According to studies, incorporating quality characteristics in physical education (PE) will result in a comprehensive development in students’ lives in various domains including physical and motor skills (Emmanouel et al., 1992; Eather et al., 2013), academic performance (Rasberry et al., 2011; Marques et al., 2017), cognitive function (Ardoy et al., 2014), self-perception (Goni and Zulaika, 2000), self-determination and motivation (Shen et al., 2009; Wallhead et al., 2014; Cuevas et al., 2016), self-efficacy and enjoyment (Dishman et al., 2004, 2005) and social interactions and reciprocal relationships (Wallhead et al., 2013; Garn and Wallhead, 2015). Globally, there is a shift to holistic PE that incorporates social and cultural factors (Lynch, 2019). The emergence of holistic PE in curriculums can help countries meet some of the United Nations’ Sustainable Development Goals such as quality education and gender equality (Baena-Morales et al., 2021). Quality PE fosters a healthy lifestyle and enjoyment in students, and they can learn social and interactive skills (National Association for Sport and Physical Education, 2013). Social interaction plays an important role in the students’ skill-learning process outside of their comfort zone, and social activities can shape their learning process (Koekoek and Knoppers, 2015).

Recent studies have also demonstrated young people’s general acceptance of applications related to health and learning settings (Casey et al., 2017; Goodyear et al., 2019; Lee and Gao, 2020). For this reason, teachers should consider using applications and tablets (e.g., iPads) to help students achieve PE learning goals. They are also encouraged to use a selection of apps for efficient classroom management or for emphasizing other learning domains (Lee and Gao, 2020). Casey et al. (2017) have argued that with the appropriate pedagogical considerations, digital technology can offer potential benefits to students’ learning and performance. Recently, there has been a rise in literature exploring app-integrated PE in which various apps are used on the iPad in PE to facilitate the teaching and learning process (Watterson, 2012; Zhu and Dragon, 2016; Krause et al., 2020; Thacker et al., 2021). As part of this growing trend, tablets have become popular in PE, as they provide a portable multimedia platform to download a host of applications to accomplish a variety of tasks (Weir and Connor, 2009; McManis and Gunnewig, 2012; Koekoek et al., 2018).

It is critical that when technology is being used that it has appropriately scaffolded lesson plans to keep students on-task, engaging them in higher levels of thinking and performance (Ward et al., 2005; Yelland and Masters, 2007; Mosston and Ashworth, 2008; Metzler, 2011; Koehler et al., 2013). Because of these benefits, applications need to be carefully constructed to adhere to a pedagogical curriculum (Lee and Gao, 2020; Yu, 2020). Since it is necessary to develop curriculum programming for PE (Wallhead et al., 2021), effective use of technology, specifically through tablet computers, may offer a way to provide scaffolded content knowledge in a PE classroom (Gubacs-Collins and Juniu, 2009; Leight et al., 2009; Sinelnikov, 2012; Sun, 2012; O’Loughlin et al., 2013; Palao et al., 2015; Zhu and Dragon, 2016).

It has been shown that the traditional teaching methods in PE limit the student to the reproduction of observed behavior (Pritchard et al., 2008; Moy et al., 2014). Observed behavior involves the teacher modeling a skill followed by the students attempting to replicate the movement and practicing privately while receiving feedback individually from the teacher (Mosston and Ashworth, 2008). As a kind of instructional technology, video-modeling (VM) encompasses the individual watching a video of visual and auditory processes, which is modeled by a peer, a sibling, an instructor, an athlete, or the self, and then, similarly, performing the sequence of skills (Mechling, 2005; Bellini and Akullian, 2007). Video-modeling is a unique feature that is made possible through a combination of multiple technologies such as iPads, video cameras, and health-related apps and wearable devices (Aldi et al., 2016). This feature can reinforce observational learning through video analysis (Koekoek et al., 2019; Lee and Gao, 2020; Thacker et al., 2021). It is also easier to keep track of activities, and monitoring and evaluation can be facilitated by technology (Jeong and Hmelo-Silver, 2016), which is a suitable apparatus for improving students’ understanding and assessment of their movements while promoting their shared understanding (Koekoek et al., 2019; Yu, 2020; Thacker et al., 2021).

Further, watching VM could result in decreasing mental effort by focusing attention on the relevant stimuli (Charlop-Christy et al., 2000) and improving learning achievements (Kant et al., 2017). It can also facilitate retention of the skill elements and is attractive for children of all ages (Charlop-Christy et al., 2000). VM has been reported as an effective method for teaching, and for the acquisition of fundamental movement skills (Obrusnikova and Rattigan, 2016; Obrusnikova and Cavalier, 2018; Thacker et al., 2021), locomotor skills (Bulca et al., 2020), sporting skills such as throwing the discus and hammer (Maryam et al., 2009), and gymnastic skills (Boyer et al., 2009; Bouazizi et al., 2014) as well. It has also helped enhance positive social interactions in a natural setting focused on free play, and social and emotional development (Green et al., 2017). Another study reported improved on-task behavior after an intervention including self-monitoring and VM to math tasks (King et al., 2014).

Collaborative learning is another advantage provided by technology (Resta and Laferrière, 2007; Jeong and Hmelo-Silver, 2016; Bodsworth and Goodyear, 2017), which has the potential to foster peer-to-peer interactions (Resta and Laferrière, 2007). Peer-to-peer (P2P) learning occurs when peers are directly involved in the teaching and evaluation process, increasing student engagement, collaboration, and responsiveness (Byra and Marks, 1993; Topping and Ehly, 1998; Jenkinson et al., 2014; Kelly and Katz, 2016; Thacker et al., 2021). P2P learning is already being adopted in many situations due to increased engagement, improved performance, and constructive behaviors (Chai and Koh, 2017). Overall, technology may improve peer learning (Li and Gao, 2015) and provide support for PE, developing positive social interactions among students (Krause et al., 2020).

Video-modeling can also shape the aggregate and nonspecific feedback and make them specific and related to the task (Jeong and Hmelo-Silver, 2016). Enriching the quality of how feedback is provided can play a critical role in fostering effective peer evaluation and peer learning (Zher et al., 2016). High-quality constructive feedback could also help improve the learning process (Li et al., 2010; Waggoner Denton, 2018) and strengthen social interactions between peers (Lee and Gao, 2020).

Finding an effective combination of collaborative technology and scaffolding pedagogy through peer learning is important (Goodyear et al., 2014; Fernandez-Rio et al., 2017). The P2P learning model, as an effective procedure for collaborative learning, can be enhanced through pedagogical strategies and scaffolding while fostering social interactions (Jeong and Hmelo-Silver, 2016). In addition, there are still key issues regarding the suitability of technology and software for young learners that must be addressed. Measuring the quality of interaction between students offers a pathway to understanding more effective teaching and learning strategies in a P2P environment, which could also lead to improving collaborative performance and evaluation (Johnson and Ward, 2001; Jenkinson et al., 2014).

Using VM, the purpose of this study was to examine the nature and quality of interactions between elementary students in three different P2P evaluation methods and conditions. The nature and quality of the peer interaction, not necessarily the performance of the skill, was of interest. To assess student interactions, we compared three different evaluation methods and conditions: Structured Paper, Unstructured Video, and Structured Video. Nature (the duration and comments’ type) and quality of interaction (we developed four categories related to the specificity of feedback) were determined by videotaping the conversations between peers and systematically assessing them based on oral transcripts.

Sixty-two interactions were collected over 3 days of data collection: 20 from the SP group, 24 from UV group, and 18 from SV group. Four files were lost from the SV group due to technical errors during data transfer.

Table 3 displays the Estimated Marginal Mean and Standard Error of variables for evaluators by the duration of interaction and comment type. The GEE analysis revealed statistically significant condition differences for total time (p < 0.05), on-task time (p < 0.001), off-task time (p < 0.01), and positive comments (p < 0.01) for the evaluator role (Table 3).

Table 4 provides the pairwise comparison analysis for total duration, evaluator on-task time, evaluator off-task time, and evaluator comments type by evaluation condition. Pairwise comparison further revealed that for the total duration, significant differences were seen between the SP group (56.5 ± 5.8) and SV group (84.5 ± 10.1), and between UV (54.0 ± 4.8) and SV group (84.5 ± 10.1). For evaluator on-task time, significant differences were seen between the SP group (46.4 ± 5.5) and SV group (76.6 ± 9.1), and between UV (39.4 ± 3.9) and SV group (76.6 ± 9.1). For evaluator off-task time, significant differences were only seen between the UV group (6.6 ± 2.4) and the SV group (0.5 ± 0.4). Regarding evaluator comment types, significant differences were seen between the SP group (0.8 ± 0.2) and SV group (2.5 ± 0.4), and between UV (1.1 ± 0.2) and SV group (2.5 ± 0.4). Evaluators in the SV groups had more positive feedback overall.

Table 5 displays the Estimated Marginal Mean and Standard Error of variables for evaluators and performers by quality of feedback recorded. GEE analysis revealed statistically significant condition differences for evaluator comments in Category 2 (p < 0.01) and in total comments (p < 0.05). Category 0 and Category 1 showed trends (p < 0.10) and there were no differences in Category 3 level comments.

Table 6 shows the pairwise comparison analysis for quality of evaluator comment rating between groups. Pairwise comparison further revealed that there was a significant difference between SP (2.34 ± 0.45) and SV (1.02 ± 0.31), and UV (0.76 ± 0.26) groups in Category 2 feedback. In total feedback, there was a significant difference between the SP (6.31 ± 0.68) and UV groups (4.68 ± 0.55).

The results of the performers were also analyzed using the same methods as the evaluators. For performer engagement time and comment types, significant differences were found only for time on-task (p < 0.01) between the UV (36.4 ± 3.9) and SV group (71.4 ± 8.9), and positive comments (p < 0.05) between the UV (0.5 ± 0.1) and SV group (1.7 ± 0.2). No other significant differences were found between the total time, time off-task, and negative and constructive comments. In terms of the quality of feedback, pairwise comparison further revealed a significant difference in Category 2 feedback for performers (p < 0.05) between the students in the SP (0.92 ± 0.32) and UV groups (0.18 ± 0.10).

The purpose of this study was to examine the nature and quality of interactions between peers in an elementary PE class using VM and in a peer-evaluation setting with three different P2P evaluation methods. Nature of interaction was determined by the duration of interaction (total time, on-task time, off-task time, neutral), and the type of comments (positive, constructive, negative). The quality of interaction was measured by the four categories of feedback from both the evaluators and performers. The evaluators in the SV condition (video playback with structured evaluation) spent significantly more time on-task while interacting with their partners and spent less time off-task compared to the SP and the UV conditions. Students (both evaluators and performers) were observed following the instructions and reviewing the video of their performance under the SV condition. Further, they provided more feedback that is positive to their peers than when under the SP and UV conditions. Similarly, participants in the performer role spent more time on-task in the SV condition compared to the other conditions. When provided with video feedback and playback options, the students had an objective way to evaluate their partners’ performances instead of solely relying on their memory of them. There was statistically significantly less off-task time for the SV condition compared to the SP or the UV feedback conditions. When in the SV condition, students spent more time on-task and increased their time interacting with their peers than in the other teaching methods. While there was an increase in positive feedback with the use of SV, no difference was observed in the frequency of constructive and negative comments between the three conditions for both the evaluator and performer roles. When examining the quality of the positive feedback given, there was more Category 2 feedback given by the SP group compared to the other two conditions (Table 6). There were no other significant differences observed when examining the quality of the feedback between the three conditions.

To our knowledge, this is the first study that examined the nature and quality of social interactions between peers based on a tablet-based intervention providing P2P learning along with VM, specifically while learning a basketball skill in three different evaluation conditions. Compared to the SP and UV conditions, the SV group increased on-task time behavior and had more positive feedback during discussions between peers, which shows that students are more likely to encourage their peers when there is video playback and structured assessments. This finding is in line with Law and Baer’s (2020) study which reported that by facilitating peer assessment using a structured software program, students improved their ability to provide quality feedback regarding their classmates’ writing skills.

Bartimote-Aufflick et al. (2016) noted that when students’ psychological needs are met, they become more actively involved in the learning process. According to their review of 64 studies, pedagogical strategies that could help students meet their needs include using videos, structured learning activities, and task analysis. Structured peer evaluation using technology, in combination with access to video recall, could lead to more engagement in task performance while decreasing the frequency of unrelated or disruptive behaviors (King et al., 2014). According to Obrusnikova and Rattigan (2016), using the VM technique improved fundamental movement performance through increasing the students’ attention and motivation to the skill complexity and production. Furthermore, when students provide peer feedback, it makes them think more critically about the assessment criteria, engaging them in the process for a longer period of time, and ultimately, sustaining that process (Zher et al., 2016). This is likely the case because cooperative learning and peer evaluation foster positive experiences and skills such as enjoyment, connection, teamwork, and creativity (Fernandez-Rio et al., 2017).

Another possible reason for finding significant improvement in positive comments and extending on-task time for the SV group could be the use of a rating criteria for peer assessment. This observation aligns with previous research that showed having explicit rating criteria makes the peer evaluation process more fruitful (Li and Gao, 2015), leading to students reaching more learning goals. In general, integrating technology in peer assessment seems to be more beneficial in the learning process compared to paper-based peer evaluation. This is supported by Li et al.’s (2010) study which reported that computer-mediated peer evaluation was associated with greater learning achievement compared to paper-based peer evaluation. The efficacy of technology-based feedback on students’ motivation and learning has been well-documented (Roure et al., 2019).

The findings in this research show the potential of structured feedback with VM to provide positive feedback during the peer-learning process and enhance the quality of interactions between students in peer-assessed performance analysis. One of the goals of physical education involves psychomotor performances that can be effectively captured by digital video. While digital video is beneficial for the PE classroom, providing structure for evaluation, such as a prototype app in this study, can further enhance engagement. Designing digital video tools that incorporate structure and can provide feedback for the self and in peer evaluations leads to increasing participant engagement and meeting knowledge outcomes.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Conjoint Health Research Ethics Board, University of Calgary. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

NS was the lead author, participated in the design, led the data collection, managed the data, interpreted the audio recordings, prepared the first draft of the study including the tables, and contributed significantly to each draft (six revisions). HL participated in the study design, had a major role in the data collection, and provided feedback on the drafts. VB helped to organize the study, assisted in the data collection, edited the first draft, and conducted the initial literature review. TF was the team statistician, analyzed the data, provided advice and feedback on the results, and assisted with the study design. HR reviewed the literature for both the introduction and discussion, expanded the study focus, added creatively to the discussion, and provided feedback on the drafts. LK developed the application, designed the study, participated in the data collection, edited the drafts, and supervised the team. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.