer interaction. They included, for example, learners’ individual differences, teachers’ guidance, grouping, and learners’ financial and family backgrounds. The teachers also highlighted the importance of using peer interaction for YLs in online classes and suggested various strategies to maximize its benefits. Additionally, the teachers stated that the schools’ regulations and resources, learners’ affective aspects, and parental support affected and determined the extent to which they would be using peer interaction in their online classes.

Keywords:

• Synchronous online PI
• task-based interaction
• teacher belief
• young learners

Từ khóa:

• Tương tác với bạn đồng lứa trong lớp học trực tuyến
• tương tác dựa trên hoạt động học
• quan điểm của giáo viên
• học sinh nhỏ tuổi

Introduction

Following the development of diverse second language (L2) pedagogical approaches such as communicative language teaching (CLT) and task-based language teaching (TBLT) that place learners at the center of all teaching and learning activities, learner-learner interaction, henceforth peer interaction (PI), has become a prevalent type of face-to-face and/or online interaction in L2 classrooms. The current study investigated this type of interaction in technology-supported modality because of the following reasons. First, despite an increasing body of studies using interactional data to document benefits and issues of face-to-face PI (Philp et al., 2014; Sato & Ballinger, 2016), little research to date has investigated teachers’ perceptions of synchronous online peer interaction (SOPI) regarding its benefits, issues, and implementation. Second, research has largely investigated face-to-face PI among adult learners, with few studies exploring synchronous online PI among young learners (YLs), especially in terms of teachers’ perceptions (Carless, 2002; Oliver & Azkarai, 2019; Pinter, 2007; Shintani, 2014). This points to the gap where there is little understanding of how teachers perceive and implement this kind of SOPI for YLs. Third, since synchronous online language classes for YLs are currently being delivered in some EFL contexts, it is important to examine teachers’ perceptions of SOPI so that appropriate pedagogical implications could be provided. Finally, previous research suggests that teachers’ beliefs are likely to affect or determine their classroom practices (Barcelos & Kalaja, 2013; Li & Walsh, 2011). It is therefore essential to understand their beliefs about SOPI among YLs so that relevant attempts could be conducted to raise teachers’ awareness of the benefits and issues of this type of interaction. Against this background, the current study explores teachers’ perceptions of SOPI among young EFL learners (aged 7 to 15), specifically focusing on its benefits, implementation, and mediating factors.

Literature review

Theoretical underpinnings for the role of PI in L2 learning

Before discussing its theoretical underpinnings, it is necessary to describe what PI refers to. PI is a type of interaction where learners interact with each other to complete a communicative task. Learners in PI often include those who assumingly share learning goals and equivalent skills, are at a similar language proficiency level and probably at a similar age, and function as classmates in a language classroom (Philp et al., 2014). PI can comprise multiple parties (two or more learners) perceived as symmetrical in various aspects of abilities, and at times PI may involve teacher’s (minimal) participation.

The facilitative role of PI in L2 learning could be explained from multiple perspectives: cognitive-interactionist, skill-acquisition, and sociocultural perspectives. From the cognitive-interactionist perspective, interaction is perceived as a context for negotiation of meaning and conversational modifications through which learners attend to language features, connect input and output, and benefit from interactional feedback (Long, 1996). Interaction also provides learners with opportunities for correcting, experimenting, and polishing language (Philp et al., 2014), all considered central to L2 development. The benefits of PI could be seen in the process of producing language in which learners are ‘pushed’ to think about language to express meaning (Philp & Iwashita, 2013), to raise their awareness of the relationship between form, function and meaning (Kowal & Swain, 1994), and to modify their output to make it become more precise, coherent, and appropriate (Swain, 1985). The role of PI in L2 development could also be explained through skill-acquisition theory, emphasizing the contextualized practice which drives learners to transform their declarative knowledge (i.e. mental representation of skills) to procedural knowledge (i.e. actual execution of knowledge to perform cognitive operations with automatization) (Anderson, 2005). PI represents a contextualized practice which could contribute to improved, faster, and more accurate processing of L2 in both comprehension and production (Sato & McDonough, 2019).

From the holistic view of sociocultural theory in which L2 learning concerns all individual and environmental elements taken into the process of knowledge building (van Compernolle, 2015), PI is perceived as context where learners reflect on language production, and co-construct language knowledge through collaborating and mutually assisting each other to resolve language- and task-related issues. The collaborative dialogue in PI is likely to enable learners to move from being unable to perform a task individually to completing it independently thanks to peer assistance and scaffolding (Lantolf et al., 2015). Thus, PI engages learners in the process of collaborative learning, which is considered crucial to knowledge building.

PI among young learners

To date, L2 research has largely investigated PI among adult learners, documenting its benefits (see Philp et al., 2014; Sato & Ballinger, 2016 for reviews), mediating factors (e.g. task-related factors, proficiency, perceptions and context) (see Dao & McDonough, 2017, 2018; Erlam & Pimentel-Hellier, 2017; Philp et al., 2010, Sato & Viveros, 2016; Sato & Lyster, 2007; Yoshida, 2008, 2010), and pedagogical interventions to address PI’s issues (see Dao, 2020; Kim & McDonough, 2011; Sato, 2013; Sato & Ballinger, 2012; Sato & Lyster, 2012). Extending this body of research, some research has explored PI among YLs to examine the extent to which findings about adult learners apply to YLs. Expectedly, young learners possess different capabilities, needs, and perspectives, which set them apart from adult learners (Ellis, 2014; Mackey & Gass, 2005). Thus, it is essential to obtain information about YLs in PI (Oliver & Azkarai, 2017).

Among the first studies investigating PI among YLs (see also Chaudron, 1988; Ellis, 1984) is Swain and Lapkin’s (1998) study that analysed grade 8 French immersion learners’ PI when performing a text construction task. They found that during PI young learners were engaged in resolving language problems and helping each other complete the task. Their engagement in resolving language issues, so-called language-related episode (LREs), is argued to enable them to activate their mental processes and thus serves as learning opportunities. Additionally, when analysing grade 6 learners’ PI, Gagné and Parks (2013) found that YLs provided diverse scaffolding to peers. Similarly, Pinter (2007) explored PI between 10-year-old L2 learners with a very low proficiency level and found that YLs were able to carry out PI and benefited from it through assisting each other, attending to and using language for communication, and completing the tasks. These YLs also expressed positive perceptions about PI and recognized positive changes in their performances.

Regarding pair/group dynamics in PI (cf. Storch, 2002), recent studies have also documented some patterns unique to YLs. For example, Butler and Zeng (2014, 2015) found that during PI 4^th grade Chinese learners of English experienced more difficulties in providing information to each other and in completing tasks as compared to 6^th grade learners. Comparing PI between two groups of Spanish YLs of English (aged 8—9 vs. 9—10) in two task repetition conditions (i.e. procedural and exact task repetition), García Mayo and Imaz Agirre (2016) found that the younger learner group demonstrated more collaborative patterns than the older learner group in both conditions (see also Sample & Michel, 2015; Shintani, 2012, 2014 for the benefits of task repetition for L2 interaction and learning). Additionally, Ahmadian and Tajabadi (2017) explored pair dynamics among Iranian YLs of English and found that YLs mostly showed a collaborative pattern over the six learning sessions, despite some pairs revealing variations from being less collaborative to more collaborative in PI. Their results also provide some evidence for the link between collaborative interaction patterns and vocabulary learning. These results are supported by Lázaro Ibarrola and Hidalgo’s (2017) study that documented positive impacts of PI on YLs’ language development, especially when implemented via task repetition (see also Mackey & Oliver, 2002; Mackey & Philp, 1998; Mackey & Silver, 2005; Tedick & Young, 2016, for positive evidence of young learners’ PI on language development).

In relation to the proficiency variable, Oliver (1995, 1998) examined PI between two groups of YLs (aged 8 to 13): native versus non-native speakers of English. She found that young native-speaker learners provided feedback to their non-native peers who then used this feedback in their interaction. They also negotiated for meaning using a wide range of strategies. In another study investigating YLs (aged 8 to 13), Oliver (2002) found that learners were engaged in negotiation, and this negotiation increased when learners were of lower proficiency level as compared to peers of higher proficiency level.

When compared to adult learners, research has shown that YLs (aged 8 to 13) tended to use mostly self- and other-repetition for negotiating for meaning as opposed to comprehension checks used by adult learners (Oliver, 1998, 2002). They were more likely to modify their output as opposed to adults (Mackey et al., 2003). Although Mackey et al. (2003) and Pinter (2007) found differences between young and adult learners in strategies used for completing tasks, they both suggested that like adults, interactions among YLs facilitate L2 learning due to the opportunities created within the interaction for comprehending input, producing and modifying output following feedback. Young learners’ PI was also found to be affected by different variables like adult learners. For instance, YLs tended to be more willing to negotiate for meaning, modifying output, and taking more risks when performing familiar tasks than unfamiliar ones (Mackey et al., 2007).

With regard to PI among YLs in synchronous online platforms or SOPI, research on this area is relatively scant, although there have been an increasing number of studies on adults’ SOPI (see Lin, 2015). Studies on adult learners’ SOPI have largely focused on learners’ interactional behaviours such as learners’ discussion of language features, negotiation of meaning, patterns of interaction, turn-taking (Yanguas, 2010, 2012), repair moves (Jepson, 2005), speech act (i.e. refusals) (Sykes, 2013), language-related episodes (LREs) (Bueno-Alastuey, 2013; Yanguas & Bergin, 2018), noticing in communication breakdowns (Bueno-Alastuey, 2010), amount of language production (Bueno-Alastuey, 2010, 2011), and identity-based interactional behaviours (Helm & Dooly, 2017). Similar to these studies with adult learners behaviours in SOPI is also in its early stage. For instance, Kopf (2012) is one among very few studies that investigated YLs’ interactional strategies in SOPI. The results show that YLs used a wide range of interactional strategies which enhanced their comprehension of input and production of output, and that task types mediated their negotiation routines in SOPI. These results indicate that YLs were able to interact online, suggesting potentials of SOPI for L2 learning.

Overall, L2 research that has primarily focused on adults and YLs’ interactional behaviours in face-to-face and online PI has provided important information about the benefits and distinctive features of young learners’ PI (as opposed to adult learners’ PI) and its link to L2 learning. However, less is known as to how teachers perceive young learners’ SOPI, and it appears that no studies, to the authors’ knowledge, have investigated how teachers perceive and use SOPI for YLs (e.g. aged 7 to 15) in online L2 classes. Understanding teachers’ perceptions of young learners’ SOPI is pedagogically important since it would help understand the extent to which SOPI is used, and how teachers implement it in synchronous online classes.

Teachers’ beliefs (perceptions) about PI

In the field of L2 teaching, teachers’ beliefs or perceptions are defined as evaluative propositions that individuals regard as true and that have a strong affective component (Borg, 2011). Teachers’ stated beliefs can also be perceived as ‘statements teachers made about their ideas, thoughts, and knowledge that are expressed as evaluations of what should be done, should be the case, and/or is preferable’ (Basturkmen et al., 2004, p. 244). Notably, teachers’ beliefs are often linked to their interpretive ability to make sense of the social world around them and to developing a sense of self as teachers.

Understanding teachers’ beliefs is essential because it could shed light on many pedagogical questions that are fundamental to L2 teaching practices (Barcelos & Kalaja, 2011; Borg, 2003, 2006, 2015). These questions may concern whether teacher beliefs about teaching (e.g. understanding of SOPI) match with their practices in the classroom (i.e. implementing SOPI), and whether there are mismatches (or matches) between teachers’ beliefs and their practice. Additionally, studying teachers’ beliefs is important because beliefs held by teachers are related to their reflections about their own practices (i.e. using SOPI), so they are likely to affect their practices, actions, and decisions in their everyday practice (i.e. implementing SOPI), their responses to changes or innovations, and their process of learning to teach. Once there is a comprehensive understanding of the relationship between teachers’ beliefs and specific pedagogical issues, (e.g. SOPI among YLs), appropriate pedagogical suggestions could be generated.

Given the important role of teachers’ beliefs in influencing their practices, existing research has explored teachers’ beliefs about different classroom practices such as teaching methods, including CLT and TBLT (Liu & Ren, 2021; Wesely et al., 2021), corrective feedback (Kamiya, 2016; Kartchava et al., 2020), teaching pronunciation (Baker, 2014; Couper, 2017), incidental focus on form (Basturkmen et al., 2004), grammar instruction (Graus & Coppen, 2016), pedagogical use of online interaction tools (Taghizadeh & Ejtehadi, 2021) and so on. However, little research has directly looked at teachers’ beliefs of SOPI regarding its frequency of use, benefits, and issues.

Despite not focusing directly on PI, some previous studies indicated that teachers hold positive beliefs about the facilitative role of PI. For instance, Liu and Ren (2021) found that most teachers stated that they supported the use of PI as reflected in their endorsement of Task-Based Language Teaching (TBLT) approach for which PI is a central element. These teachers believed that their role in classroom interaction, especially in PI, has shifted from the authoritative and knowledge transmitter to the role of a facilitator, a guide, a task organizer or even the audience of PI. Teachers’ beliefs on the role of oral communication and PI are also demonstrated in Li and Walsh (2011) study, in which an experienced teacher stated that ‘communication [including PI] goes first, and language is not only a subject but a medium they use when going into another community’ (p. 49). The comment shows teachers’ positive beliefs of the facilitative role of communication and PI in L2 learning.

Despite these positive beliefs it is not always the case that teachers implement classroom interaction or use PI (Li & Walsh, 2011; also see Borg, 2003, 2011; Li, 2013). For example, primary Pakistani teachers in Arshad’s (2017) study believed that although PI has potential for L2 learning, it poses many issues. These teachers also indicated that they were aware of the benefits of PI, but the contextual and individual-related factors prevented them from using it (see also Carless, 2002, 2003, 2004, 2007 for issues related to the implementation of TBLT and PI for YLs). These contextual factors included learners’ lack of autonomy in learning, teachers’ lack of support, school facilities, and teacher training on how to implement PI effectively, and their burden of required teaching-unrelated tasks. Overall, the existing research suggests that L2 teachers generally hold positive beliefs about the benefits of PI, but multiple factors were cited preventing them from using it. Despite being informative, this existing research has largely explored indirectly teachers’ beliefs of face-to-face PI (mostly among adult learners) in relation to the implementation of TBLT. To the authors’ knowledge, no studies so far have looked at teachers’ beliefs of SOPI among YLs in online classes. To address these gaps, the current study explores teachers’ perceptions of SOPI among YLs, focusing on various aspects: benefits, frequency of use, implementation, strategies to promote SOPI’s effectiveness, and mediating factors.

Research questions

RQ1. What are teachers’ perceptions of synchronous online PI among YLs regarding its benefits, mediating factors, and strategies promoting its effectiveness?

RQ2. What are teachers’ reported degrees of using PI in their synchronous online class of YLs and factors affecting its frequency?

RQ3. To what extent do teachers’ perceptions of synchronous online PI differ according to their teaching experience and previous training in synchronous online teaching?

Method

Participants and instructional context

Participants were 38 teachers of English (female = 24; male = 14) recruited from various private language centres in different regions of Vietnam, that provided both synchronous online and face-to-face English classes to Vietnamese EFL learners. Their age mean was 30.39 (SD = 7.64) and they had diverse L1 backgrounds (22 Vietnamese, 4 Americans, 4 Filipinos, 3 British, 1 Australian, 1 Malaysian, 1 Serbian, 1 Russian, and 1 Romanian). Regarding their English proficiency level based on the Common European Framework of References for Languages (CEFR), 17 teachers reported to be at C2 or native-speaker level, with 15 teachers stating to achieve C1 level as opposed to six teachers at B2 level. All teachers reported to hold a BA in language teaching or a certificate in TESOL.

They had an average of 6.64 years (SD = 6.02) of teaching experience, and their average year of synchronous online teaching experience was 1.46 (SD = 1.17). Twenty-four teachers reported to attend at least one professional development (PD) training on synchronous online teaching from their institutions and/or free-online workshops, whereas the rest (n = 14) had not attended any training. The platforms they used to teach synchronous online English classes included Zoom, Google Meet, Facebook Messenger, Zalo, Skype and Microsoft Teams, among which Zoom and Google Meet were reported to be used by a majority (n = 34). The participants reported to teach Vietnamese EFL learners of various age groups, ranging from 7 years old to 15 years old. Their class size was often small, ranging from a group of five to 15 learners.

Study’s design and materials

This study explored teachers’ perceptions of SOPI among YLs regarding its benefits, factors affecting its effectiveness, strategies to enhance its effectiveness, teachers’ frequency of use and factors affecting their use, and possible variations in teachers’ perceptions and use of SOPI. To achieve this goal, the study adopted a mixed-methods design using both a survey (quantitative) and an individual in-depth interviews (qualitative) to investigate various aspects of SOPI as mentioned above. The survey comprised two sections, with the first section using open-ended questions to obtain teachers’ background information and teaching contexts, and the second section using five-point Likert scale items to gain insights into teachers’ perceptions of SOPI among YLs (Appendix A). The Likert-scale items were organised around distinct parts in line with areas of focus of the study: 1) general perceptions of SOPI, 2) perceived benefits of SOPI, 3) factors affecting SOPI, 4) strategies to promote effectiveness of SOPI, and 5) factors affecting the frequency of use of SOPI. These Likert-scale items were developed based on two sources: the preliminary data from interviews with five teachers and findings from previous research that investigated PI (see Adams & Oliver, 2019; Philp et al., 2014; Sato & Ballinger, 2016). The in-depth interviews also focused on teachers’ perceptions of the five aforementioned themes (Appendix B).

Procedure

The data collection lasted for two months, scheduled according to participants’ availability. Prior to the data collection, the second author conducted five pilot interviews with five synchronous online English teachers using the preliminary interview questions. The pilot data were used in combination with findings from previous research about PI to write up the Likert-scale items and to revise the interviews questions further. Once the Likert-scale survey was created and further piloted with two synchronous online English teachers for revising the wording, it was delivered to all participants. It should be noted that the pilot participants were not included in the final pool of participants, and their pilot data were not analysed for this study. The in-depth interview with 11 synchronous online English teacher-volunteers, who indicated in the survey their willingness to participate in the subsequent interview, were conducted online after they completed the survey. All the interviews (approximately 40 minutes) were audio-recorded and conducted in English since all participants preferred it.

Analysis

The survey responses were entered into the SPSS software (version 27) and checked for outliers and missing data. A series of (four) exploratory factor analyses (EFA) (i.e. principal component analyses) were performed on four separate sections of the Likert-scale items to identify emergent latent variables according to the four target areas of investigation: 1) perceived benefits, 2) factors affecting the benefits, 3) strategies to promote SOPI, and 4) factors affecting frequency of using SOPI. Since the four sections of the survey conceptually targeted different areas of investigation, we performed four separate CPAs. Descriptive and inferential data were then reported to answer the first two research questions which asked about teachers’ perceptions of benefits, factors mediating the effectiveness, frequency of use, factors affecting the frequent of use, and teachers’ strategies to promote SOPI’s effectiveness. To answer the third research question which explored possible variations in teachers’ perceptions of SOPI according to teaching experience (face-to-face and online), non-parametric Mann-Whitney U tests were performed. To create groups of comparison according to teaching experience, we followed Huberman’s (1989, 2000) and Araújo et al. (2015) framework of teacher’s life cycle phases in which teachers with fewer than three years of teaching experience were categorised into the ‘entry’ group and teachers with three years or more formed the ‘stabilization’ group. For online teaching experience, since all teachers in this study had fewer than three years of online teaching experience and some had just a few months of online teaching experience, we, for the practical purpose, then categorised them into two groups: a group of teachers with less than 1 year of online teaching experience versus a group of just more than one year but fewer than two years of experience.

For the qualitative data, the interviews were transcribed verbatim and then a content-based analysis was employed to identify common themes. The first and third authors first read through all the transcripts independently to get the general understanding and then code separately to identify emergent topics related to the five aforementioned areas of focus in this study. Then, the two coders discussed and grouped emergent topics/ideas into common themes to supplement the quantitative results.

Results

Perceived benefits of synchronous online PI among young learners

An EFA was performed on eight questionnaire items (Part B of the survey) to identify teachers’ perceived benefits of SOPI for YLs’ L2 learning. Data screening showed no multicollinearity issues and no correlations coefficients among items being higher than .90. The Kaiser-Meyer-Olkin (KMO) measure verified the sampling adequacy for the analysis (KMO = 5.88), suggesting relatively compact patterns of correlations and the identified factors being reliable and distinct. The results of Bartlett’s test of sphericity showed significance (p < .01). Following the Kaiser criterion for communalities after extractions with eigenvalues being greater than 1, two factors were extracted, which in combination explained 63.08% of the variance. These two factors concerned learners’ focus on language aspects, and their peer-related and interaction environment (Table 1).

Table 1. Teachers’ perceived benefits of SOPI among young learners.

Items	Factor loadings
Focus on language aspects (Factor 1)
•SOPI promotes YLs’ focus on grammatical aspect of language.	.88
•SOPI promotes YLs’ focus on the spellings of words.	.76
•SOPI promotes YLs’ focus on pronunciation aspect of language.	.46
Peer-related and interaction environment (Factor 2)
•SOPI promotes peer supports among YLs in task completion.	.83
•SOPI promotes YLs’ willingness to provide feedback to peers.	.72
•SOPI promotes YLs’ positive feelings when receiving peers’ feedback.	.63
•SOPI promotes a stress-free learning environment for YLs to learners with peers.	.44

Note: SOPI = Synchronous online peer interaction; Factor loadings <.4 were suppressed.

In Table 1, three items loaded highly on Factor 1, which was focused on language aspects, including grammar, pronunciation, and spellings. The mean score of Factor 1 (an average score of three items) based on the 5-point Likert scale was 3.01 (SD = 1.06), indicating the surveyed teachers agreed that online PI promoted YLs’ attention to different aspects of language. For Factor 2, four items loaded highly on this factor. These items concerned peers’ support, learners’ willingness to provide feedback to peers, positive feelings of receiving peer feedback, and stress-free attitude when interacting with peers. The mean score of 3.16 (SD = 1.11) for Factor 2 indicated that the teachers perceived online PI as a facilitative condition for learning an L2 with peers, where YLs were more likely to support each other for task completion, provide feedback, and feel positive during a stress-free interaction.

The in-depth interviews with the teachers (n = 11) also confirmed the quantitative results, with all teachers (100%) stating that SOPI enabled YLs to focus on language aspects. Teacher P11 commented on this aspect as below.

When kids are drawn to different aspects of language during the interaction, for example, pronunciation, language use or grammar, they would immediately start yelling and correcting each other. They would get really excited about this and not afraid of insulting each other. I think the vast majority of kids focused on pronunciation. Like…kids said ‘that is what it is’ or ‘this mean this or that or something like that’. They also write down the spellings for words as well. Kids often ‘criticized’ each other but tried to support each other like ‘oh you said that. That’s not how you say it. You should say it like this’. Kids are often excited about correcting pronunciation. It seems they are very proud if they could correct their friends. That is what kids love to do. They wanted to feel helpful. [Teacher P11]

The teachers (54.45%) also further elaborated that as YLs felt stress-free and more socially connected with peers, they would be able to focus on other different aspects of language (Teacher P1), build more confidence in speaking and communicative skills (Teacher P9) and thus enhance their memory of what [aspects of language] has been learned and produce more language (Teacher P10).

Also corroborating the quantitative results, the teachers (63.63%) agreed that SOPI among YLs created a facilitative and comfortable learning condition in which they could build collaboration and help and teach each other because all kids have their own strengths and weakness (Teacher P4). Because of these perceived benefits, all interviewed teachers (100%) and surveyed teachers confirmed in their questionnaire responses (Part A of the survey) that PI among Vietnamese young learners should be promoted in [synchronous] online classes (M = 4.47, SD = .82) and that online classrooms are suitable environment for using PI (M = 3.86, SD = .09).

Factors affecting synchronous online PI’s effectiveness

To explore factors that affect the benefits of SOPI among YLs, an EFA was conducted on 13 questionnaire items (Part C of the survey) and, the results revealed a KMO value of .66 confirming the sampling adequacy, and Bartlett’s test of sphericity being significant (p < .01). Four factors that had eigenvalues over Kaiser’s criterion of 1 were extracted and they altogether explained 75.21% of the total variance. The four factors concerned learner individual differences, teachers’ provision of instruction, grouping of learners, and family backgrounds (Table 2).

Table 2. Factors affecting SOPI.

Items	Factor loadings
Learner individual differences (Factor 1)
•YLs’ English proficiency affects the effectiveness of their online PI.	.69
•YLs’ interest in interacting with peers affects their online PI.	.85
•YLs’ interest in online learning affects their SOPI.	.73
Teachers’ provision of instruction (Factor 2)
•My instruction on how to interact with peers affects young learners’ SOPI.	.68
•My instruction on how to do the tasks affects young learners’ SOPI.	.86
•My instruction on how to use online interaction and learning tools affects young learners’ SOPI.	.87
Grouping of learners in PI (Factor 3)
•Grouping YLs of different levels negatively affects SOPI.	.86
•Pair work positively affects young learners’ SOPI.	.85
•Small group work positively affects young learners’ SOPI.	.69
Family backgrounds (Factor 4)
•YLs’ financial background (i.e. rich or poor families) affects their SOPI.	.81
•The assistance of parents in the online class positively affects young learners’ SOPI.	.83

In Table 2, three items loaded highly on Factor 1 that was related to learner individual differences. These items concerned learners’ level of proficiency, and their interests in interacting with peers and online learning modality. The mean score for Factor 1 was 3.36 (SD = 1.19), indicating that young learner individual differences (i.e. proficiency, interest in talking to peers and online learning) were considered to affect the benefits of SOPI.

The in-depth interviews revealed that all teachers (100%) were concerned with individual learner differences when conducting SOPI. Specifically, for YLs’ proficiency level, the teachers stated that SOPI will not work if the kids’ proficiency was low (Teacher P11). They (81.81%) also expressed that when kids did not like their peers, they would not talk and thus PI was just ineffective. Some kids just cried when paired with someone they don’t want (Teacher P8). Additionally, some teachers (54.55%) shared that SOPI was a bit difficult to run effectively when kids were not interested in learning online and thus they just kept silent and there was no way to open their mouth (Teacher P4).

Regarding Factor 2 that concerned teachers’ instruction or guide, there were three items loading highly on it. These items concerned teachers’ instruction or guide on how to interact with each other online, how to carry out the task, and how to use online interaction/learning tools. An average score of these items (M = 4.04, SD = 1.01) indicated a high agreement among teachers on the impact of their instruction on the effectiveness of PI. The interview results also revealed teachers’ (72.72%) belief that for SOPI among YLs to be effective, they needed to provide lots of support especially on how to carry out the interaction and how to complete the task (Teacher P11). They further explained that these supports should be based on the pedagogical knowledge and skills related to online classroom management (Teacher P9). Besides, given that the PI was conducted online, teachers explained that it was very necessary to provide training about technology and increase young learners’ technological literacy (Teacher P5).

For Factor 3 on which three items loaded highly, it concerned the grouping of YLs for SOPI. The mean score of the three items (M = 3.32, SD = 1.22) indicated the teachers agreed on the impact of grouping (i.e. mixed proficiency pairs and small group work) on the effectiveness of SOPI among YLs. The interview data suggested that many teachers (81.81%) believed grouping YLs of different proficiency levels in PI could cause issues such as kids’ imbalance in the amount of talking time, and dominance and passiveness in interaction (Teacher P1). In addition, all teachers (100%) expressed that pair work and small group work were appropriate for online interaction among kids and could work effectively when YLs were of similar proficiency level (Teacher P10).

Finally, Factor 4, which had two items loaded highly on it, concerned family backgrounds. The family backgrounds were related to parents’ educational level and financial condition. The interview showed the teachers (54.54%) expressed the importance of the involvement of parents in their children’s online learning: it was so important for parents to be guiding and keeping their young kids focused during synchronous online PI because kids could not concentrate long, and at times they chat about off-task topics. The parents also needed to know a bit of technology to resolve their kids’ issues such as internet disconnection. This was an issue for low-educated parents who did not know how to help their kids (Teacher P3).

Besides, the participants (36.37%) commented that financial situations partly affected kids’ experience in online interaction and learning because they used old slow-working computers and could not afford a decent internet broadband for good connection. They [kids] were just lagging all the time (Teacher P6). Although the qualitative data confirmed that financial condition and parents’ education level were factors affecting SOPI, the mean score of these two items was 2.49 (SD = 1.28), indicating that not all teachers shared a high level of agreement on this factor.

Teachers’ strategies to promote online PI’s effectiveness

To identify strategies perceived by teachers as effective for promoting the effectiveness of SOPI among YLs, an EFA was performed on four questionnaire items (Part D of the survey). The KMO value was 7.23, confirming the sampling adequacy, and the results of Bartlett’s test of sphericity were significant (p < .01). Following Kaiser’s criterion of eigenvalues being of 1, one factor was extracted (Table 3), explaining 67.24% of the total variance.

Table 3. Teachers’ strategies to promote effectiveness of PI.

Items	Factor loadings
Teachers’ pedagogical strategies (Factor 1)
•I teach my YLs how to work with peers during SOPI.	.83
•I teach my YLs how to do their tasks during SOPI.	.86
•I teach my YLs how to use their online interaction tools.	.75
•I facilitate SOPI among YLs throughout the process of task completion.	.84

In Table 3, all four items loaded highly on one factor, labelled teachers’ pedagogical strategies. These strategies focused on teachers’ facilitating YLs’ interaction during task completion and teaching them about how to work with peers, how to carry out the tasks and how to use online interaction tools. The mean score of these four items was 4.04 (SD = 1.01), indicating teachers’ high agreement on how these strategies were considered useful to promote the effectiveness of SOPI among YLs. In the individual interviews, all teachers (100%) stated that teaching kids how to work with peers, how to carry out task and how to use online interaction tools prior to their online interaction was essential for the success of synchronous online PI (Teacher P9). In addition, all the interviewed teachers stressed that it was necessary to monitor and facilitate synchronous online PI closely because the learners were young and thus could side-track during the interaction, especially when it was online (Teacher P6).

Implementing synchronous online PI and factors affecting its frequency

The teachers responding to the item PI activities are often used in my online classes (M = 3.87, SD = 2.09) in Part A of the survey indicated that they used PI often in their online classes for YLs. These quantitative results were supported by interview responses. Specifically, the teachers (63.63%) reported that they always or often used PI in their online classes. However, two teachers (18.18%) said that PI for kids was just used at times in their online classes and the other two teachers (18.18%) said they rarely used PI for kids. The interviews with the four teachers who said they used PI at times or rarely in their online classroom revealed that there were a variety of factors determining their degree of using PI in their online classes as summarised in Table 4.

Table 4. Factors affecting the teachers’ frequency of use of PI.

Items	Factor loadings
Schools’ regulations and resources (Factor 1)
•School policies	.84
•Curriculum mandates	.59
•Standardized paper-pencil tests as final assessments	.73
•Availability of online teaching resources	.84
•Availability of online learning resources for YLs	.65
Learners’ affective aspects (Factor 2)
•YLs’ willingness to interact with peers	.90
•YLs’ interest in online learning and interaction	.80
Parents’ support (Factor 3)
•YLs’ parental support regarding technical issues	.84
•YLs’ parental support regarding online learning issues	.89

To explore factors affecting teachers’ frequency of using PI for YLs, an EFA was performed on nine questionnaire items (Part E of the survey). The sample adequacy of this EFA was verified by the KMO value of .63 and the results for Bartlett’s test of sphericity were significant (p < .001). Following the Kaiser criterion for communalities after extraction in which eigenvalues greater than 1 was set, three factors were extracted, and they altogether explained 71.46% of the total variance.

In Table 4, five items loaded highly on Factor 1 which was concerned with schools’ regulations (e.g. policies, curriculum mandates and assessment) and resources for online teaching and learning. The mean score of these five items was 3.75 (SD = 1.14), indicating that teachers considered these school-related areas as factors determining their frequency of using SOPI for YLs.

The second factor (Factor 2), which had two items loading on it highly, concerned YLs’ affective aspects (i.e. willingness to talk to peers and interest in online learning and interaction). The quantitative results showed teachers’ high agreement (M = 4.09, SD = 1.20), indicating that YLs’ individual differences were of teachers’ concerns when deciding whether PI should be used. The final factor (Factor 3), which had two items loading highly on it, was related to parents’ support regarding technical issues and online learning issues. The mean score of the two items was 3.43 (SD = 1.21), indicating that parents’ support was important for the effectiveness of SOPI. These factors were also stressed by all teachers (100%) in the interviews as crucial for not only SOPI but also online learning.

Variations in teachers’ perceptions

To examine teachers’ perceptions of SOPI among YLs in relation to their teaching experience and previous training on online teaching, the survey responses were compared, and the results are summarised in Tables 5–7.

Table 5. Teachers’ perceptions of SOPI according to online teaching experience.

	Online teaching experience
	< 1 year (n = 8)	> 1 year & < 2 years (n = 30)	Mann-Whitney U Test
Perceived benefits of SOPI	M(SD)	M(SD)	z	p	r
•Focus on language aspects (Factor 1)	2.86(.81)	3.41(.71)	1.47	.14	.24
•Peer-related and interaction environment (Factor 2)	3.05(.89)	3.59(.49)	1.62	.11	.26
Factors affecting SOPI
•Learner individual differences (Factor 1)	3.51(1.00)	4.17(.94)	1.72	.08	.28
•Teachers’ provision of instruction (Factor 2)	3.72(.87)	4.38(.48)	2.11	.03	.34
•Grouping of learners in PI (Factor 3)	3.23(1.06)	3.67(.87)	1.06	.28	.17
•Family backgrounds (Factor 4)	2.48(1.15)	2.50(1.36)	.07	.94	.01
Strategies to promote effectiveness of PI
•Teachers’ pedagogical strategies (Factor 1)	4.01(.90)	4.15(.49)	.09	.93	.02
Factors affecting the frequency use of PI
•Schools’ regulations and resources (Factor 1)	3.57(.80)	4.37(.64)	2.52	.01	.41
•Learners’ affective aspects (Factor 2)	3.97(1.15)	4.56(.56)	1.31	.18	.21
•Parents’ support (Factor 3)	3.45(1.16)	3.37(1.25)	1.82	.85	.29

Note. An effect size (r = Z/√N) of 10, .30 and above .50 was considered small, medium and large, respectively.

Table 6. Teachers’ perceptions of SOPI according to teaching experience.

	Face-to-face teaching experience
	> 3 years (n = 28)	< 3 years (n = 10)	Mann-Whitney U Test
Perceived benefits of SOPI	M(SD)	M(SD)	z	p	r
•Focus on language aspects (Factor 1)	2.86(.81)	3.42(.71)	1.46	.14	.23
•Peer-related and interaction environment (Factor 2)	3.08(.89)	3.46(.60)	.94	.34	.15
Factors affecting synchronous online PI
•Learner individual differences (Factor 1)	3.57(1.05)	3.95(.84)	.90	.36	.14
•Teachers’ provision of instruction (Factor 2)	3.74(.89)	4.23(.37)	1.78	.07	.29
•Grouping of learners in PI (Factor 3)	3.23(1.09)	3.67(.69)	1.20	.22	.19
•Family backgrounds (Factor 4)	2.58(1.21)	2.12(.02)	.91	.36	.15
Strategies to promote effectiveness of PI
•Teachers’ pedagogical strategies (Factor 1)	3.98(89)	4.25(.50)	.67	.50	.11
Factors affecting the frequency use of PI
•Schools’ regulations and resources (Factor 1)	3.77(.83)	3.77(.91)	.28	.77	.05
•Learners’ affective aspects (Factor 2)	4.00(1.17)	4.43(.49)	.53	.59	.09
•Parents’ support (Factor 3)	3.40(.15)	3.56(.26)	.61	.53	.10

Note. An effect size (r = Z/√N) of 10, .30 and above .50 was considered small, medium and large, respectively.

Table 7. Teachers’ perceptions of SOPI according to previous training on online teaching.

	Training on online teaching
	Yes (n = 24)	No (n = 14)	Mann-Whitney U Test
Perceived benefits of SOPI	M(SD)	M(SD)	z	p	r
•Focus on language aspects (Factor 1)	2.97(.83)	2.97(.82)	.09	.93	.01
•Peer-related and interaction environment (Factor 2)	3.03(.84)	3.29(.86)	.59	.55	.09
Factors affecting synchronous online PI
•Learner individual differences (Factor 1)	3.47(.13)	3.75(.94)	.93	.35	.15
•Teachers’ provision of instruction (Factor 2)	3.66(.72)	3.97(.90)	1.54	.12	.25
•Grouping of learners in PI (Factor 3)	3.14(.01)	3.43(.05)	.70	.48	.11
•Family backgrounds (Factor 4)	2.03(.24)	2.75(.07)	1.84	.07	.29
Strategies to promote effectiveness of PI
•Teachers’ pedagogical strategies (Factor 1)	4.05(.74)	4.03(.88)	.17	.86	.03
Factors affecting the frequency use of PI
•Schools’ regulations and resources (Factor 1)	3.18(.72)	3.70(.91)	.52	.60	.08
•Learners’ affective aspects (Factor 2)	4.03(.11)	4.12(.07)	.32	.74	.05
•Parents’ support (Factor 3)	3.10(.14)	3.62(.15)	1.52	.12	.24

Note. An effect size (r = Z/√N) of 10, .30 and above .50 was considered small, medium and large, respectively.

In Table 5, there were no significant differences among teachers regarding their perceived benefits of SOPI and strategies promoting the effectiveness of PI among YLs. However, teachers with more online teaching experiences (greater than 1 year) had significantly stronger beliefs (with medium effect sizes) about the impact of ‘their provision of instruction’ (Factor 2) and ‘school’s regulation and resources’ (Factor 1) on the effectiveness of SOPI and their use of PI in their online classes, respectively. Regardless of online teaching experience, all teachers shared similar beliefs about the impact of YLs’ individual differences, grouping, family background, YLs’ affective attitude, and parents’ support on the effectiveness and use of SOPI.

Despite differences in teachers’ perceptions of factors affecting the effectiveness and the frequency use of SOPI among YLs, Table 6 shows that face-to-face teaching experience did not impact on any aspects of their perceptions of SOPI among YLs. Similarly, whether the teachers received pedagogical training on online teaching did not affect their perceptions and the frequency use of SOPI for YLs since no significant differences were observed between two groups of teachers (Table 7).

Discussion

This study explored teachers’ perceptions about various aspects of PI for YLs in online language classes. The results show that teachers considered SOPI facilitative of drawing YLs’ attention to language issues related to grammar, pronunciation, and spellings. These results are in line with previous research suggesting that PI promotes learners’ attention to language form (Dao & McDonough, 2017; Sato, 2013). Additionally, the teachers in this study felt that their Vietnamese YLs tended to focus more on pronunciation and grammatical aspects as opposed to lexical items. One possible explanation is that the teachers often drawn learners’ attention to pronunciation and grammar given that these are two main areas of concern and weaknesses that teachers felt need to be addressed for Vietnamese YLs. The teachers said YLs liked to correct each other’s grammar and pronunciation since they felt helpful and proud when doing so. Also, these young Vietnamese EFL learners were taught grammar and pronunciation explicitly in their language classes at school, so they seemed to have grammatical and pronunciation knowledge, which thus enables them to comment on each other’s grammar and pronunciation issues.

Another finding was that teachers believed SOPI created a comfortable and stress-free context for YLs to provide and receive support and feedback from each other. These results indicate that during SOPI, young learners were not only able to provide each other with feedback, which is essential to creating L2 learning opportunities, but also willing to and not afraid of ‘insulting’ each other. The results seem to point to the characteristics of YLs who, as on teacher claimed, ‘often feel proud of being helpful’ to others when correcting each other and might have not yet been fully aware of ‘being polite or impolite’, thereby comfortably commenting on each other’s language issues. These results thus indicate differences in YLs’ behaviours in SOPI as opposed to adult learners who at times do not feel comfortable, thus constraining themselves from providing feedback due to avoiding others’ face-losing or the possibility of harming their relationship (Naughton, 2006).

In addition, the results revealed multiple factors that teachers believed to affect the effectiveness of SOPI for L2 learning. The first factor concerned learners’ individual differences such as low proficiency level. These results supported previous research that reported teachers’ similar concerns about the effectiveness of PI (Dao & McDonough, 2017, see also Sato & Ballinger, 2016). However, previous research suggests that with careful guidance and procedure (i.e. task repetition) and interesting tasks, very young learners (e.g. 10-year-olds) with low proficiency could carry out PI effectively (Pinter, 2007). Thus, it seems important to raise teachers’ awareness about the possibility of implementing PI effectively among young and low-proficiency learners. The results also revealed that teachers believed learners’ relationship or interests in interacting with peers determines the effectiveness of SOPI. the teachers said that ‘some kids would start crying when paired with those they do not like’ (Teacher 8) and ‘some kids were not really interested in learning online and they just kept silent’ (Teacher P4). These results tend to point to the characteristics of YLs who often express explicitly emotions and preference, which therefore needs to be taken into consideration when implementing SOPI for YLs.

Moreover, the results show that teachers believed their instruction on how to interact with peers, how to do the task, and how to use online tools affected the effectiveness of SOPI among YLs. Previous research suggests that even for adult learners they need modelling (Kim & McDonough, 2011) and training on how to interact, provide feedback, and use communicative strategies to perform tasks effectively (Dao, 2020; Sato, 2013; Sato & Lyster, 2012). For YLs, teachers’ instruction and guidance even become more important given that learners are young and interact online. Thus, the teachers suggested that one of the ways to create productive SOPI among kids and address its issues was to provide guidance on various aspects and facilitate YLs’ interaction (Table 3).

The results also show that teachers perceived grouping learners of different levels would create some negative impacts on young learners’ SOPI such as learners’ dominance and/or passiveness in interaction or unequal amount of talking time. These results were similar to previous research’s findings on PI among adult learners who demonstrated different negative dynamics such as dominant-passive or passive-passive patterns of interaction due to mixed proficiency level (Dao & McDonough, 2017; Storch, 2002; Watanabe & Swain, 2007, 2008). However, previous research suggested that mixed-proficiency pairing could be conductive to learners’ greater attention to language issues when they were assigned with appropriate task role, carefully managed, or modelled on how to interact collaboratively (Dao, 2020; Dao & McDonough, 2017; Kim & McDonough, 2011). Thus, it is important to raise teachers’ awareness about the possible benefits of mixed-proficiency pairing of YLs in SOPI. Besides, the teachers reported that grouping learners in pairs or groups instead of doing teacher-led whole class interaction would positively affect their PI. This indicates that teachers appear to embrace the learner-centered approach, where learners are given more opportunities to talk and take control of their activities (Li & Walsh, 2011; see also Carless, 2004, 2007).

Notably, the teachers reported that YLs’ family background (parents’ assistance and finance) contribute to the effectiveness of SOPI. The teachers’ reports of the importance of parents’ involvement in young learners’ SOPI seem to be typical given that kids are believed not able to hold attention or concentration during an entire online session and might not be able to handle emerging technological issues. These results indicate significant differences between YLs and adult learners during SOPI. That is, teachers believed that parents’ participation in YLs’ SOPI would help YLs become better focused during the interaction and resolve technology-related matters that might occur. However, the teachers pointed out variations in the effectiveness of parents’ help to YLs during SOPI due to their educational background differences. The results seem to indicate a typical context of the Vietnamese participants where low-educated parents of some learners did not know how to help their kids as opposed to those parents who had high level of education, experience and knowledge of technologies, thereby being able to help kids during the SOPI. Besides, the results indicate that YLs’ financial condition contribute to the effectiveness of SOPI because some learners might not be able to afford a decent Internet connection contract and a computer device. However, the results should be interpreted with care and further explored given that not all teachers agreed on this aspect (see Factor 4 in Table 2).

Regarding teachers’ frequency of using SOPI for YLs, the results show that although teachers reported their strong belief in the need of using PI in their online classes for YLs, they reported multiples factors (e.g. school regulation and resources, learners’ willingness and interest, and parental support) affecting their actual frequency of use (Table 4). These results support previous research that documented similar contextual factors affecting the use of PI (Arshad, 2017). However, it seems that these factors, especially Factors 1 and 2 (i.e. school regulations, resources, and learners’ characteristics) were related to teachers’ perceptions rather than actual constrains of the context or learner-related issues, which thus could be possibly addressed via pedagogical adjustments. Notably, teachers noted that the parental assistance to YLs during SOPI is necessary, indicating a very distinctive characteristic of SOPI among YLs and thus needing to be taken into consideration when it comes to teaching YLs online. Despite focusing on young learners’ SOPI only, these results raise an interesting question of whether YLs should study and interact online when parental support is not available.

Furthermore, the results revealed that there were no significant differences in teachers’ perceptions of SOPI among YLs in relation to their face-to-face teaching experience, but they differed in their perceptions in terms of the impact of their provision of instruction/training for YLs and school regulations and resources on the effectiveness and frequency of using SOPI, respectively. These results suggest that teachers’ face-to-face teaching experience (6.64 years on average) did not seem to affect their perceptions of SOPI among YLs. However, their short online teaching experience (1.46 years on average) appeared to impact their perceptions, with teachers having more than one year of online teaching experience holding stronger beliefs in the role of their instruction on training YLs on various aspects in determining the SOPI’s effectiveness. Also, teachers with more online teaching experience held stronger beliefs in the impact of school regulations and resources on their frequency of using SOPI for YLs. The results also pointed out to an interesting relationship in which teachers’ face-to-face teaching experience, regardless of the amount, did not seem to be related to their perceptions of SOPI, which suggests that for SOPI to be implemented effectively, greater online teaching experience seems to be needed.

Finally, the results show that whether teachers received training on online teaching did not affect their perceptions and frequency of using SOPI for YLs. These results seem to differ with findings from previous research which suggests the significant impact of teacher education and training on teachers’ perceptions and teaching practices (Borg, 2011; Nguyen & Walkinshaw, 2018). A possible explanation for these contrasting results is related to the quality, the length, and the focus of training that teachers of the current study received. That is, although the teachers reported to receive training, their training was generally brief and focused mainly on mechanical technological aspects (i.e. how to use certain software and run some apps or breakout rooms). Teacher P7’s comment clarifies this aspect.

For the training, to be honest, we were just asked ‘Do you know this apps? Do you know Zoom? Do you know Google meet or do you know the Messenger?’ And then we were given like how to operate it and how to add the students and then the rest is up to you. We and our students will just figure out ourselves, research ourselves, find out for ourselves. It’s like that. (Teacher P7).

Teachers P7’s comment indicates that training on online teaching was limited in terms of both quality and quantity, which thus could partly explain the lack of impact of the training on teachers’ perceptions, particularly of SOPI among YLs. Thus, a more well-designed and high-quality PD training seems necessary for creating positive impacts on teachers’ practices.

Conclusion

This study explored teachers’ perceptions of various aspects of SOPI among YLs. The results showed that PI was perceived to promote learners’ attention to language and create facilitative peer-supported learning environment. However, its effectiveness was believed to be affected by multiple factors. To address issues and enhance PI’s effectiveness, the teachers suggested different pedagogical strategies. Additionally, although they agreed on the importance of using SOPI for YLs, multiple learner- and context-related factors affected their use of SOPI. The results also revealed no impact of previous brief trainings on online teaching and face-to-face teaching experience on their perceptions of SOPI, but it seems that online teaching experience, albeit short, mediated their perceptions of SOPI. The results overall suggest that the teachers hold positive perceptions of SOPI for YLs and were aware of learner- and context-related factors affecting PI’s effectiveness and its frequency of use. The results point to the necessity of raising teachers’ awareness about the possibilities of addressing PI’s issues via pedagogical adjustments and the need of more effective training for teachers on implementing SOPI.

It is necessary to note some limitations of the current study. This study primarily focused on teachers’ perceptions of SOPI but did not investigate teachers’ practices of SOPI. Given that there might be variations between teachers’ perceptions and actual practices, it is important to document the frequency and how teachers use SOPI in online classes using observations rather than relying on teachers’ self-reports as in this study. Also, the study did not include learners’ perceptions of SOPI, so it is necessary to explore SOPI from both teachers’ and learners’ perspectives for comparison. Methodologically, categorizing teachers into groups of comparison based on the number of years of teaching experience might not reflect the actual difference in the teachers’ pedagogical and social knowledge accumulated over years, so future research might need to collect further qualitative evidence regarding this issue to triangulate with the quantitative results reported in this study. Finally, the study focused on teachers of homogenous groups of learners in one EFL context, so the results might not be generalisable to other contexts (e.g. ESL contexts). Despite the limitations, the study contributes to our general understanding of teachers’ perceptions of SOPI for YLs and its use in EFL contexts.

Disclosure statement

No potential conflict of interest was reported by the authors.

Appendix A

Introduction: This survey is to explore teachers’ perceptions of and implementation of online peer interaction between young Vietnamese EFL learners in online classes.

Section 1: Background information and context

Instruction: Please provide short answers to the following questions

1. What is your first language (mother tongue)?

2. How old are you?

3. What is your gender?

4. How do you rate your English proficiency from A1 to C2 according to CEFR?

5. What is your highest degree?

6. Which area of Vietnam and which education sector are you working in? Public or language centers?

7. Where is your

8. How many years have you taught English?

9. How many years have you taught English online?

10. Did you participate in a training on online teaching? If so, please describe the training that you received?

11. Which software/platform are you using for your online classes?

12. What is your students’ age range in your online classes?

13. What is the average size of your online English classes?

Section 2. Your opinions about online peer interaction

Instruction: Please indicate your opinion regarding the following statement by selecting a number on the scale from 1 (strongly disagree) to 5 (strongly agree). There are no right or wrong answers in this survey, and your responses will be used for research purposes only.

Note: PI = Peer interaction

Table

Part A. Your perceptions of PI in online classes
PI among Vietnamese young learners should be promoted in online classes.
Online classrooms are suitable environments for using PI.
PI activities are often used in my online classes.
Part B. Benefits and reasons for using synchronous online PI among young learners
SOPI promotes young learners’ focus on word meaning.
SOPI promotes young learners’ focus on the spellings of words.
SOPI promotes young learners’ focus on pronunciation aspect of language.
SOPI promotes young learners’ focus on grammatical aspect of language.
SOPI promotes a stress-free learning environment for young learners to learners with peers.
SOPI promotes young learners’ willingness to provide feedback to peers.
SOPI promotes young learners’ positive feelings when receiving peers’ feedback.
SOPI promotes peer supports among young learners in task completion.
Part C. Factors affecting synchronous online PI
Young learners’ English proficiency affects the effectiveness of their SOPI.
Young learners’ interest in interacting with peers affects their SOPI.
Young learners’ interest in online learning affects their SOPI.
My instruction on how to interact with peers affects young learners’ SOPI.
My instruction on how to do the tasks affects young learners’ SOPI.
My instruction how to use online interaction and learning tools affects young learners’ SOPI.
Grouping young learners of different levels affects negatively SOPI.
Grouping young learners of similar levels positively affects SOPI.
Pair work positively affects young learners’ SOPI.
Small group work positively affects young learners’ SOPI.
Whole class interaction affects negatively young learners’ SOPI.
Young learners’ financial background (rich or poor families) affects their SOPI.
The assistance of parents in the online class positively affects young learners’ SOPI.
Part D. Your strategies to promote effectiveness of PI
I teach my young learners how to work with peers during SOPI.
I teach my young learners how to do their tasks during SOPI.
I teach my young learners how to use their online interaction tools.
I facilitate young learners’ SOPI throughout the process of task completion.
Part E. The following factors determining the frequency of using synchronous online PI
School policies
Curriculum mandates
Standardized paper-pencil tests as final assessments
Availability of online teaching resources
Availability of online learning resources for young learners
Young learners’ willingness to interact with peers
Young learners’ interest in online learning and interaction
Young learners’ parental support regarding technical issues
Young learners’ parental support regarding online learning issues

Appendix B

Interview prompts

• In your opinion, should peer interaction among young learners in online classes be promote? If so, why?

• How often do you use PI in your online classes?

• Do you think online classrooms are suitable environments for using peer interaction?

• In your opinion, what are the benefits and issues of online peer interaction among young learners?

• What factors in your opinion affect the effectiveness of online peer interaction among young learners?

• What are your strategies to promote the effectiveness of online peer interaction if it is used in your online classes?

• Are there any factors affecting your frequency of using peer interaction between young learners in your online classes?

• Did you receive any training on teaching young learners online? If so, can you describe the training?

• Is there anything else that you would like to share about your use of online peer interaction and your online teaching of young learners in general?