“Character Is the Most Important Thing in Soccer”—A Pilot Study on the Development and Effectiveness of a Character Building Program in High-Performance Youth Soccer ()
1. Introduction
The connection between strength of character and sport-related performance outcomes is well established. Athletes who have developed a strong performance and moral character based on well-formed habits can build on this foundation to strive for true excellence (Coyle, 2009; Ericsson, Krampe, & Tesch-Römer, 1993). In order to effect character development, a structured sporting environment with a clearly defined goal of developing the character of young players is required, as well as appropriate teaching and coaching methods (Doty, 2006; Rudd & Stoll, 2004).
In recent years, character coaching has become more established in sports (Naylor & Yeager, 2013; Collins & MacNamara, 2012). Youth sports, however, still pay little attention to explicit character development as athletic youth programs often fail their responsibility to be engaged in the personal development of their athletes. Moreover, the influence of athlete and coach character on performance and talent development is often not considered when designing youth programs (Yeager, Fisher, & Shearon, 2011).
An exception from sport practice is the Danish soccer club FC Nordsjælland that takes a pioneering role in implicit and explicit character development in youth soccer. The club established concepts guided by social responsibility and implemented a character building school for their youth teams (FCN, 2022). The club further partnered with the Right-to-Dream-Academy in Ghana, which educates and trains their players based on social responsibility and value orientation.
Although there are good developments in character training in high performance sports, there has been little scientific investigation of these programs (Pulido, Leo, González-Ponce, López-Gajardo, & Sánchez-Oliva, 2021). Existing interventions also focus primarily on coaches without considering athlete’s perspectives and target outcomes besides character development (Lupori, 2015; Harwood, Barker, & Anderson, 2015). Moreover, although there are already studies comparing high-performance soccer players with amateurs in terms of several parameters (Verburgh, Scherder, Van Lange, & Oosterlaan, 2016; Vaeyens et al., 2006), there is still a lack of evidence here on variables related to character strength. This paper therefore presents one of the first initiatives of a theory-based character bildung program in youth soccer and tests its application into sport practice. In addition, we compare amateur and high-performance youth players in terms of their strength of character and implicit theories of intelligence and talent. Our research aims to conceptualize an intervention that targets explicit character development and cultivates a growth mindset to increase the effectiveness of talent development programs. In doing so, we aim to help making character strength a fundamental component of strategies that optimize athletes’ performance.
The aim of this pilot study was thus to describe the theoretical background, development and implementation of a new character building program for youth soccer players. The study further used a mixed-methods approach to present initial results of the program. We defined three research questions:
1) Is there a difference in implicit theories of intelligence and talent as well as the strength of character between children in an amateur soccer club and a high-performance youth academy?
2) Does a character building intervention in a soccer team at a high-performance youth academy improves test scores in implicit theories of intelligence and talent and strength of character?
3) What contents of the character building program are remembered by the players after the intervention?
Accordingly, we will first describe the implementation and contents of the character building program as well as its development based on a theoretical foundation.
The character building program
The character building program (called “Charakterschule”) has been developed at the youth academy of the professional German soccer club SpVgg Greuther Fürth (SGF) for players at age levels Under 12 (U12) to Under 15 (U15) since this age group corresponds to pubescence where personality development and high neuroplasticity take place (Yeager, Trzesniewski, & Dweck, 2013). The goal of the program was to strengthen the youth athlete’s character building and personal development in individuality, personality, and social competence. Guided by the transformational leadership approach, the program aimed to create a supportive, mastery-oriented environment for youth players and serve as an opportunity to actively teach and discuss basic values with SGF’s youth academy players. The framework schedule consisted of twelve weekly sessions with different lecturers selected for their individual fields of expertise. Table 1 presents an overview of the framework schedule and each lesson’s.
Theoretical background
The design of the character building program was based on a theoretical framework that includes and links the “Achievement Goal Theory” (AGT), “Motivational Climate” (MC), and “Transformational Leadership” (TL) (see Figure 1).
The AGT with two mindsets (fix and incremental) at its core (Dweck & Leggett, 1988) forms the framework concept of the program. AGT states that children with a so-called “growth mindset” hold an incremental theory of intelligence and talent, believing that intelligence is a malleable trait that can be increased and controlled. In contrast to this is the “fixed mindset”, where intelligence or other basic talents are understood as fixed and uncontrollable (Dweck, 2017).
![]()
Figure 1. Relationship between AGT, MC and TL as theoretical Framework of the character building program.
![]()
Table 1. Framework schedule of the character building program.
Previous research has shown that children who believe that intelligence is increasable pursue the learning goal of developing their competence, whereas children who believe that intelligence is a fixed entity are more likely to pursue the performance goal of securing positive judgments about it or preventing negative judgments about it (Bandura & Dweck, 1985). In sport context, athletes with a growth mindset and therefore high mastery orientation report a higher sense of competence, more enjoyment of the activity and higher intrinsic motivation and effort compared to ego-orientated athletes with a fixed mindset (Duda, 2005; Kavussanu & Roberts, 1996).
The mindset of young athletes can be influenced by their coaches (e.g., by their definition of failure) (Cimpian, Arce, Markman, & Dweck, 2007; Gunderson et al., 2013; Haimovitz & Dweck, 2016; Kamins & Dweck, 1999; Mueller & Dweck, 1998). In the context of the character building program, the method of TL was used to transfer the AGT to the MC. The TL-paradigm (Longshore & Bass, 2006) is increasingly considered in the sport context (Álvarez, Castillo, Molina-García, & Balague, 2016; Hoption, Phelan, & Barling, 2007; Turnnidge & Côté, 2018), but empirical research is still relatively new (Arthur, Bastardoz, & Eklund, 2017). Several previous studies, however, have already shown that transformational coach behaviours are positively related to athlete performance outcomes (Álvarez, Falco, Estevan, Molina-García, & Castillo, 2013; Morgan, Fletcher, & Sarkar, 2015), team cohesion (Cronin, Arthur, Hardy, & Callow, 2015; Price & Weiss, 2013), satisfaction (Kao & Tsai, 2016), less aggressive behaviour (Tucker, Turner, Barling, & McEvoy, 2010), and athlete well-being (Stenling & Tafvelin, 2014).
The TL forms the tool to convey the contents of the program in a MC, which itself is an important situational construct in AGT, and can significantly influence athletes’ behaviour and performance strategies (Ames, 1995). In the sport context, MC refers to individual athletes’ perceptions of how their coaches define success or failure and how they evaluate competence in training and play. Consistent with theoretical expectations, a climate of mastery is associated with stronger incremental beliefs related to intelligence and talent in athletes, and a climate of ego is associated with stronger fixed beliefs (Duda, 2005; Roberts, Treasure, & Conroy, 2007). Previous studies have shown that the cognitive-behavioural approach to MC interventions with coaches led to improved self-esteem in children and young athletes (Barnett, Smoll, & Smith, 1992; Smith & Smoll, 1978; Smith, Smoll, & Ptacek, 1990), increased their enjoyment of sport (Cumming, Smoll, Smith, & Grossbard, 2007) and the development of a championship-oriented growth mindset (Smith, Smoll, & Cumming, 2009; Smoll, Smith, & Cumming, 2007). According to Duda and Balaguer (2007) research on MC in a sport context suggests that creating a task climate has positive effects on athletes, while creating an ego climate leads to negative sport experiences (Duda & Balaguer, 2007). Thus, players with a growth mindset pursue the goal of increasing their competencies and their behavioural patterns evolve towards seeking challenges rather than avoiding them and are characterised by high persistence (Diener & Dweck, 1980).
Strength of character
The SGF youth section promotes the core values and mental strength of their young players. The program was intended to be a character building measure in which even more emphasis was placed on the teaching of core values to educate young players to become responsible and strong people of character. The strength of character is a set of overriding qualities, principles, and values like honesty, authenticity, integrity, and the engagement for others (Smoll, 2015). To achieve the development of favourable skills, qualities and values in the players, the program aimed to cultivate the development of an incremental theory of intelligence and talent (growth mindset) in the young soccer players (Dweck, 2017; Smith et al., 2009). Prerequisites for the expression and development of a growth mindset are a mastery-oriented environment, which is characterized by personal growth and team spirit (Ames, 1992; Smith, Shoda, Cumming, & Smoll, 2009). To operationalize and create a clearer understanding of the construct of strength of character seven sub-scales were created from overriding propositions, qualities, and values (Smith & Smoll, 1978; Smoll, 2015). This construct provided the framework for the character building intervention, with one of the main goals being the transfer of the concepts presented in Table 2.
2. Material and Methods
2.1. Participants and Design
This pilot study used a pre-post study design. The intervention group was formed by the current squad of the U12 SGF team players (n = 14). For baseline assessment two control groups, one club-internal group CG1 (U13 SGF team; n = 14) and one amateur sports club CG2 (U12 team of an amateur sports club; n = 14) were introduced to the study and tested again after the intervention period. The character building intervention for the IG took place weekly for a total of 15 weeks until the end of 2020.
2.2. Measures
Implicit Theories of Intelligence Scale
The German version of the Implicit Theories of Intelligence Scale (ITIS) was used to analyze the implicit theories of intelligence. The ITIS explores respondents’ beliefs about mastery (incremental) and performance (entity) mindsets (Dweck, 1999). The scale contains eight items with four items related to entity beliefs about intelligence and four items related to incremental beliefs. Responses are given using a six-point Likert scale (1 = strongly disagree; 6 = strongly agree). After reversing the items on the entity theory, high sum scores indicate a strong belief in an incremental theory and low sum scores in an entity theory. Regarding the psychometric properties, Blackwell et al. (2007) found an internal reliability of 0.78 with a test-retest reliability of 0.77 over a two-week period
![]()
Table 2. Sub-scale concepts of the character building program.
(Blackwell, Trzesniewski, & Dweck, 2007). Other findings support the validity of the ITIS scores (Cook, Castillo, Gas, & Artino Jr, 2017) and report high internal consistency for each domain (incremental items: Cronbach’s α = 0.81, entity items: Cronbach’s α = 0.78) (Da Fonseca et al., 2007; De Castella & Byrne, 2015). Results on the German version indicated a similar consistency and construct validity compared to the original version (Diseth, Meland, & Breidablik, 2014; Robins & Pals, 2002). We supplemented the ITIS with eight items to assess the implicit theories of talent in soccer with four items related to entity beliefs about talent and four items related to incremental beliefs.
Charakterschule-Questionnaire
The strength of character was measured by using a newly developed 23-item questionnaire (Charakterschule-Fragebogen, CSFB) which measured the seven main constructs of strength of character (see Table 2) using a six-point Likert scale (1 = strongly disagree; 2 = disagree; 3 = mostly disagree; 4 = mostly agree; 5 = agree; 6 = strongly agree). The CSFB items are scored that higher scores correspond with higher strength of character (the items 2, 6, 10 - 14, 16, 20 - 22 are reverse coded for the purpose of this study so that higher scores correspond with higher strength of character). The internal consistency was determined as acceptable (Cronbach’s α = 0.74) (Streiner, 2003) and test-retest reliability was established as moderate (intraclass-correlation-coefficient = 0.452) (Cicchetti, 1994).
2.3. Statistical Analysis
A mixed-method analysis was conducted. To analyze research question (1) a Mann-Whitney-U-Test was performed. The Mann-Whitney-U-Test is one of the most powerful non-parametric tests where the statistical power corresponds to the probability of rejecting a false null hypothesis (Landers, 1981). The test is applicable for small sample sizes (Nachar, 2008). We compared the independent groups IG and CG2 at baseline with respect to the dependent variables (H0: there is no statistical difference between the IG and CG2 in implicit theories of intelligence, implicit theories of talent and strength of character). For the second research question (2) the Wilcoxon’s signed-rank t-Test, a non-parametric analog to the paired sample t-Test, was conducted. Median within group differences in scores between baseline and post-intervention time-points (after 15 weeks) were analyzed for all three groups. The null hypothesis for the IG was: the character building intervention has no effect on the implicit theories of intelligent, implicit theories of talent and strength of character in the IG). The statistical tests were two-sided with an α risk of 0.05. All analyses were carried out by using the SPSS Statistics, Version 22. Research question (3) was analyzed qualitatively. After the end of the intervention, the IG wrote essays about their impressions of the character building project and the lessons, they had learned from the intervention phase. All essays were thematically analyzed by generating codes and then content themes by examining frequencies, relationships, and interconnections between themes. The results of the qualitative analysis are not discussed in the results section, but are taken up in the discussion. Appendix shows the four main themes identified with exemplary statements from the players’ essays.
3. Results
The mean age of the whole sample (N = 42) was 11.3 years. In IG and CG1 all individuals completed pre- and post-testing, in the CG2 post data from 12 players was collected.
3.1. Baseline Differences between Amateur and High-Performance Youth Soccer Clubs
The implicit theories of intelligence scores in the IG (Mdn = 30.00) did not differ significantly from the CG2 (Mdn = 29.50) at baseline, U = 77.00, z = −0.97, p = 0.352, r = −0.18. Implicit theories of talent scores were significantly higher in the IG (Mdn = 32.50) compared to the CG2 (Mdn = 28.50) at baseline, U = 48.50, z = −2.28, p = 0.021, r = −0.43. The strength of character scores were also significantly higher in the IG (Mdn = 98.00) compared to the CG2 (Mdn = 88.00) at baseline, U = 30.00, z = −3.13, p = 0.001, r = −0.59. Figure 2 further presents the means for the IT IS and CSFB scores as well as the seven domains of strength of character. The results indicated that the IG scored significantly higher than the CG2 in the domains work ethic, dealing with mistakes and handling pressure at baseline.
3.2. Effects of the Character Building Intervention
The implicit theories of intelligence scores were significantly higher in the IG after the intervention (Mdn = 36.50) compared to baseline (Mdn = 30.00), z = −2.93, p = 0.003, r = −0.55 while for the implicit theories of talent, scores did not
![]()
Figure 2. Baseline mean values for IG and CG2. *p < 0.05, nIG = 14, nCG2 = 14.
differ significantly between post-test measure (Mdn = 36.50) and baseline (Mdn = 32.50), z = −1.79, p = 0.074, r = −0.34. The strength of character scores for the IG were significantly higher after the intervention (Mdn = 107.50) than at baseline (Mdn = 98.00), z = −2.89, p = 0.004, r = −0.55. Wilcoxon-test further showed significant improvements in the strength of character domains perception of mistakes, effort and self-worth. No significant changes were found for CG1 and CG2 (see Figure 3).
4. Discussion
4.1. Main Findings
The purpose of this pilot study was to explore the impact of a character building program on the implicit theories of intelligence and talent as well as the strength of character. It was demonstrated that implicit theories of talent and the strength of character were higher in the players of the high-performance youth academy than in the amateur youth team at baseline. No difference was found for incremental beliefs about intelligence, which is similar to previous research (Best, 2018). Given the high performance youth academy players’ mastery goal orientations, it would be logical for the players to have higher ratings of growth mindset as well. Sigmundsson et al. (2020) compared football players and university students and found no significant difference in the scores on mindset, but mainly regarding passion and grit. They argued that both groups, professional and non-professional players may need growth mindset for their achievement (Dweck, 2017; Sigmundsson, Clemente, & Loftesnes, 2020).
![]()
Figure 3. Baseline and post-intervention mean values for the IG, CG1, CG2. p < 0.05. Note: nCG2post = 12.
In terms of the effectiveness of the intervention, the results revealed that the IG had significantly higher scores on strength of character and implicit theories of intelligence after the intervention. As hypothesised, the character building intervention led to a move towards a growth mindset in the players. The qualitative analysis reflected this result. Thus, the most prevalent theme within the essays was the importance of characteristics and personality traits one should possess in order to be a person of strong character (e.g., differences between growth and fixed mindset, ambition and work ethic, dedication, responsibility, self-efficacy, strength of character). Another frequent theme was overriding values and principles, where the players described and discussed the core values they had learned in the character building program (e.g., cohesion, joy, respect) (see Appendix).
The results of this study support the previous finding by Dweck and Leggett (1988) that implicit theories are malleable, in the short term (Dweck & Leggett, 1988). However, while the effectiveness of growth mindset interventions in education has been well investigated, there is still limited and inconclusive research on interventions in the sports context. Results from growth mindset interventions in other sports (softball, field/cross-country) also showed higher levels of growth mindset and lower level of fixed mindset from pre- to post-intervention, however, these changes were not statistically different (Williams, 2018).
4.2. Strength and Limitations
The character building program was conceptualized based on a solid scientific background, integrating a theoretical framework of AGT, MC and TL. In addition, the study explicitly connects these theoretical concepts to conceptualize a character building intervention in high-performance athletic youth setting as one of the first. This pilot study serves as one of the first investigations and longitudinal studies in the field of explicit and implicit character building. To date, only a few studies in athletic settings have used a longitudinal approach of this kind (Papaioannou, Marsh, & Theodorakis, 2004).
Like many other interventions in athletic settings, the sample size was small and non-randomized. In addition, coaching philosophies and leadership styles in the different teams varied from coach to coach, which could have influenced the outcomes. This comes along with limitations of generalizability and transferability of the results. Moreover, the Covid-19 pandemic led to adjustments of the intervention as from November 2nd meetings were held in an online setting via zoom-platform. From a growth mindset perspective, this challenge also posed an opportunity as this thesis is able to demonstrate that character building is not only possible in real-life environments (e.g., in classrooms), but also in online-settings. Similar interventions also demonstrated that the cultivation of a growth mindset can take place in virtual spaces (Blackwell et al., 2007; O’Rourke, Haimovitz, Ballweber, Dweck, & Popović, 2014). As half of this study was conducted online, the results indicate the possibility of delivering character building programs to youth athletes via internet. To the best of our knowledge, this is the first and only intervention of this kind in the athletic domain. Another issue worth noting is that the CSFB-questionnaire has been designed just for this intervention. To increase the quality of psychometric properties, internal consistency and test-retest reliability was determined. Another limitation is the lack of a follow-up measure to examine mindset changes beyond the intervention period.
4.3. Implications for Future Research and Practice
Adding to existing literature, this pilot study supports the need for greater incorporation of theoretical concepts such as AGT, MC, as well as the TL. Greater emphasis should also be placed on values-based leadership to enhance positive sport experiences (Castillo, Adell, & Alvarez, 2018), athlete performance outcomes (Álvarez et al., 2013; Morgan et al., 2015), team task cohesion (Price & Weiss, 2013), and other positive outcomes in youth soccer and talent development. Future interventions should involve coaches, parents, and other responsible parties (e.g., athletic directors, coordinators) to support the development of a growth mindset in young athletes. Involving caregivers can create a motivational environment that can foster player learning and development. In addition, transformative leadership styles can be promoted among coaches (Cumming et al., 2007; Smith & Smoll, 1978; Sousa, Smith, & Cruz, 2008). In addition, it should be ensured that commonly agreed upon core values, once established, are lived by all protagonists in the organisation every day even after the intervention (Castillo et al., 2018). Integrity and an infinite-minded long-term approach to talent development is of utmost importance when it comes to the sustainability of these programs (Sinek, 2019). Previous studies already examined the relationship between character development and academic performance (Olowookere, Alao, Odukoya, Adekeye, & Agbude, 2015). Future analogous findings related to athletic performance could further support the importance of character training in sports. Future studies should investigate this research topic of this work, while focusing on bigger, randomized sample sizes and on the validation of questionnaires to assess mindset, and the construct of strength of character.
5. Conclusion
The character building program described here supported the formation of strength of character and implicit theories of intelligence among the players of a high-performance youth academy soccer team. The development and success of such programs support the role of the athlete beyond purely physical athletic performance, bringing to the fore the social responsibility of youth athletes coaches in terms of psychological aspects.
Appendix
![]()
Table A1. Main themes identifies within the qualitative analysis with examples from the player’s essays.
Ubuntu is of African origin and means a collection of values and practices that make people authentic human beings. Ubuntu points out that an authentic individual human being is part of a larger and more significant relational, communal, societal, environmental and spiritual world (Mugumbate & Chereni, 2020).