Emotional Intelligence and Self-Esteem in U10–U12 Football Players: A Montessori-Informed Approach Beyond Motor Skill Development

1. Introduction

Sport performance in children and adolescents is a multidimensional construct influenced by motor, cognitive, emotional, and social factors. In youth soccer, physical abilities such as speed, agility, and coordination are essential for performance development. However, recent research emphasizes that psychological variables also play an important role in athletes’ adaptation to training and competition (Côté et al., 2007; Lloyd et al., 2015). Among these psychological dimensions, emotional intelligence and self-esteem have been associated with emotional regulation, stress management, motivation, and behavioural adaptation in sport contexts. Athletes with higher emotional intelligence demonstrate better coping strategies and greater emotional stability during competitive situations, while self-esteem influences confidence, persistence, and engagement in training (Laborde et al., 2016; Orth & Robins, 2014). Moreover, emotional intelligence can be improved through targeted interventions and educational experiences, supporting athletes’ adaptive capacities and psychological functioning (Kotsou et al., 2019).

Furthermore, regular physical activity has been positively associated with life satisfaction and psychosocial well-being among children and adolescents (Ubago-Jiménez et al., 2023). Despite increasing interest in psycho-emotional factors in sport, limited research has examined the relationship between emotional intelligence, self-esteem, and motor performance in preadolescent soccer players. Furthermore, the role of educational environments that promote autonomy, active learning, responsibility, cooperation, and emotional self-regulation (Ericsson et al., 2018), such as Montessori-inspired systems remains comparatively underexplored in youth sport (Gentaz & Richard, 2022; Demangeon et al., 2023). The empirical evidence supporting Montessori education generally remains relatively limited, with the existing body of research being constrained by several methodological limitations, including small sample sizes, inadequate control conditions, and variability in the fidelity with which Montessori principles and practices are implemented across studies (Gentaz & Richard, 2022).

In physical activity, the findings suggest that the educational setting and instructional practices may play an important role in shaping children’s physical activity behaviour, with children attending Montessori centres reported to be more physically active than their counterparts attending traditional childcare centres (Pate et al, 2022). These findings highlight the potential relevance of child-centred educational environments for promoting physical activity, while also indicating the need for further research to determine whether Montessori-based principles can be systematically translated into structured physical activity and sport interventions.

Other evidence highlights the importance of providing children with structured, developmentally appropriate opportunities to practise fundamental movement skills, supported by well-informed educators and an environment that facilitates meaningful engagement in physical activity. Dobell et al. (2023) emphasize that effective interventions should address not only the child’s individual capabilities but also the interpersonal and organisational contexts in which movement learning occurs. This perspective supports the development of pedagogical approaches that integrate structured practice with child-centred learning environments and may provide a relevant foundation for extending fundamental movement skill development into more specialised physical activity and sport contexts. Taken together, these findings provide a rationale for investigating whether the principles underlying Montessori education, particularly its emphasis on child-centred learning and the learning environment, can be adapted to structured sport settings and may be associated with developmental outcomes in young athletes. Within the context of youth soccer training, these principles may contribute not only to psychological development but also to improved adaptation to sport-specific motor demands.

The Role of Emotional Intelligence and Self-Esteem in Youth Sports

Such educational approaches are consistent with contemporary perspectives on youth sport that emphasize the interpersonal and developmental dimensions of coaching. Emotional intelligence and transformational leadership have been associated with the quality of coach–athlete relationships and with positive developmental experiences in youth sport. Research involving adolescent soccer players has shown that transformational leadership behaviours and the quality of the coach–athlete relationship are positively associated with developmental experiences, highlighting the importance of individualized consideration, intellectual stimulation, and positive role modelling in youth coaching (Vella et al., 2013).

More recent evidence similarly indicates that transformational leadership behaviours are related to the quality of coach–athlete relationships, particularly through individualized consideration, role modelling, and the promotion of group goals (López de Subijana et al., 2021; Zhao & Jowett, 2023). Emotional intelligence may further contribute to the quality of these relationships, with evidence suggesting that emotional intelligence and relationship quality are interconnected with athletes’ and coaches’ satisfaction (Zhao & Jowett, 2023). Within this developmental framework, self-esteem should not be conceptualised solely as a consequence of sporting success. Rather, it represents an important component of young athletes’ self-perceptions and may interact with perceived competence, motivation, and engagement in sport (Ryan & Deci, 2000). Longitudinal research in children has demonstrated associations among global self-esteem, perceived athletic competence, and physical activity (individual or team practice), although these relationships may vary according to developmental and individual characteristics (Noordstar et al., 2016; Šagát et al., 2021).

In youth soccer, therefore, psychological development should be considered alongside the acquisition of sport-specific skills. Organized sport provides opportunities not only for the development of physical and technical competencies but also for the development of confidence, social connection, responsibility, and adaptive behaviours. Evidence from youth sport research indicates that positive coaching environments and supportive coach–athlete relationships can contribute to these broader developmental outcomes.

Accordingly, contemporary youth soccer training may be conceptualized as a multidimensional educational process in which motor learning, psychological development, interpersonal relationships, and opportunities for autonomy and competence are closely interconnected. Yu et al. (2024) demonstrated that self-esteem and collective self-esteem may represent important psychological mechanisms linking football participation with sport trait confidence among adolescents. Zhao et al. (2025) found that athlete engagement and collective self-esteem jointly mediated the relationship between soccer participation and team cohesion in adolescent soccer players.

Research on youth soccer has also emphasized the importance of autonomy-supportive coaching and increasingly emphasized athlete-centred leadership and the alignment between coaches’ behaviours and athletes’ perceptions. Sevil-Serrano et al. (2021) showed that athletes’ perceptions of coaching behaviours are closely related to the satisfaction and frustration of basic psychological needs and to sport commitment. The study by Buda et al. (2026) highlights the relevance of individualized consideration, emotional support, interpersonal adaptability, and athlete-centred communication within effective coaching environments.

Therefore, the present study examines the relationship between emotional intelligence, self-esteem, and motor performance in soccer players aged 10–12 years and compares psychological and motor performance profiles between athletes participating in Montessori-informed and traditional training environments.

2. Materials and Methods

2.1. Aim of the Study

The present study aimed to examine the relationships between emotional intelligence, self-esteem, and motor performance among U10–U12 soccer players while investigating whether participation in a Montessori-informed educational environment contributes to differences in psycho-emotional functioning and sport-related motor development. Specifically, the study sought to compare psychological and motor performance profiles between athletes trained within a traditional football setting and those enrolled in a Montessori-based educational program, as well as to explore the extent to which emotional intelligence and self-esteem may contribute to variability in youth athletic performance.

To achieve this aim, the study focused on assessing emotional intelligence and self-esteem levels among young soccer players, analysing motor performance through soccer-specific standardized tests, and identifying potential differences between athletes trained within a traditional coaching system and those integrated into a Montessori-based training environment.

The present research was designed within a multidimensional framework of youth sports development, according to which athletic performance at early ages is not determined exclusively by motor abilities but emerges from the complex interaction between physical, emotional, and psychosocial characteristics of young athletes, as proposed by long-term athlete development models (Lloyd et al., 2015; Fernandes et al., 2024).

2.2. Research Objectives

To address the study aim, the research was structured around the following objectives:

• To characterize the motor performance through standardized soccer-specific assessments, including agility, dribbling efficiency, and sprint speed, using the T-test, Zig-Zag dribbling test, and 10-meter sprint test (Montenegro Bonilla et al., 2024).
• To identify the levels of emotional intelligence among U10–U12 soccer players using the Bar-On Emotional Quotient Inventory – Youth Version (EQ-i:YV) (Bar-On, 1997), with particular attention to adaptability, stress management, interpersonal functioning, and general well-being.
• To assess the self-esteem profiles among young athletes using the Coopersmith Self-Esteem Inventory – Students (CSEI-S) (Coopersmith, 1967), examining both global and domain-specific dimensions of self-perception relevant to youth sport participation
• To compare psycho-emotional and motor performance profiles between athletes trained within a traditional football setting and those participating in a Montessori-informed educational environment, in order to identify potential developmental differences.

2.3. Research Hypotheses

The research was based on the following hypotheses:

• Athletes exposed to a Montessori-inspired, autonomy-supportive training environment will demonstrate greater improvements in agility and adaptation to game-specific motor tasks than athletes trained in a traditional coach-centered environment.

The rationale for H1 is that guided choice, variable task constraints, and op-portunities for self-correction require players to participate actively in solving mo-vement problems. For example, spatial restrictions and changes in possession create repeated opportunities to adjust movement direction, pace, and ball control. Such activities connect practice with the coordination demands represented by the T-test and Zig-Zag dribbling assessment, while repeated accelerations also relate to 10-m sprint performance. Within the skill-acquisition framework discussed above, the coach prepares appropriately challenging tasks and offers concise guidance, allo-wing the player to explore effective solutions rather than relying continuously on an externally prescribed response. The expected advantage therefore concerns the qu-ality of engagement with motor practice and the opportunity to adapt learned skills to changing task conditions. The three motor tests provide complementary indica-tors through which this expectation can be examined.

• Athletes participating in a Montessori training environment will demonstrate higher scores on selected dimensions of emotional intelligence than athletes participating in a traditional training environment.

H2 follows from the emotional and interpersonal demands embedded in the intervention. Choosing among possible actions, coping with errors, and responding to changing task conditions provide opportunities to practise adaptability and stress management. Cooperative activities also require players to communicate, recognize teammates’ needs, and coordinate their responses, making interpersonal function-ing particularly relevant. When the coach treats errors as information for learning and combines clear rules with meaningful choice, players may experience a greater sense of competence, inclusion, and enjoyment. These features offer a conceptual explanation for expecting favourable General Mood scores, consistent with the autonomy, competence, and relatedness framework introduced earlier. The hypothesis thus concerns a differentiated emotional-intelligence profile: the strongest expectations relate to capacities repeatedly engaged in training, rather than assuming an identical response across every psychological dimension.

• Athletes trained according to Montessori pedagogical principles will report higher levels of self-esteem and more favorable self-perception profiles than athletes trained in a traditional environment.

The rationale for H3 centers on how children interpret their participation and progress. Opportunities to make decisions, recognize personal improvement, and contribute to shared activities may help players experience themselves as capable and valued participants. Individualized feedback can direct attention towards achievable progress and constructive responses to mistakes, providing a basis for more favorable self-perception. Cooperative tasks also give children opportunities to experience recognition and belonging within the group, making the social domain of self-esteem relevant to the training context. This reasoning is consistent with the relationships among perceived competence, self-esteem, and sport participation discussed in the Introduction. The hypothesis concerns substantive self-esteem dimensions; the inventory’s social-desirability scale is considered separately as an indicator of response tendencies, rather than as evidence of improved self-esteem.

2.4. Participants

A non-probability convenience sampling strategy was used. Participants were recruited from two pre-existing training environments: School Sports Club No. 1 Bucharest, a conventional football club, and the Keep Going – Montessori School educational programme. All athletes enrolled at the two participating institutions were invited to participate in the study. Their parents were informed at meetings held at each institution, during which the study procedures were explained and written informed consent forms were distributed. Recruitment took place between 1 and 15 March 2024.

A total of 51 children expressed interest in participating in the study. The inclusion criteria were: (1) being aged 10–12 years at the time of the study; (2) having 2–4 years of formal soccer training experience; (3) having participated in local-level soccer competitions; and (4) being enrolled in one of the two designated training environments. Of these, 11 were not included because they did not confirm their availability to attend all scheduled training sessions. The final study sample comprised 40 soccer players, with 20 participants from each training environment. Participation was voluntary, and written informed consent was obtained from the parents or legal guardians of all participants in accordance with ethical principles governing research involving minors.

The traditional training group comprised athletes enrolled at School Sports Club No. 1 Bucharest. This group followed the club’s conventional football training model, characterised by coach-directed instruction, structured drills, direct corrective feedback, and comparatively greater coach control over task execution and decision-making. The Montessori-based group comprised athletes participating in the Keep Going – Montessori School educational programme. These athletes trained with the same coach as the traditional group but followed a Montessori-informed approach (Table 1).

Group membership therefore reflected participants’ pre-existing training environments rather than random allocation by the researchers. Accordingly, the study employed a quasi-experimental rather than a randomised design.

2.5. Research Methods

From a methodological perspective, the study employed a pedagogical experiment and systematic pedagogical observation. The primary between-group factor was training environment (traditional vs. Montessori-informed), while the outcome variables comprised emotional intelligence, self-esteem, and motor performance. This multidimensional research design allowed the investigation of the interaction between psycho-emotional factors and athletic performance within the context of youth football training.

Motor performance was assessed using standardized tests commonly applied in youth soccer evaluation, namely the T-Test for agility, the Zig-Zag dribbling test (Figure 1) and the 10-meter sprint test. These assessments represent relevant indicators of functional performance in modern soccer, where reaction speed and rapid directional changes constitute key determinants of competitive success (Montenegro Bonilla et al., 2024).

These instruments were selected due to their ecological validity and their relevance for the functional demands of modern football, where rapid accelerations, multidirectional movements, agility, and technical execution under time pressure represent essential components of competitive performance. Existing literature (Montenegro Bonilla et al., 2024) emphasizes that speed and change-of-direction ability constitute reliable indicators of motor efficiency and game adaptability in youth football players.

Psychological dimensions were evaluated using internationally validated instruments. Emotional intelligence was measured using the Bar-On Emotional Quotient Inventory – Youth Version (EQ-i:YV) (BarOn & Parker, 2022) is a 60-item self-report for children and adolescents aged 6–18 years. The measure takes approximately 30 minutes to complete and is grounded in the Bar-On model of emotional and social intelligence, which encompasses competencies related to self-awareness, interpersonal relationships, adaptability, and emotion regulation. It yields scores on five scales—Intrapersonal, Interpersonal, Stress Management, Adaptability, and General Mood. The inventory also includes a Positive Impression scale and an Inconsistency Index to assess response validity. The four core scales—Intrapersonal, Interpersonal, Stress Management, and Adaptability—contribute to the Total EQ score, while General Mood provides additional information on emotional functioning. Each of the 60 items is rated on a four-point Likert-type scale ranging from 1 (Very seldom true of me) to 4 (Very often true of me). The responses are used to calculate scores for the instrument's emotional and social intelligence dimensions, according to the standardized scoring procedures. The Romanian version used in this study has been validated for the Romanian population and was approved by the Romanian College of Psychologists, in 2023, for use in clinical psychology, educational psychology, school counseling, and vocational counseling.

Self-esteem was evaluated through the Coopersmith Self-Esteem Inventory – for Students (CSEI-S) (Coopersmith, 2017), in its Romanian standardized version for children and adolescents aged 9–18 years. It measures evaluative self-attitudes across various domains of an individual's life, including social, academic, family, and personal contexts. Within the CSEI framework, self-esteem is defined as a subjective judgment of personal worth, expressed through an individual's attitudes toward themselves. Self-perception is considered an indicator of perceived competence, success, significance, and personal worth. The CSEI-S comprises 58 items scored using a dichotomous response format (Yes/No), with an estimated completion time of approximately 25–30 minutes. Fifty items assess self-esteem, while the remaining eight constitute the Distortion Scale, which evaluates the defensive response patterns that may influence a child's answers to the inventory. The self-esteem items yield a total score and can also be used to calculate four separate subscale scores: General Self-Esteem, Peer-Related Self-Esteem, Home-Related Self-Esteem (reflecting relationships with parents), and School-Related Self-Esteem (reflecting academic performance). The interpretation of its scores followed the standardized scoring procedures specified for the instrument.

The intervention was delivered in Romanian, which was the participants' native language and the language used for all instructional activities and communication throughout the study. Evidence supporting the validity of the Romanian version of both instruments was reported in the specific references authorized by the Romanian College of Psychologist (Bar-On & Parker, 2022; Coopersmith, 2017). In the present study, internal consistency reliability was assessed using Cronbach’s alpha; no additional instrument-validation analyses were conducted.

The combined use of these instruments enabled an integrated analysis of psychological mechanisms underlying youth sports participation, consistent with recent findings emphasizing the role of emotional intelligence in behavioral self-regulation and sustained sport motivation (Huang et al., 2025). Both instruments have demonstrated evidence of validity and reliability in previous research, and their use in the present study provided standardized measures of the assessed psycho-emotional dimensions. Their inclusion in the present research ensured an objective evaluation of psycho-emotional variables relevant to sport participation and athlete development.

Internal consistency

Internal consistency was assessed using item-level responses for both the Bar-On Emotional Quotient Inventory – Youth Version (EQ-i:YV) and the Coopersmith Self-Esteem Inventory – Students (CSEI). Cronbach’s alpha coefficients were calculated for the total sample and separately for each training group. Reverse-keyed items were recoded before analysis in accordance with the respective scoring instructions.

For the EQ-i:YV, internal consistency was assessed separately for the four core dimensions: Intrapersonal, Interpersonal, Stress Management, and Adaptability. Items from the General Mood and Positive Impression scales were excluded from the analysis of the combined 40-item core scale. In the total sample, Cronbach’s alpha was .762 for Intrapersonal, .731 for Interpersonal, .610 for Stress Management, and .875 for Adaptability. The coefficient for the combined core scale was .862. For this combined scale, group-specific coefficients were .895 in the traditional football group and .731 in the Montessori-informed Keep Going group.

For the CSEI items, excluding the eight Lie Scale items. The coefficient for the total self-esteem scale was .842 in the total sample, .722 in the traditional football group, and .829 in the Montessori-informed Keep Going group.

Collected data were analyzed using both descriptive and inferential statistical methods. Statistical analyses were conducted using IBM SPSS Statistics (Version 29/30). Distribution normality was assessed using the Shapiro–Wilk test, while homogeneity of variances was verified through Levene’s test. Within-group changes from baseline (T1) to follow-up (T2) were assessed with the Wilcoxon signed-rank test, and between-group differences at T2 with the Mann–Whitney U test. Where appropriate, T2-to-T1 change scores were calculated and compared between groups to assess differences in the magnitude of change. Effect sizes were reported as rank-biserial correlations. Exploratory associations between psychological and motor-performance variables were examined using regression analyses. The significance threshold was set at α = 0.05.

2.6. Research Methodology

The research protocol was implemented over a period of nine months, during the 2024–2025 competitive season. Two assessment sessions were organized: an initial evaluation (T1, September 2024) at the beginning of the championship period and a final evaluation (T2, May 2025) at the end of the season. Training sessions were conducted in Bucharest, three times per week, complemented by weekly official matches organized within local and regional youth football competitions. This longitudinal structure enabled the assessment of changes in motor and psycho-emotional variables over one competitive season.

The intervention program was developed according to the principles of Montessori pedagogy and adapted to the specific demands of football training. (Table 1) The originality of the intervention consisted in integrating football-specific motor objectives with socio-affective and self-regulatory learning components. The training environment was intentionally designed to encourage autonomy, active participation, cooperation, decision-making, and self-regulation during game-related situations.

A central principle of the intervention involved the use of constraint-based learning situations, in which the environment itself generated corrective feedback through task conditions, limited time, spatial restrictions, or loss of ball possession. This approach was intended to encourage players to identify solutions independently and adapt their behaviour to contextual demands. Within this framework, autonomy was operationalized through the players’ capacity to make effective tactical and technical decisions without continuous external intervention from the coach. Furthermore, the coach adopted a facilitative role, providing guidance while minimizing direct control over players’ decision-making. Training activities were organized around game-based tasks and small-sided situations that promoted exploration, responsibility, and experiential learning.

These learning conditions are consistent with contemporary theories of expertise development, emphasizing deliberate practice, experiential learning, and the progressive acquisition of performance-related skills (Ericsson et al., 2018).

Consequently, the Montessori-informed approach aimed not only to improve motor performance but also to stimulate emotional adaptability, communication skills, intrinsic motivation, and psychological resilience, dimensions considered essential for long-term athlete development in contemporary youth football.

3. Results

The analysis of the results was conducted both separately for each assessment instrument and through the examination of the relationships between motor and psycho-emotional variables. This approach allowed a multidimensional interpretation of the effects generated by the Montessori-informed intervention program on the athletic and psychological development of young football players. The term “Montessori-informed” is used to indicate that selected principles of Montessori pedagogy were adapted to football training rather than that the intervention constituted a full Montessori educational program.

3.1. Analysis of Motor Performance

Regarding motor performance, the initial assessment (T1) did not reveal statistically significant differences between the two groups in any of the selected motor tests, suggesting no statistically significant baseline differences between the groups on the assessed motor-performance measures. This aspect is methodologically important, as it supports the assumption that subsequent differences identified at the final assessment may be associated with the characteristics of the implemented intervention rather than with pre-existing disparities between participants.

Table 2 presents the results of the intragroup analysis of motor performance, highlighting significant improvements in agility, dribbling ability, and sprint speed between the initial and final assessments. Within-group changes were examined using Wilcoxon signed-rank tests. In the traditional group, statistically significant differences were observed between the initial and final assessments for T_L1–T_L2, W = 78, p = .002, the Zig-Zag test, W = 3, p = .009, and the Sprint test, W = 42, p = .024. However, the corresponding rank-biserial correlations indicated very small effects (r_rb = .10, .12, and .06, respectively).

In the Montessori group, statistically significant improvements were observed across all three measures. The T_L1–T_L2 comparison yielded W = 105, p = .001, with a moderate effect (r_rb = .35). The Zig-Zag assessment showed a highly significant difference, W = 171, p < .001, accompanied by a very large effect (r_rb = .85). Similarly, the Sprint assessment demonstrated a highly significant difference, W = 210, p < .001, with a large effect (r_rb = .55). Overall, both groups demonstrated statistically significant changes from the initial to the final assessment; however, the magnitude of the observed changes was substantially greater in the Montessori group. The strongest within-group effect was observed for the Zig-Zag performance (r_rb = .85), followed by Sprint performance (r_rb = .55). In contrast, the traditional group demonstrated only very small effects across all assessed motor-performance measures. The analysis demonstrated significant improvements in both groups across the evaluated motor parameters.

At the second evaluation session (T2/follow-up assessment), intergroup differences became more pronounced. Table 3 summarizes the intergroup significant differences (p<0.05) in motor performance observed at the final assessment, highlighting the superior performance achieved by athletes from the Montessori-based program.

The Mann–Whitney U test revealed statistically significant differences between the two groups across all three motor performance measures assessed at T2. For the T-test, a statistically significant between-group difference was observed, U = 98.5, p = .006, with a mean difference of 0.3 points (95% CI [0.1, 0.5]). The rank-biserial correlation indicated a medium effect size (r_rb = .50).

A particularly pronounced between-group difference was observed for the Zig-Zag test, U = 2.0, p < .001. The mean difference was 1.4 points (95% CI [1.2, 1.6]), with a rank-biserial correlation of r_rb = .99, indicating an extremely large effect and a near-complete separation between the two groups in terms of ranked performance.

Similarly, the Sprint test showed a statistically significant difference between groups, U = 42.5, p < .001. The mean difference was 0.6 points (95% CI [0.6, 0.8]), and the rank-biserial correlation indicated a large effect (r_rb = .78).

Overall, the findings indicate statistically significant and substantial between-group differences in all three motor performance measures at T2, with the medium to large effect observed for the T-test, the Sprint and the Zig-Zag test measures.

3.2. Analysis of the Psycho-Emotional Variables

The analysis of psycho-emotional variables assessed using the Bar-On Emotional Quotient Inventory revealed consistent differences favouring the Montessori group across most assessed dimensions. Figure 2 illustrates the mean values obtained by the two groups across the emotional intelligence dimensions assessed by the Bar-On Emotional Quotient Inventory at the final evaluation (T2).

As shown in Figure 2, athletes from the Montessori group obtained higher scores in Adaptability, Stress Management, Interpersonal Skills, and Intrapersonal functioning compared to the traditional group, with the largest difference being observed for the Interpersonal dimension.

The Mann–Whitney U test indicated statistically significant differences between the two groups for four of the five assessed dimensions. A significant difference was observed for Adaptability, U = 111, p = .02, with a mean difference of −7 and a moderate effect size (r_rb = .45). Similarly, Stress differed significantly between groups, U = 125, p = .04, with a mean difference of −3 and a moderate effect size (r_rb = .38). A statistically significant difference was also identified for the Interpersonal dimension, U = 120, p = .03, with a mean difference of −5 and a moderate effect size (r_rb = .40). For General Mood, the analysis revealed a significant between-group difference, U = 114, p = .02, with a mean difference of −18 and a moderate effect size (r_rb = .43). In contrast, no statistically significant difference was found for the Intrapersonal dimension, U = 166, p = .35. The mean difference was small (−1), and the reported effect size was negligible (r_rb = 0). (Table 4)

From a self-esteem perspective, higher self-perception scores and more stable self-evaluative patterns, along with lower social desirability scores observed in Group 2, suggest enhanced self-awareness and a favourable self-perception profile and lower socially desirable response tendencies. (Table 5) Mann–Whitney U tests revealed statistically significant between-group differences in three of the four assessed dimensions. A significant difference was observed for Self-perception, U = 89.5, p = .003, with a mean difference of −2 points (95% CI [−3, −1]) and a moderate-to-large rank-biserial correlation (r_rb = .55). Social-self also differed significantly between groups, U = 127, p = .041, with a mean difference of 1 point (95% CI [1, 5.57]) and a moderate effect (r_rb = .37). No statistically significant difference was found for the educational contexts dimension (School), U = 153.5, p = .184, with a mean difference of 3.67 points (95% CI [−5.94, 1]) and a small effect (r_rb = .23). Finally, Social desirability showed a highly significant between-group difference, U = 78.5, p < .001, with a mean difference of 1 point (95% CI [1, 1]) and a large effect (r_rb = .61). Overall, the findings indicate that the groups differed significantly in self-perception, social-self, and social desirability, whereas no significant difference was observed for the educational dimension. (Table 5)

An exploratory linear regression analysis was conducted in the Montessori group (n = 20) to examine the association between T2 motor performance and General Mood. The final model included Zig-Zag performance as the predictor and was statistically significant, F(1,18) = 5.64, p = .029, R² = .24. Thus, Zig-Zag performance accounted for approximately 24% of the variance in General Mood in the analysed sample. Given the exploratory nature and limited sample size of this analysis, the finding should be interpreted as an association rather than evidence of a predictive or causal effect, suggesting that motor performance may represent a meaningful correlate and potential predictor of General Mood in young football players, although these findings should be interpreted cautiously and verified in larger samples. (Table 6).