The effects of a tactical games model unit on students’ volleyball performances in elementary school

Abstract

The purpose of this study was to examine the effects of a tactical games model instructional plan on game-play volleyball performances of elementary school students, taking into account their skill level. In total, 39 fourth-grade students (average age: 8.9 years) participated in a 13-week unit, in which each lesson exaggerated the use of small-sided games. In-game performances were assessed via the Team Sport Assessment Procedure, while students played a 10-minute modified game (four versus four). Data were collected pre- and post-intervention, and after the summer vacation (retention test). A 2 (skill level) × 3 (time) analysis of variance with repeated measures was used to compare students’ performance, and the relevant effects were interpreted mainly by means of confidence intervals and effect size measures. At the end of the instructional period, all participants had an overall moderate to large improvement, and this global improvement seems to have remained at least until the end of the summer vacation. Lower-skilled students attained a larger and more established improvement than high-skilled students did. However, some detrimental effects on in-game students’ performance existed at the end of the instructional period. Therefore, teachers have to take into account students’ skill levels when designing their lessons because, if small-sided games are adequately considered and managed, students’ learning processes can be enhanced. Furthermore, the students should be assigned appropriate learning activities to avoid summer learning loss in physical education.

Keywords

Physical education skill level small-sided games sport education longitudinal study

Introduction

Siedentop (1998) suggested that to play well in a variety of sports and games, a student must become a competent sportsperson – that is, a student with sufficient skill competence (i.e. technical proficiency) and adequate games knowledge (i.e. tactical awareness). In this regard, sport pedagogy researchers have performed a great deal of research on several methods for teaching sport-related games during physical education (PE) (Mitchell et al., 2020). Among the findings that emerged from these studies, the following were the most shared: (a) learning processes have to be designed to preserve a motor task’s complexity (i.e. task simplification) rather than its decomposition into several motor programmes before they are rebuilt together (i.e. part-task decomposition; Seifert and Davids, 2017); (b) teachers have to create the conditions that facilitate an on-going learning process of sport concepts and skills that emerge by means of an adaptive and functional relationship between the task condition, improving skill-technical motor learning, and developing tactical cognition process, rather than by precise description and prescription of movement patterns and tactics (Nathan, 2018). In this respect, using a game-centred approach (GCA) has been widely recognized in the literature as a valid, effective, and authentic teaching strategy (Harvey and Jarrett, 2014; Mitchell et al., 2020). This model’s pedagogical perspective is to organize student-centred and situated learning experiences that stimulate the intertwined development of technical and cognitive skills (Kirk and MacPhail, 2002) via teaching strategies (i.e. the exaggeration of modified game play), which “link tactics and skills by emphasizing the appropriate timing of skill practice and application within the tactical context of the game” (Mitchell et al., 2020: 3).

Teaching games for understanding (TGfU) was identified as the forerunner of the GCA philosophy, but several methods have further developed to extend this model (Hodges et al., 2018). In this regard, the tactical games model (TGM) has been developed as an extension because of the following: it uses a simplified framework to develop a lesson (i.e. the three-stage versus the six-stage model of TGfU); it uses modified game play to facilitate the development of game sense; and it uses a variety of tactical complexity levels, distinctly for games and sports, which can be used as a proxy for learning progression (Harvey and Jarrett, 2014; Hodges et al., 2018). Mitchell et al. (2020) suggested the use of small-sided games (SSGs) to implement modified game play because they permit the teacher to develop a learning experience based on all students’ full involvement and on the simplification of an authentic tactical problem by modifying or adapting the length of a field, the number of players, and/or the rules of each selected game. In addition, by using SSGs, teachers can organize authentic and ecologically valid assessments to collect data on students’ performance. Furthermore, Van Acker et al. (2010) have identified the use of modified game play and small-sided teams as an effective strategy to support students’ learning and development through PE without the influence of differences in gender or skill level.

From a pedagogical perspective, it is relevant to consider students’ initial skill level to design developmentally appropriate learning processes that guarantee students of lower skill level an equal opportunity to learn (Silverman et al., 1999). In this regard, Mahedero et al. (2015) examined the effect of a sport education season on secondary school students’ performances of game play components (i.e. decision-making, skill execution, and game performance), when students were assigned to three different skill groups. They reported an overall improvement for all students at the end of the instructional period, but middle-skill-level students improved much more than the others. Similar game-play components were analysed in the study by Araújo et al. (2016), which divided their volleyball students (average age: 11.8 years) into two skill-level groups (i.e. lower and higher) and verified between-group differences – in terms of game play components, lower-skill students gained significantly more than the higher-skill students.

Nevertheless, although the importance of accounting for students’ skill level when examining the effects of instructional interventions is well known (Hastie et al., 2011), most of the research on GCAs does not consider students’ skill level as a mediating factor (e.g. Arias-Estero et al., 2020; Dania and Harvey, 2020). In addition, when Harvey and Jarrett (2014) reviewed research on using GCAs, they identified a lack of studies that utilized ecological assessment methods to address skill development related to GCA intervention and a lack of evidence related to long-term perspectives on GCA use.

Therefore, taking into account students’ skill levels, a long-term examination of the effects of utilizing a TGM instructional approach to teach volleyball skills to elementary school students can extend the current literature. Consequently, this study aimed to assess the effect of a longitudinal TGM-based instructional plan on game-play volleyball performances of elementary school students, when they were grouped by their skill levels (i.e. as a subsequent aim).

Methods

Design

Following Ployhart and Vandenberg’s (2010) suggestions about longitudinal research, we conducted an explanatory longitudinal study and hypothesized that the educational intervention would have been the main reason for explaining a nonlinear form of change in the students’ performances. By taking into account Harvey and Jarrett’s (2014) recommendations on sampling and selecting participants to guarantee ecological validity in education research, two intact classes were randomly selected to attend an alternative PE lesson. We conducted a pre-test (PreT) on students’ game-play performance before the instructional period began, a post-test (PostT) when the instructional period ended (a few days before the end of the school year), and a retention test (ReT) after the summer vacation, one week after the new school year started.

Participants and procedures

The study was nine months long, from January 2019 to October 2019, and 39 children (i.e. 18 girls and 21 boys; average age: 8.9 years) participated. In January, children attended the fourth year of an elementary school in the south of Italy, whereas in October they were in the fifth year. The instructional period was 13 weeks. All children were enrolled in two PE lessons per week, with each lesson lasting 60 minutes. The students had no previous experience with TGM and learning volleyball. All lessons and the assessments were performed in the school’s gym. The PE teacher had more than 15 years of experience as a physical educator. An expert in PE supported her during this project. The expert, who had a master’s degree in PE, was in the final year of his PhD course in PE, and had two years of experience in using TGM for teaching sport in secondary school, supported the teacher in applying the TGM framework during the design of the unit and helped her with the setup of the learning context. He did not teach any of the lessons. This expert’s involvement was sustained by the national project “Classroom sport”, which aimed to improve PE and sport activities in primary school (WHO, 2018). The Board of the School approved the current protocol, and all participants provided informed consent forms that their parents or legal guardians had signed. The Ethical Committee of the University of Enna approved the design and the methodological procedures used in this study.

TGM instructional plan

A games-based teaching intervention was designed, using net games and, at times, a hands-based invasion game. We used these game typologies to support a multi-sport learning process and to somewhat account for the transfer of learning. Volleyball was the net game used to develop game forms and task experiences throughout each lesson and was also used for students’ performance assessments. For these activities, according to the teacher’s evaluation of students’ volleyball abilities, 10 mixed-ability teams of four students each were formed. Across the assessments, each match was organized as a small-sided volleyball game: it was played four versus four, was 10 minutes in length, the size of the playing field was reduced (i.e. 6 m × 12 m), the net height was 1.60 m, and the ball was soft and provided an easy bounce. Furthermore, each match was scheduled to allow the same playing time for all students, and teams with similar skill levels played each other. We chose to play four versus four matches because the fourth player permitted us to increase the development of tactical understanding beyond the basic triad formation. The teams and match schedule remained the same across the assessments. According to the TGM framework (Mitchell et al., 2020), each lesson accounted for defining a tactical problem, a lesson focus, and an objective. Each lesson was organized according to this framework: Game 1, Practice task, Game 2, Closure. Game 1 was 15 minutes long and was used to introduce, through modified games, the lesson’s tactical problem. Practice task was 15 minutes long and was focused on developing movements and skills related to the lesson’s tactical problem. Game 2 was 20 minutes long and was used to reinforce the lesson’s focus. Closure was 10 minutes long and was organized as a questioning time used to verify how the students had focused on each lesson’s tactical problem and which strategies were proposed to solve that problem. Some modifications concerning volleyball rules and equipment were made during the initial stage of the unit to avoid loss of interest and disappointments in lower-skill students – for example, using an overhand serve was initially replaced with a movement from the bottom (i.e. underhand serve) and the possibility of two ball rebounds on the ground was permitted. Table 1 describes the main characteristics (i.e. tactical problem and/or focus) of the lessons throughout the unit.

Table 1.

TGM lesson plan with specifications of tactical problems and game-play characteristics.

Week	Tactical problem	Movements (off-the-ball) and skills (on-the-ball)
1	Pre-training assessment
2	Maintaining a rally	Moving to catch, reading, anticipating, and underhand throwing
3	Setting up an attack	Seeing court spaces, opening up to teammates, passing, and shooting
4	Winning a point	Attacking spaces, passing, and shooting
5	Defending space	Base positioning, covering the court as a team, and sliding
6	Defending against attacks	Backing up teammates and shifting to cover
7	Mini-tournament	Six-minute matches, four versus four
8	Initiating play	Underhand and overhand service and receiving the serve
9	Maintaining a rally	Tossing, setting, and striking sequences
10	Setting up an attack	Shooting for depth, approach shot, and dropping shot
11	Winning a point	Attacking spaces, making power versus accuracy decisions, and spiking
12	Defending against attacks	Base positioning, covering the court as a team, and sliding
13	Post-training assessment

Instructional and treatment validity

Following Metzler’s (2005) and Hastie and Casey’s (2014) suggestions about the need to develop an instructional process consistent with accepted standards and to respect the essential contextual conditions required to determine a particular teaching strategy’s influence on students’ learning, some actions were implemented in the period before the project started. First, the teacher and expert were previously trained for six months on using TGM in elementary school. The protocol was designed to emphasize the pedagogical implications (e.g. the teacher is a facilitator, students are active learners, etc.) Dyson et al. (2004) provided and was focused on the unit progression for the elementary school context that Mitchell et al. (2020) proposed. An external professor with several years of expertise in using TGM in elementary school delivered the protocol. Second, an external researcher with two years of expertise in using TGM positively reviewed the instructional plan the teacher developed through the benchmarks for faithful implementation of TGM, as Metzler (2005) suggested. Finally, the same researcher verified that the gym and the equipment available were adequate to maximize students’ involvement and achievements. Moreover, only those students who participated in 85% of the instructional plan and completed the three assessments were included in further analysis.

Game-play performance measures

Students’ game-play performance during three matches (i.e. one for each assessment) was rated using the Team Sport Assessment Procedure (TSAP; Grehaigne et al., 1997). This is an authentic assessment instrument for reflecting student technical learning in relation to the in-game context. Arias-Estero and Castejón (2014) asserted that the TSAP has been mainly used as a tool to assess learning progression at the end of the teaching-learning process. Even though this instrument has been validated for participants aged between 15 and 18 years, it has been used in studies with participants aged less than 15 years and it showed a predisposition for use with other ages (Arias-Estero and Castejón, 2014; Sgrò et al., 2018).

The TSAP’s components are associated with on-the-ball skills and are codified according to two sequential steps related to the following questions: “How did the player gain possession of the ball?” and “How did the player dispose possession of the ball?”. The answer to the first question may be conquering the ball from an opponent (CB) or receiving the ball from a teammate (RB). The answer to the second question can be chosen from the following possession outcomes, which have been customized from the original description to account for volleyball characteristics:

Playing a neutral ball (NB): the ball has passed among teammates (i.e. first or second touch within a rally), or the ball has passed over the net, but does not put the other team in jeopardy.

Losing the ball (LB): the ball is hit over the net, but the opposing team recovers it.

Playing an offensive ball (OB): this is identified with the second touch of possession (i.e. the pass from passer to hitter) when the hitter most often scores a point or puts pressure on the opposing team.

Executing a successful shot (SS): the team scores a point, or the ball is returned to the team following the execution of an attack.

According to these components, each player’s behaviour was summarized by means of the following performance indexes: volume of play (VP), efficiency index (EI), and performance score (PS). From a learning perspective, VP represents the player’s general involvement in the game; the EI represents each student’s effective play; and the PS serves as an overall performance indicator and is built by combining VP and EI (Grehaigne et al., 1997). The formulas used to estimate the aforementioned indexes followed the indications provided by Grehaigne et al. (1997).

Each match was video-recorded using two action-cameras (i.e. YI Camera, resolution: 720 p; frame rate: 60 Hz) located in two opposite corners of the volleyball court. At the end of each assessment, two skilled operators, who were preliminarily trained on using TSAP and had sufficient about- and in-game knowledge of volleyball, saw the videos and scored each participant’s actions according to the TSAP components. These operators were not involved in the instructional plan, they did not know the participants, and, to reduce the chance of memory biases, the number and the colour of the jackets the players wore were changed in each assessment. The software Kinovea was used to reproduce videos and to note each participant’s relevant actions (Kinovea, version 0.8.15, http://www.kinovea.org). The levels of agreement between the observations of these operators were verified following this procedure: (a) preliminarily, each operator examined each TSAP component and then they reached a consistent common accordance on component definition; (b) they scored each participant’s performance separately; and (c) for each assessment, the inter-operator reliability was assessed through an agreement percentage calculated by using the PS scores of eight randomly selected players (i.e. approximately 20% of the final participants and 24 performances overall). According to the recommendations for using TSAP (Grehaigne et al., 1997), the percentage of agreement between the operators had to be greater than 0.80.

Data analysis

The average values of the TSAP performance indexes were checked preliminarily to verify if assumptions required by Repeated Measures Analysis of Variance (RM ANOVA) have been met. More specifically, we checked for missing values, univariate outliers, and homogeneity of variance. According to the subsequent aim of this study, a non-hierarchical k-means cluster method, with the number of clusters fixed at two, was used to determine the skill level of groups according to the scores estimated for the TSAP indexes in the pre-intervention assessment. Then, we compared the students’ performances across the three assessments (i.e. PreT, PostT, and ReT) by means of 2 (skill level) × 3 (time) RM ANOVA. Main effects and interactions were estimated for each analysis and subsequent Bonferroni–Holm post hoc tests were used for multiple comparison. Concerning the interpretation of the results, according to Cumming’s (2014) suggestions about the need to go beyond dichotomizing null hypothesis testing, we reported different statistics for between- and within-group analysis, respectively. We provided the mean and 95% confidence interval (CI) of each group for between-group comparison, whereas we reported mean differences and the relative 95% CI for within-group analysis. Furthermore, the effect size (ES) of main effects and interactions were measured by means of eta-squared and interpreted with the following cut-offs: 0.01 = small; 0.06 = medium; and 0.13 = large (Cohen, 2013). Concerning multiple-comparison analyses, the ES was estimated by means of Cohen’s d measure and was interpreted with the following cut-offs: 0.2 = small; 0.5 = medium; and 0.8 = large (Cohen, 2013). The analyses were performed by means of R for Mac OS X and the alpha test was set to 0.05.

Results

The level of agreement between the operators resulted in 0.98 for PreT, 0.99 for PostT, and 0.95 for ReT; therefore, the recommendation about the reliability for using TSAP was verified. Data screening revealed that all the participants met the criteria discussed in the previous section (i.e. 85% presence rate during the intervention and in the assessments) and five students were univariate outliers. Because no other violations were verified, these participants’ data were removed, and parametric analyses were carried out by considering 34 students altogether (i.e. 18 girls and 16 boys). The skill-level groups, resulting from a non-hierarchical k-means cluster analysis, accounted for 16 students (i.e. seven girls and nine boys) in the higher-skill group and 18 students (i.e. 11 girls and seven boys) in the lower-skill group. To improve the readability of the clustering results in a practical sense, we provided the values of the three indexes for the closest between-cluster cases as a proxy of cut-off points: low-level-skill case: VP = 8, EI = 0.53, PS = 9.3; high-level-skill case: VP = 9; EI = 0.55; PS = 10.

Students’ game-play scores for all performance indexes are shown in Table 2.

Table 2.

Students’ performance indexes measured by means of TSAP.

				Time effect
						PreT–PostT comparison		PostT–ReT comparison
	PreT M (SD)	PostT M (SD)	ReT M (SD)	F p	η ²	MD 95% CI	ES	MD 95% CI	ES
VP	8.88 (3.30)	12.47 (3.60)	12.00 (4.43)	11.32 <.001	0.26	–3.58 (–5.60, –1.57)	–0.75	0.47 (–1.54, 2.48)	0.01
EI	0.50 (0.15)	0.99 (0.32)	0.83 (0.32)	34.56 <.001	0.51	–0.49 (–0.64, –0.35)	–1.40	0.16 (0.01, 0.31)	0.47
PS	9.71 (3.49)	16.42 (4.52)	14.30 (5.14)	28.52 <.001	0.46	–6.71 (–8.94, –4.48)	–1.27	2.12 (–0.11, 4.34)	0.07

PreT: pre-intervention assessment; PostT: post-intervention assessment; ReT: retention assessment; VP: volume of play; EI: efficiency index; PS: performance score; M: mean; SD: standard deviation; MD: mean difference; 95% CI: 95% confidence interval for mean difference; ES: Cohen’s d measure.

Overall, the data showed large improvements in all the participants’ scores from pre- to post-intervention, and these improvements remained at least until the end of the summer vacation (ReT). The EI showed a relevant regression from PostT to ReT (negative moderate effect).

Table 3 shows the results of the RM ANOVA for within-group effects.

Table 3.

Main and interaction effects of TGM unit on students’ performances.

	Lower (n = 18)			Higher (n = 16)			Time effect		Skill level × time interaction
	PreT M (SD)	PostT M (SD)	ReT M (SD)	PreT M (SD)	PostT M (SD)	ReT M (SD)	F p	η ²	F p	η ²
VP	6.27 (1.87)	11.11 (2.88)	11.11 (4.06)	11.81 (3.25)	14.00 (3.79)	13.00 (4.74)	11.28 <.001	0.14	2.77 0.07	0.03
EI	0.41 (0.08)	0.85 (0.24)	0.75 (0.22)	0.59 (0.15)	1.15 (0.32)	0.91 (0.39)	34.50 <.001	0.37	0.68 0.50	0.01
PS	7.25 (1.67)	14.50 (3.56)	13.10 (4.10)	12.47 (2.88)	18.57 (4.60)	15.65 (5.95)	28.10 <.001	0.29	1.08 0.33	0.01

PreT: pre-intervention assessment; PostT: post-intervention assessment; ReT: retention assessment; VP: volume of play; EI: efficiency index; PS: performance score; M: mean; SD: standard deviation.

Main effects resulted for time factor, whereas no relevant skill level × time interaction was found. Between-group analyses provided the following results: VP: F(1,32) = 19.20, p < .001, η ² = 0.17; EI: F(1,32) = 18.30, p < .001, η ² = 0.10; PS: F(1,32) = 20.13, p < .001, η ² = 0.14. Figure 1 shows the outcomes for higher- and lower-skill-level students across the assessments and provides the relevant comparisons between two pairs of assessments (i.e. PreT–PostT and PostT–ReT).

Figure 1.

Students’ performance indexes, grouped by skill level, across the three assessments. Within-group comparisons (mean differences and 95% CIs) were as follows: volume of play: PreT–PostT: –3.51 (–5.16, –1.86); PostT–ReT: 0.50 (–1.86, 2.86). Efficiency index: PreT–PostT: –0.49 (–0.63, –0.36); PostT–ReT: 0.17 (–0.004, 0.34). Performance score: PreT–PostT: –6.68 (–8.55, –4.80); PostT–ReT: 2.16 (–0.66, 4.98).

The students’ performances comparison between higher- and lower-skill-level students resulted in large to medium differences, respectively, in PreT and PostT assessments (i.e. for each index, the mean and the 95% CI of higher and lower groups did not overlap or just touched each other), whereas there were no clear differences in ReT. Post-hoc analysis revealed significant improvements for all the indexes between PreT and PostT, and these improvements seemed to remain at least until ReT. Specifically, Table 4 reports the within-group statistics for each skill level.

Table 4.

Students’ performance differences according to their skill level.

		Time effect
		PreT–PostT comparison		PostT–ReT comparison
		MD 95% CI	ES	MD 95% CI	ES
Lower (n = 18)	VP	–4.83 (–7.67, –1.99)	1.99	0.00 (–4.07, 4.07)	0.00
	EI	–0.44 (–0.67, –0.20)	2.45	0.10 (–0.19, 0.41)	0.43
	PS	–7.24 (–11.05, –3.45)	2.60	1.40 (–2.40, 5.20)	0.36
Higher (n = 16)	VP	–2.18 (–5.20, 0.82)	0.67	1.00 (–3.32, 5.32)	0.23
	EI	–0.55 (–0.80, –0.31)	2.24	0.23 (–0.08, 0.55)	0.67
	PS	–6.10 (–10.13, –2.07)	1.58	2.92 (–1.11, 6.96)	0.55

Concerning the lower-skill students, the measures revealed large improvements mostly caused by the intervention. A small regression resulted between PostT and ReT for EI and PS indexes (see mean difference and ES values in Table 4). Performances of higher-skill students seemed to mainly improve, with moderate to large amplitude, throughout the intervention period. After the intervention ended, its effect showed a loss of efficacy more than for the lower-skill students (i.e. the negative effects were small for VP and moderate for EI and PS, respectively).

Discussion and conclusion

This study aimed to assess the effect of a TGM-based unit on students’ volleyball performances in a longitudinal perspective. In addition, we aimed to assess the teaching strategy’s effect on students’ game-play performances by taking into account their original skill level.

The overall results showed a moderate to large positive effect of the TGM unit, even if it partially lost its effect after it ended. From a learning perspective, the indexes’ form of change throughout the unit was in agreement with the hypothesized outcomes assumed by applying an explanatory longitudinal design. In addition, these results are of interest because they can be read as good acceptance of the proposed instructional approach. However, as is well known, the effect of a teaching process tends to be lacking if it is not always supported by developmentally appropriate activities (Derri et al., 2008; Mesquita et al., 2012). In this regard, the presence of negative effects on students’ performances, mostly caused by the summer vacation, were verified (see PostT–ReT statistics in Table 4); therefore, the authors argue that it is necessary to organize summer school programmes to address these effects.

The RM ANOVA results revealed that between-group differences at the PreT tended to reduce in amplitude across the other assessments. The scores estimated for all outcomes in ReT for lower-skill-level students reached or exceeded the scores of higher-skill-level students at the PreT assessment. Within-group analysis (see Figure 1 and Table 3) revealed a large improvement in the overall scores for each index in the PreT–PostT comparison. The ES estimation and CI of each index (see Table 4) help to explain the extent to which the TGM plan had an impact over time. In the PreT–PostT comparison, the improvements of lower-skill-level students were larger than those of higher-skill-level students. Furthermore, these comparisons were supported by narrower CIs, which also do not include null value. Finally, the PostT–ReT comparison revealed similar performance behaviours for both groups, even if the decline of the indexes’ scores seemed to be much clearer for the higher-skill-level students. The findings are similar to what Araújo et al. (2016) identified for students involved in a volleyball teaching programme, based on fusing sport education and Step-Game Approach, even if the assessments discussed in the study by Araújo et al. (2016) were mainly related to tactical outcomes (i.e. decision-making, adjusting, skill efficacy). By considering similar outcomes in a mini-volleyball context, Mahedero et al. (2015) identified significant improvements for high- and low-skill-level students in decision-making, but only the middle-skilled students also improved on the skill execution index. More specifically, both of the previous studies addressed performances related to a different perspective (i.e. development of tactical awareness) within the psychomotor learning domain. However, according to the trend this previous evidence showed, current higher-skill-level students were probably limited in their improvements by a ceiling effect, related to using the same SSG format as the lower-skill-level students (i.e. four versus four) during the lessons and the assessments. In addition, the choice of using mixed-ability teams probably further limited the improvements of more skilled players throughout the unit. In this respect, when teachers identify proficiency differences within the class, they can adapt the lesson contents to lead students to work in their zone of proximal development (Vygotsky, 1978) – for example, SSGs for higher-skill-level students may be modified to improve the proposed learning tasks’ complexity by increasing the number of players (five versus five), the field’s dimensions, the length of a match, or by adapting the scoring rules. Furthermore, if adequate from a learning perspective, the composition of a team can be changed throughout the unit with the aim to support higher levels of motivation for each student with new challenges, but by choosing players belonging to the same ability group to preserve the competition’s fairness (MacPhail et al., 2008).

Overall, this study has shown that the TGM plan led students to improve their performance, and that the relative effect was also positive at the end of the summer vacation. According to the results of this study, we can suggest using this or a similar approach to reach in-game learning aims related to the psychomotor domain with elementary school students. Indeed, as shown in Table 2, the length of the CIs of each outcome in PreT-PostT comparison was similar to the one of the same outcome in Post-ReT comparison, and the level of uncertainty of means difference estimation was comparable. Regarding the practical significance of the CIs, the small sample size of each group negatively affected the level of precision of the mean difference intervals, but each group’s size was in accordance with a real educational setting. Therefore, the current results can be meaningful, as they are valid and ecological from a teaching-learning perspective. Finally, to the best of our knowledge, very few previous results related to this sport pedagogy topic have accounted for similar CIs and/or related ES analysis. Therefore, as an additional key outcome, we are confident that the current results may lead either to providing a trustworthy interpretation of the current design and analysis or to providing useful quantitative data for planning adequate future studies (i.e. preliminary sample size estimation and/or power analysis), as suggested by Abt et al. (2020).

Nevertheless, the current results are somewhat limited, because they do not account for the overall complexity underlying the teaching–learning process, such as the intertwined relationship between psychomotor and affective domains of learning. In this regard, future studies will assess the TGM plan’s effect also on data related to cognitive and affective domains of learning. In addition, we are also aware that our results do not take into account the possible confounder effect of the maturity status’s change throughout the study. Therefore, future studies will be planned and delivered by considering contrasting groups (i.e. the same teacher who teaches a TGM plan to one group and a different PE plan to the other group) to also take into account the aforementioned effect. Finally, using different sport-based games (e.g. invasion and net games) with similar or more longitudinal data collection can also be used to address a TGM instructional plan’s effect on students’ transfer of learning.

In conclusion, this study’s results strongly support the feasibility of using a GCA to engender improvements in technical in-game volleyball outcomes among elementary school students. At the same time, the results confirm the need for teachers to pay attention to the differences in students’ skill proficiency level for designing appropriate learning experiences. Furthermore, the detrimental effects of the summer vacation reflected in the current results support the need to design developmentally appropriate learning experiences during this period. In this respect, we believe it is appropriate to establish an educational agreement between school and extra-curricular stakeholders (i.e. local sport societies) because it can be the ground on which to develop sport-based educational programmes during this time. If realized, this strategy’s effects can be useful to support students’ participation in well-structured sport activities beyond the PE lessons and may be in alignment with the strategies supported by the World Health Organization for promoting active lifestyles for people of all ages (WHO, 2019: 7).

Footnotes

Acknowledgements

The authors would like to thank the Editor and the reviewers for their precious comments and suggestions, which have significantly helped us to improve the quality of this manuscript.

Declaration of conflicting interests

The author(s) declare no potential conflicts of interest to exist with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Abt

Boreham

Davison

, et al. (2020) Power, precision, and sample size estimation in sport and exercise science research. Journal of Sport Sciences 38(17): 1933–1935.

Araújo

Mesquita

Hastie

, et al. (2016) Students’ game performance improvements during a hybrid sport education–step-game-approach volleyball unit. European Physical Education Review 22(2): 185–200.

Arias-Estero

Castejón

(2014) Using instruments for tactical assessment in physical education and extra-curricular sports. European Physical Education Review 20(4): 525–535.

Arias-Estero

Jaquero

Martínez-López

, et al. (2020) Effects of two TGfU lessons period on game performance, knowledge and psychosocial variables in elementary physical education. International Journal of Environmental Research and Public Health 17(10): 3378.

Cohen

(2013) Statistical Power Analysis for the Behavioral Sciences. New York: Academic Press.

Cumming

(2014) The new statistics: Why and how. Psychological Science 25(1): 7–29.

Dania

Harvey

(2020) Teaching basketball to sampling-year athletes: A game-centered and situated learning perspective. Journal of Physical Education and Sport 20(2): 529–538.

Derri

Emmanouilidou

Kioumourtzoglou

, et al. (2008) Relationship between academic learning time-physical education and skill concepts acquisition and retention. Physical Educator 65(3): 134–145.

Dyson

Griffin

Hastie

(2004) Sport education, tactical games, and cooperative learning: Theoretical and pedagogical considerations. Quest 56(2): 226–240.

10.

Grehaigne

Godbout

Bouthier

(1997) Performance assessment in team sports. Journal of Teaching in Physical Education 16(4): 500–516.

11.

Harvey

Jarrett

(2014) A review of the game-centered approaches to teaching and coaching literature since 2006. Physical Education and Sport Pedagogy 19(3): 278–300.

12.

Hastie

Casey

(2014) Fidelity in models-based practice research in sport pedagogy: A guide for future investigations. Journal of Teaching in Physical Education 33(3): 422–431.

13.

Hastie

de Ojeda

Calderón

(2011) A review of research on Sport Education: 2004 to the present. Physical Education and Sport Pedagogy 16(2): 103–132.

14.

Hodges

Wicke

Flores-Marti

(2018) Tactical games model and its effects on student physical activity and gameplay performance in secondary physical education. Physical Educator 75(1): 99–115.

15.

Kirk

MacPhail

(2002) Teaching games for understanding and situated learning: Rethinking the Bunker-Thorpe model. Journal of Teaching in Physical Education 21(2): 177–192.

16.

MacPhail

Kirk

Griffin

(2008) Throwing and catching as relational skills in game play: Situated learning in a modified game unit. Journal of Teaching in Physical Education 27(1): 100–115.

17.

Mahedero

Calderón

Arias-Estero

, et al. (2015) Effects of student skill level on knowledge, decision making, skill execution and game performance in a mini-volleyball Sport Education season. Journal of Teaching in Physical Education 34(4): 626–641.

18.

Mesquita

Farias

Hastie

(2012) The impact of a hybrid sport education–invasion games competence model soccer unit on students’ decision making, skill execution and overall game performance. European Physical Education Review 18(2): 205–219.

19.

Metzler

(2005) Instructional Models for Physical Education. 2nd ed. Tempe, AZ: Holcomb Hathaway.

20.

Mitchell

Oslin

Griffin

(2020) Teaching Sport Concepts and Skills: A Tactical Games Approach for Ages 7 to 18. Champaign, IL: Human Kinetics.

21.

Nathan

(2018) Nonlinear pedagogy game instruction. In: Serra-Olivares

(ed) Sport Pedagogy: Recent Approach to Technical-Tactical Alphabetization. London: IntechOpen , pp. 63–78.

22.

Ployhart

Vandenberg

(2010) Longitudinal research: The theory, design, and analysis of change. Journal of Management 36(1): 94–120.

23.

Siedentop

(1998) What is sport education and how does it work? Journal of Physical Education, Recreation & Dance 69(4): 18–20.

24.

Silverman

Woods

Subramaniam

(1999) Feedback and practice in physical education: Interrelationships with task structures and student skill level. Journal of Human Movement Studies 36: 203–224.

25.

Seifert

Davids

(2017) Ecological dynamics: A theoretical framework for understanding sport performance, physical education and physical activity. In: Bourgine

Collet

Parrend

(eds) First Complex Systems Digital Campus World E-Conference 2015. Cham: Springer, 29–40.

26.

Sgrò

Pignato

Lipoma

(2018) Assessing the impact of gender and sport practice on students’ performance required in team games. Journal of Physical Education and Sport 18(1): 497–502.

27.

Van Acker

Carreiro da Costa

De Bourdeaudhuij

, et al. (2010) Sex equity and physical activity levels in coeducational physical education: Exploring the potential of modified game forms. Physical Education and Sport Pedagogy 15(2): 159–173.

28.

Vygotsky

(1978) Mind in Society. Cambridge: Harvard University Press.

29.

World Health Organization (WHO) (2018) Promoting physical activity in the education sector. Current status and success stories from the European Union Member Stated of the WHO European Region. Available at: https://www.euro.who.int/en/health-topics/disease-prevention/physical-activity/publications/2018/promoting-physical-activity-in-the-education-sector-2018 (accessed 10 January 2021).

30.

World Health Organization (WHO) (2019) Global action plan on physical activity 2018-2030: More active people for a healthier world. Available at: https://apps.who.int/iris/bitstream/handle/10665/272722/9789241514187-eng.pdf (accessed 28 January 2021).