An ecological meta-analysis of coach education for sport-based youth development

Abstract

Playing sports has been a favorite activity among young people for decades and holds considerable promise for both health promotion and educational purposes. Evidence highlights the central role of coaches and the value of coach education program interventions (CEPIs). Yet, the extent to which these programs foster social, cognitive, and emotional development in youth athletes remains underexplored. To address this gap, outcomes (k = 193) were extracted from 27 studies (N = 4345) conducted over the past two decades. Results from frequentist and Bayesian meta-analyses indicate that CEPIs are associated with positive youth development, with effect sizes ranging from small to moderate across individual- and micro-level outcomes. These findings underscore the role of trained coaches in creating mastery-oriented and autonomy-supportive motivational climates that foster learning, social interaction, and skill development through sport participation. Taken together, the results extend understanding of theory-driven coach education as a mechanism for sport-based youth development from an ecological perspective and point to directions for future research on program design and validation.

Keywords

Autonomy support emotional skills motivational climate positive youth development social interaction

Sport participation provides numerous opportunities to support youth development and contribute to population health.^1,2 In the United States, more than 60 million children aged 6–17 years participate in sports annually.^3,4 Beyond promoting health and well-being, sport offers rich educative contexts for learning and skill development.^3,5 For young people, these opportunities include cultivating skills such as goal setting, decision making, problem solving, responsibility, leadership, and prosocial behaviors.⁶ Yet, sport participation does not inherently yield positive developmental outcomes.^7,8 In some cases, it has been linked to antisocial conduct, substance use, depression, injury, and burnout.^9,10

Coaches play a central role in shaping the youth sport environment and athlete experiences.^11–13 Effective coaches not only ensure safe participation but also foster the holistic development of young athletes.¹⁴ A growing body of literature highlights the potential of coach education programs to enhance developmental outcomes in youth athletes.¹⁵ As Jowett¹⁶ emphasized, coach–athlete relationships sit at the heart of coaching effectiveness. In youth sport, effectiveness rests on athlete-centered coaching, which prioritizes holistic development and athletic performance.^14,17,18 Over the past decades, numerous theory-driven coach education program interventions (CEPIs) have been developed to support youth athletes’ basic psychological needs and development through sport participation.

More specifically, informed by theories and frameworks in education and psychology, CEPIs have been developed using, for example, Self-Determination Theory (SDT),¹⁹ Achievement Goal Theory (AGT),²⁰ and Positive Youth Development (PYD).²¹ Programs such as Coach Effectiveness Training (CET) and the Mastery Approach to Coaching (MAC) are rooted in AGT and designed primarily to foster coaching behaviors that facilitate positive athlete behavior and a mastery-oriented motivational climate, an environment where effort, task mastery, self-improvement, cooperative learning, and having fun are valued over winning and outperforming others.^22–24 Others draw upon SDT, advocating autonomy-supportive coaching behaviors (e.g., offering choices, providing rationale for tasks and limits, providing structure and involvement) over controlling (e.g., intimidation, contingent rewards) to prepare coaches with strategies to meet athletes’ needs for autonomy, competence, and relatedness.^25,26 Research advocates for the contextual integration of PYD-focused material into coach education^27,28 and suggests it can be prioritized within the courses²⁹ in order to support coaches’ ability to deliver targeted developmental outcomes for the athletes.^30–33 These outcomes have been conceptualized as life skills, defined as “ internal personal assets, characteristics and skills such as goal setting, emotional control, self-esteem, and hard work ethic that can be facilitated or developed in sport and are transferred for use in non-sport settings” (p. 60).³⁴ When CEPIs focus on both implicit and explicit structures for teaching life skills development and transfer, coaches are better positioned to foster PYD and apply PYD principles and strategies into coaching practice.^33,35 However, the coach's ability to foster PYD principles, affect and sustain change after completing an educational course is not guaranteed,³² often due to the lack of a practical component in the training among other reasons.³³ In addition, other CEPIs address areas including mental health,³⁶ ethical and moral reasoning,³⁷ humanistic coaching,³⁸ and concussion awareness and management.^39,40 This evolution of CEPIs beyond tactical and technical competencies toward holistic approaches reflective of broader societal shifts is also evident in recommendations to reimagine the PYD-focused coach education through a social justice perspective by integrating social justice life skills into coach education and development.⁴¹ According to Newman and colleagues,⁴¹ the goal is to help coaches empower youth to engage with complex social systems and work towards more socially just society by providing youth with opportunities to learn, discuss, and practice social justice life skills (e.g., allyship, advocacy, anti-racism).

More than 70% of coach education studies have reported positive coach and athlete outcomes following CEPIs.¹⁵ CEPIs demonstrated effectiveness as coaches were able to apply program principles rooted in SDT and AGT and improve coaching practices: for instance, increased use of game-based play-form activities and application of strategies to support athlete's basic needs for autonomy, relatedness, and competence.^42–44 Additional outcomes also showed significant improvements, including increased coaching knowledge for ensuring psychological and physical safety^36,45–47 and enhanced interpersonal and leadership skills.^48,49 These outcomes reflect an increasing emphasis in coach education on coaching as a relational and safeguarding practice rather than purely technical instruction. Overall, the coaches indicated that participation in CEPIs helped them become more aware of how to create a supportive and motivational environment for their athletes, which they believed would contribute to athletes’ well-being and long-term participation in the sport.

Beyond coach professional development, CEPIs are expected to prepare coaches with the knowledge and competencies needed to support youth development in safe, mastery-oriented and autonomy-supportive sport environments. It is important to note that CEPIs do not directly “produce” outcomes for the athletes, rather they are intended to re-shape coach attitudes and behaviors (i.e., how coaches create the participation environment), which, in turn, impact athlete self-reported experiences in sport. Several studies provide evidence of the effectiveness of different types of interpersonal coach development programs on a variety of positive youth outcomes: increased perception of task-related climate, reduced anxiety, increased self-esteem, enhanced team cohesion, improved perceptions of coach competence, and increased fun and enjoyment (see Bengtsson et al.⁶ for review). For instance, athletes who played for MAC trained coaches reported a stronger motivational climate, greater increases in mastery goal orientation, and decreases in anxiety.^22,23 While athletes who played for coaches trained through the need supportive interpersonal interventions based on SDT reported higher levels of self-determined motivation⁵⁰ and lower levels of burnout.⁵¹ Similarly, athletes with coaches trained in humanistic coaching principles engaged less frequently in anti-social behaviors and reported stronger connections to their coaches, which can be attributed to relationship-building and autonomy-supportive coaching behaviors.⁵²

These findings are consistent with other studies: trained coaches may build stronger coach-athlete relationships that positively impact youth development.^9,49 When athletes perceive an increase in caring, supportive, or transformational behaviors from the coach, they experience significant gains in psychosocial experiences, such as improved self-confidence,⁵³ self-esteem,⁵⁴ and enhanced cognitive functioning and goal-setting.⁴⁹ Trained coaches’ ability to create a positive learning environment and deliver effective, game-based coaching sessions (e.g., making fun part of the skill development process, offering diversified activities to the athletes) is connected to improvements in intrinsic motivation, task orientation, game awareness, wellbeing, and enjoyment of sport for the athletes.²⁵ Taken together, these results speak to the potential efficacy of well-delivered and theoretically informed CEPIs on a coach's ability to create a more supportive and empowering environment for athletes, impact the quality of youth sports experience, and drive athlete retention. However, some studies have reported inconsistent, non-significant, or even contradictory results, particularly at the athlete level, underscoring the need to better understand the impact of CEPIs on youth development through sport participation.¹⁵

Current study

Previous reviews have examined youth athlete outcomes using a range of theoretical frameworks and perspectives.^6,15,55 Despite these contributions, a gap remains in organizing these outcomes within a broader and more integrative framework. Across CEPIs that emphasize holistic athlete development, outcomes have been assessed across domains such as social, cognitive, emotional development, life skills, and self-related beliefs.⁵⁶ These outcomes are often studied in isolation or within a specific theoretical lens, limiting the ability to synthesize findings across studies and domains.

An ecological perspective provides a coherent framework to address this limitation. Grounded in ecological systems theory,^57,58 development is understood as emerging from dynamic interactions between individuals and their environments. Such a perspective is particularly relevant in youth sport and physical activity contexts, where developmental outcomes are shaped by multiple, interconnected systems, including individual characteristics, coach–athlete relationships, team environments, and broader sociocultural structures.^31,59,60

Building on this perspective, the current study adopts the Ecological Approaches to Social Emotional Learning (EASEL) taxonomy developed by EASEL Lab.^61,62 The EASEL taxonomy was created to organize and connect non-academic developmental outcomes across theoretical frameworks commonly used in education and developmental psychology.⁶³ Rather than advancing a single unified theory, EASEL provides a structured taxonomy that identifies areas of convergence and distinction across constructs by grouping them into six broad domains: cognitive, social, emotional, identity, perspectives, and values.⁶¹

These six domains represent broad clusters of related non-academic skills and capacities. In the EASEL coding system, cognitive regulation includes processes such as attention control, working memory and planning, inhibitory control, and cognitive flexibility; emotion processes include emotion knowledge, emotion expression, and emotion and behavior regulation; and social or interpersonal skills include understanding social cues, conflict resolution, and prosocial behavior. The remaining domains capture values, including ethical, performance, intellectual, and civic values; perspectives, including optimism, gratitude, openness, and enthusiasm or zest; and identity or self-image, including self-knowledge, purpose, self-efficacy, growth mindset, and self-esteem.^61,62

The EASEL taxonomy is particularly well suited for the current study. First, youth sport outcomes are predominantly non-academic and align closely with constructs emphasized in frameworks such as SDT, AGT, and PYD. Second, these outcomes reflect interconnected developmental processes rather than discrete categories, consistent with ecological perspectives on child development.⁶² Third, the EASEL taxonomy provides a practical structure for synthesizing diverse outcome measures across studies, which is critical for meta-analytic work. Accordingly, the six-domain framework offers a coherent and ecologically grounded lens for mapping athlete outcomes reported in CEPIs research and for examining how these interventions influence development across multiple domains.

With the growing number of studies evaluating CEPIs, the current review aims to examine whether these interventions demonstrate consistent and potentially scalable effects on youth development. Specifically, the objectives are to: (a) map athlete-level outcomes using the EASEL taxonomy^61,62; and (b) estimate the impact of CEPIs on these outcomes using both frequentist and Bayesian meta- analytic approaches.

Method

The present meta-analytic review was developed by following the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines.⁶⁴ The review protocol was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO), an online database for systematic review protocols (CRD42023479392).

Search and screening

The literature search was conducted by using seven online bibliographic databases: PubMed, SPORTDiscus, PsycInfo, MEDLINE Complete, Education Resources Information Center (ERIC), Dimensions, and Google Scholar, spanning from 2003 to 2024. A three-step search strategy was used. First, a Boolean/Phrase search restricted to peer-reviewed, English-language articles. Search mode was set as “Boolean/Phrase” with “and”, “or” as Boolean operators. A combination of the following keywords was used to search the relevant articles: “youth”, “adolescents”, “young people”, “teen”, “child*”, “kids”, “young athlet*”, “youth athlet*”, “coach training”, “coach workshop”, “coach education”, “professional development”, “learning”, “coaching effectiveness”, “efficiency”, “knowledge”, “attitude”, “intention”, “behavi*”, “psychosocial”, “social”, “support*”, “autonomy”, “motivation*”, “competence”, “relatedness”, “leadership”, “mastery”, “ego”, “emotion*”, “well-being”, “wellbeing”, “mental health”, “satisfaction”, “regulation”, “safe*”, “develop*”, “positive youth development”, “PYD”, “social emotional learning”, “cognitive”, “goal orientation”, “achievement”, “interperson*”, “intraperson*”. Abstracts were screened for relevance, followed by full-text reviews to exclude ineligible studies. Cross references and reference lists of included articles were also examined, and backward searches in relevant review papers ensured comprehensive coverage.

Inclusion and exclusion criteria

Using the PICOS framework, which outlines participants, interventions, comparison conditions, outcomes, and study designs, inclusion and exclusion criteria were established based on the purpose of the study. The review included youth athletes aged 18 or younger. Studies were excluded if they involved college or professional coaches, athletes over 18 at data collection, individuals with specific health conditions, lacked detailed training descriptions, or focused solely on motor skills, physical activity, fitness, injury, or other clinical outcomes. Interventions included in-person, online, and hybrid training programs or workshops lasting 75 min to 12 months, aimed at enhancing coaching skills, promoting youth development and well-being, and supporting holistic growth in and beyond sport. The outcomes emphasize life skills as well as social, cognitive, and emotional development, all of which are connected to both individual-level and micro-level influences. Eligible designs included randomized controlled trials (RCTs), cluster RCTs, non-randomized studies, or mixed-method designs; meta-analyses and systematic reviews were excluded. Titles, abstracts, and full texts were screened for eligibility, with references managed in Mendeley. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram outlines the selection process in Figure 1, with the PRISMA checklist in Supplementary Table S1.

Figure 1.

PRISMA flowchart of search strategy and screening.

Assessment of risk of bias

Risk of bias was assessed using Cochrane Collaboration tools,⁶⁵ evaluating confounding factors, participant selection, intervention, missing data, outcome measurement, and selective reporting. Bias was rated as low, moderate, serious, or indeterminate. For non-randomized studies, the risk of bias in non-randomized studies of interventions (ROBINS-I) tool was applied as part of the Grading of Recommendations Assessment, Development and Evaluation (GRADE)'s certainty rating process.^60,66 GRADE and ROBINS-I were jointly used to ensure a comprehensive quality assessment. Findings were visualized with the Risk-of-bias VISualization (Robvis) tool,⁶⁷ with details in Supplementary Figure S1.

Quality assessment and level of evidence

Study quality was assessed using the PRISMA checklist and a modified Critical Appraisal Skills Program checklist.^68,69 A seven-question framework evaluated research question clarity, study design, methodological transparency, and appropriateness of results and interpretation. Items were scored 1 (agree) or 0 (disagree), with cumulative scores classifying studies as high (7), moderate (4–6), or low (<4) quality; low-quality studies were excluded from meta-analysis. Following Ackley and colleagues,⁷⁰ RCTs were rated as the highest evidence level, non-RCTs second, and uncontrolled studies third. Quality ratings and evidence levels are reported in Supplementary Tables S2.

Data preparation and analysis

Characteristics of each study, including demographics, design, intervention, results, were systematically reviewed for data preparation. Extracted information included authors, study location, journal, sample demographics, coaching experience, statistical methods, theoretical frameworks, coach training type and duration, key outcomes, limitations, and novel contributions. Athlete-level outcomes were then identified and categorized as either individual or micro level, with the latter encompassing measures beyond the individual such as coach–athlete interactions, perceived coaching behaviors, and team climate. Individual-level outcomes were further classified into six domains in accordance with the operational definitions outlined in the EASEL taxonomy.⁶¹ For each domain and subdomain, operational definitions, athlete-related measured outcomes, categorization criteria, and measurement tools are listed in Supplementary Table S3.

In addition, because a single study could report multiple outcomes in distinct domains, outcomes at different levels were treated as separate entities.⁷¹ For example, a study reporting both cognitive and social outcomes contributed two effect sizes. The standardized mean difference (SMD) was used to quantify effect size, calculated from sample sizes, mean changes, and standard deviations for control and intervention groups at baseline and post-test to assess mean differences over time.

Statistical analyses

Descriptive statistics were used to summarize study characteristics and quality assessment results. Meta-analyses were conducted in R (version 4.4.3; R Core Team, 2018) using both Bayesian and frequentist approaches to strengthen methodological rigor and cross-validate findings.^72–74 To address potential nesting of effect sizes within studies, three-level hierarchical meta-analyses were implemented,⁷¹ with two- versus three-level models compared to assess gains in pooled effect estimation accuracy.⁷⁵

Frequentist analyses used a random-effects model (metafor package),⁷⁶ with heterogeneity evaluated via Q and I² statistics⁷⁷ and classified as none (0%), low (25%), moderate (50%), or high (75%), according to Higgins and colleagues.⁷⁸ Small-study bias was tested with Egger's regression and funnel plot asymmetry,⁷⁹ and a three-parameter selection model was applied for publication bias sensitivity and correction at cut-points α₁ = .025 and then α₁ = .050.^80,81 Effect sizes (SMD) were interpreted as small (0.15), medium (0.36), or large (0.65) following rationales in educational and sport psychology research.^6,82

Bayesian analysis conducted with the brms package.⁸³ According to Williams and colleagues,⁸⁴ the model was specified with a weakly informative prior for the pooled effect size, μ ∼ N (0, 1), and a Half-Cauchy prior for the between-studies standard deviation, τ ∼ HC (0, 0.5). Posterior distributions for μ and τ were summarized using the mode, median, mean, standard deviation, and 95% credible intervals (CrI), with model convergence assessed via $\hat{R}$ , effective sample size, and posterior predictive checks. Sensitivity analyses compared the original model against versions with literature-informed, diffuse, skeptical, and Student-t priors for μ, and conservative or expansive priors for τ, accounting for possible overlap between prior sources and included studies. Predictive performance was compared using Leave-One-Out cross-validation.⁸⁵ Subgroup analyses examined coaching experience, athlete age, and CEPI delivery format.

Results

Characteristics of the included studies

Twenty-seven studies (N = 4345) met the inclusion criteria; these studies included 393 coaches (M = 14.04, SD = 8.19) and 3952 youth athletes (M = 141.14, SD = 75.39). Studies varied in sport, location, and research design, as detailed in Table 1. As for the selected studies, quality assessment scores ranged from five to seven, with a median of six, indicating moderate to high methodological quality. The median publication year was 2016 (M = 2016, SD = 6.32), with studies published in 20 peer-reviewed journals spanning education, psychology, and kinesiology. Coaches had a median age of 38 years (M = 37.16, SD = 8.63) and an average of 6.41 years coaching experience (SD = 3.73). Athletes, aged 7–18, participated in eight sports across 21 studies,^{5,22–26,37,49–51,53,54,86–94} while six studies reported CEPIs involving multiple sports.^{32,47,52,95–97} In studies with control arms, group allocation was balanced overall, with 52% of athletes assigned to experimental groups.

Table 1.

Characteristics and research designs of selected studies.

	References numbers	No. of studies
Type of Sports
Soccer	25, 49, 50, 53, 88, 93, 94	7
Basketball	22–24, 37, 91, 92	6
Mixed	32, 47, 52, 95, 96, 97	6
Swimming	54, 86	2
Rugby	87, 90	2
Gaelic football	51	1
Handball	89	1
Netball	5	1
Rowing	26	1
Study Sites
United States	22–24, 37, 53, 54, 86, 91, 93	9
Australia	5, 25, 47, 49, 87, 92	6
Canada	52, 94	2
Spain	50, 88	2
Portugal	32, 89	2
Poland	96, 97	2
Ireland	51, 90	2
Singapore	95	1
United Kingdom	26	1
Research Designs
Qusai-Experimental	22–24, 32, 47, 49, 50, 52, 88, 89, 90, 91, 94, 96, 97	15
RCT	5, 25, 54, 55, 87, 92	6
Mixed Methods	26, 37, 51, 53, 95, 93	6

Note. Twenty-seven studies are included in the current study: 53 = Blom et al. (2011); 54= Conroy & Coatsworth (2004); 86 = Coatsworth & Conroy (2006); 87 = Denison et al. (2023); 5 = Eather et al. (2021); 52 = Falcão et al. (2020); 92 = Guagliano et al. (2015); 25 = Jones et al. (2023); 95= Koh et al. (2024); 51 = Langan et al. (2015); 93 = Langdon et al. (2015); 32 = MacDonald et al. (2020); 26 = Mahoney et al. (2016); 94= McLaren et al. (2015); 37 = Power & Seroczynski (2015); 50 = Pulido et al. (2017); 88 = Pulido et al. (2021); 89 = Simon et al. (2010); 90 = Smith et al. (2024); 22 = Smith et al. (2007); 23 = Smoll et al. (2007a); 24 = Smoll et al. (2007b); 91 = Sullivan (2005); 47 = Vella et al. (2021); 49 = Vella et al. (2013); 96 = Wilczyńska et al. (2021); 97 = Wilczyńska et al. (2022); RCT = Randomized Control Trial; Mixed sports include multiple sports for instance soccer, basketball, football, baseball, volleyball, ice hockey, floorball, rugby, swimming, equestrian, etc.

CEPIs varied in formats and duration. Most (81%) were delivered in person,^{22–24,26,37,49–51,53,54,86–92,94–97} 15% used hybrid formats combining in-person and online sessions,^{5,25,32,47,52,93} and one was fully online, incorporating visual presentations and supplemental resources.⁹⁴ In-person CEPIs ranged from 30 min⁸⁷ to 18 h,^96,97 while hybrid programs typically spanned several weeks and were longer overall. Examples include a 1-h in-person session supplemented by self-paced online modules,⁹³ and a 4-h workshop paired with mentoring, online discussions, and a 1.5-h mid-intervention evaluation over eight weeks.²⁵

Theories and frameworks included the mediational model of coaching behaviors, as applied in CET and MAC,^{22,23,47,49,53,54,86,89,94} SDT,^{26,37,50,51,88,90,91,93} PYD,^32,52,95 and PERMA.^96,97 Some CEPIs were informed by Social Cognitive Theory⁸⁷ or integrated multiple approaches,⁹⁰ such as SDT, Explicit Learning Theories (ELT), and Implicit Learning Theories (ILT). Although all aimed to foster youth development through sport, training objectives varied, including building humanistic coaching skills, applying SDT-based strategies to support athlete need satisfaction, addressing moral reasoning, refining nonverbal behavior, promoting game-based coaching, and enhancing prosocial and emotional learning behaviors. Additional details on CEPI design and implementation protocols are provided in Supplementary Table S4. A narrative review of key outcomes and notable findings are illustrated in Table 2.

Table 2.

Key findings and unique contributions of selected studies.

Ref.	Study	Age/Grade level	Participants	Key findings	Unique or novel aspects
53	Blom et al. (2011)	14–18 years	Coach N = 9 (EG = 2, CG = 4, FDG = 3); Athlete N = 93 (EG = 21, CG = 43, FDG = 29)	Athletes in the EG exhibited a significant increase in perceived coach caring behaviors, athlete self-confidence, and intrateam attraction.	Coaches and athletes are males. Included feedback group. The training is more focused on improving socio-emotional competencies of coaching behaviors by individualized program.
86	Conroy & Coatsworth (2004)	7–18 years (M = 11.40, SD = 2.23)	Coach: N = 7 (EG = 4, CG = 3; 6F, 1 M); Athlete: N = 135 (EG = 66, CG = 69; 83F, 52 M)	The training that coaches received did not significantly predict the rate of fear of failure change in young athletes. An increase in reward/reinforcement and a decrease in encouragement for coaches in the EG.	With consideration of pretraining coach behaviors. More extensive inventions were suggested (e.g., lasting longer than 2 h, having periodic booster sessions, etc.).
55	Coatsworth & Conroy (2006)	7–18 years (M = 11.40, SD = 2.23)	Coach: N = 7 (EG = 4, CG = 3; 6F, 1 M); Athlete: N = 135 (EG = 66, CG = 69; 83F, 52 M)	Athletes demonstrated small increases in self-esteem, with no significant effects on changes in youth self-esteem. The intervention proved more successful in improving self-esteem for younger youth and for girls with initially low levels of self-esteem.	Girls and younger youth (age 11 and below) who entered the study with low levels of positive self-esteem beneﬁted most from their time with psychosocially trained coaches.
87	Denison et al. (2023)	EG: 18.00 years (SD = 1.30); CG: 17.70 years (SD = 1.20)	Coach: 13; Athlete: N = 167 (EG = 91; CG = 76)	The intervention approach did not significantly reduce athlete homophobic language use, or alter the associated norms and attitudes, relative to controls.	First RCT of anti-homophobia educational intervention. Participants are male youth athletes.
5	Eather et al. (2021)	8–16 years	Coach: N = 16 (EG = 8,CG = 8; 15F, 1 M); Athlete: N = 131	Significant interventions effects were observed for athletes’ game skills (defense, support, decision-making, and total game skills) and physical self-perceptions. No significant changes were observed for athletes’ enjoyment, motivation, or wellbeing.	The MASTER coach development program was implemented (a strengths-based and game-based approach). Provide a deeper understanding of game-based coaching pedagogies and the development of practical delivery techniques (including feedback practices) in a community sport setting.
52	Falcão, Bloom, & Sabiston (2020)	12–17 years	Coach: n = 17 (EG = 10; CG = 7); Athlete: n = 110 (EG = 74; CG = 36)	Athletes in EG showed a higher connection with their coaches. Both EG and CG showed an increase in antisocial behavior at the end of the season. However, EG showed engaging less frequently in antisocial behaviors compared to CG. No significant changes were found in competence and confidence over time by groups.	Learning humanistic principles may help youth sport coaches impact athletes’ PYD outcomes (i.e., 4Cs).
92	Guagliano et al. (2015)	9–12 years	Coach N = 9 (all M); Athlete N = 76 (EG = 38, CG = 38; all F)	Athletes in the EG spent a signiﬁcantly higher proportion of practice time being physically active compared to the CG. No signiﬁcant changes in motivation by group across time.	Coaches are males and athlete are females. Coach education sessions could increase girls’ PA and decrease inactivity without deleterious effects on their motivation.
25	Jones et al. (2023)	8–16 years (EG = 11.7, CG = 10.5)	Coach: N = 20 (EG = 10, CG = 10); Athlete: N = 200 (EG = 108, CG = 92)	Athletes in the EG displayed greater growth in their decision-making abilities in the game. Athletes and coaches in the EG signiﬁcantly increased their conﬁdence, self-perception of competence, and their enjoyment of the game.	Extending previous evidence on netball to football. The first intervention provided coaches with specific feedback training. Athletes’ game awareness was improved through quality playing-form activities, resulting in athletes becoming more conﬁdent about their football and increasing their enjoyment of the sport.
95	Koh et al. (2024)	Under 18 years	Coach: N = 27 (EG = 14, CG = 13); Athlete: N = 85 (EG = 40, CG = 45)	No significant time or interaction effect was found concerning anti- and pro-social behaviors toward teammates and towards opponents.	Evaluating the Values and Principles in Sport coach education course for promoting values-driven coaching practices in Singapore.
51	Langan et al. (2015)	11–17 years	Coach: N = 6 (EG = 3, CG = 3); Athlete: N = 87	Training was an effective approach to prevent increases in athlete burnout. Athletes in the EG generally showed little change or slight decreases in exhaustion and reduced sense of accomplishment.	This study offers the first evidence that SDT-based training for coaches is an effective approach to prevent increases in athlete burnout.
93	Langdon et al. (2015)	8.22 years (SD = 1.87)	Coach: N = 7 (2F, 5 M); Athlete: N = 46 (13F, 33 M)	No significant changes in athlete level outcome variables. Coaches believed the training influenced their coaching philosophy regarding the coach-athlete relationship and communication styles, emphasizing choice and rationales.	Focus on volunteer coaches since they represent most youth sport coaches in recreational leagues.
32	MacDonald et al. (2020)	EG: 13.90 years (SD = 2.40); CG: 13.80 years (SD = 1.90)	Coach N = 14 (EG = 8, CG = 6; 4F, 10 M); Athlete N = 270 (EG = 154, CG = 116)	Athlete reports of their developmental experiences and coach-athlete relationship were high at all time points and did not change over time. Coaches in the EG increased the frequency of their PYD-oriented behaviors during the intervention but returned to levels slightly above baseline at follow-up.	This study provided evidence on a PYD-based coach education intervention has the potential to increase the frequency of PYD-oriented behaviors in coaches but not reflect accordingly on athletes.
26	Mahoney et al. (2016)	EG: 15.33 years (SD = 1.31); CG: 14.77 years (SD = 1.68)	Coach N = 18 (EG = 10, CG = 8; 1F, 17 M); Athlete N = 113 (EG = 53, CG = 60; 78F, 35 M)	No significant effect was found for athletes perceived autonomy supportive coaching, perceived controlling coaching, psychological needs satisfaction, and mental toughness over time. Psychological needs thwarting was found higher at follow-up than post-intervention and baseline.	Qualitative data revealed that autonomy-supportive behaviors might not have been adopted due to contextual pressures (e.g., limited comprehension, relevance, competing time demands, and relapses to previous style) on the coaches.
94	McLaren, Eys, & Murray (2015)	10–18 years (M = 13.27, SD = 2.15)	Coach N = 20 (5F, 15 M); Athlete N = 243 (154F, 89 M)	Compared to the CG and AG, significant time×condition interaction across task cohesion (only to the AG), social cohesion, and task climate in the EG. Coach in the EG reported greater use of the various task-related behaviors across the duration of the study.	The study introduced an attention-control condition to discern if the intervention specifically fostered a task climate or just raised awareness.
37	Power & Seroczynski (2015)	12–16 years	Coach N = 19 (EG = 5 teams, CG = 5 teams); Athlete N = 341 (144F, 197 M)	Athletes in the EG reported statistically significant higher levels of moral engagement, perceived positive experiences of team and coach behavior, and lower levels of perceived negative coaching behaviors. Coaches in the EG accepted their role as moral educators and made an effort to foster their players’ character development.	Designed a qualitative measure consisting of four moral dilemmas with four to six accompanying questions to assess athletes’ sports-related moral reasoning, combined with the quantitative measures. Educational efficacy was found to differ between boys’ and girls’ teams, with mixed results.
50	Pulido et al. (2017)	11–15 years (EG: M = 13.92, SD = 1.23; CG: M = 13.57, SD = 1.48)	Coach: N = 8 (EG = 4, CG = 4); Athlete: N = 109 (EG = 53, CG = 56)	Athletes in the EG decreased competence and relatedness control, while significantly increased competence and relatedness need satisfaction, and sport commitment. Intrinsic motivation decreased in both groups (greater decrease in the CG).	This STD-based coach training was effective in optimizing the “bright side” (support) of motivation and reducing levels of the “dark side” (thwarting and frustration).
88	Pulido et al. (2021)	10–16 years (M = 13.78, SD = 1.38)	Coach: N = 8 (EG = 4, CG = 4); Athlete: N = 117 (EG = 59, CG = 58)	Significant changes in coaches’ game strategy competency, technique competency, character-building competency, and enjoyment as perceived by the players from the EG compared with the players from the CG. Also, both the CG and the EG athletes decreased their satisfaction with the coach.	Athlete-perceived coaching competency factors increased or else maintained their scores in the EG, opposite to Gonzalez-Ponce et al. (2018), and athletes’ intention to persist did not differ.
89	Simon et al. (2010)	13–18 years	Coach: N = 9 (EG = 4, 1F, 3 M; CG = 5, 3F, 2 M); Athlete: N = 113 (EG = 55, 12F, 33 M; CG = 58, 37F, 21 M)	Athletes in the EG demonstrated a significantly greater increase in positive evaluations of their coaches, accompanied by a decrease in anxiety levels. Coaches in the EG showed a significantly greater increase in positive self-perception.	Video feedback, particularly during the game, including timeouts and halftime breaks, was found to be effective in enhancing coaching effectiveness. Differences were found in athlete anxiety levels and coach perception scores between athletes playing in different positions.
90	Smith et al. (2024)	12.40 years (SD = 0.27); 17.90 years (0.77)	Coach: N = 4; Athlete: N = 54	Athletes in the EG showed a significantly greater increase in positive perceptions of their coaches and coach-athlete relationship. Coaches in the EG showed a greater increase in targeted behaviors, along with a positive self-perception of their coaching effectiveness.	Evaluating the Irish Rugby Football Union Coach Education Framework, which integrates SDT, ELT, and ILT.
22	Smith, Smoll, & Cumming (2007)	11–14 years (M = 11.5, SD = 1.63)	Coach: N = 37 (EG = 20, CG = 17; 4F, 33 M); Athlete: N = 216 (EG = 147, CG = 69; 99F, 117 M)	Athletes in the EG perceived their coaches as being more mastery-involving, and exhibited decreases on somatic anxiety, worry, and concentration disruption.	MLM was applied in a coach education interventional study. The findings strengthen the theoretically predicted link between a mastery-initiating motivational climate and reduced anxiety; a relationship previously demonstrated only in correlational research.
23	Smoll, Smith, & Cumming (2007a)	11–14 years (M = 11.5, SD = 1.63)	Coach: N = 37 (EG = 20, CG = 17; 4F, 33 M); Athlete: N = 225 (EG = 155, CG = 70; 99F, 126 M)	Athletes in the EG exhibited higher scores in perceived mastery climate and lower scores in ego climate. Additionally, there were significant increases in mastery goal orientation scores and significant decreases in ego-orientation scores observed across the season.	The workshop was directed not only at producing changes in motivational climate but also in influencing changes in athletes’ achievement goal orientations. A significant group difference (EG > CG) was found on the corresponding mastery score of the academic achievement goal scale in the post-season.
24	Smoll, Smith, & Cumming (2007b)	9–15 years (M = 11.6, SD = 1.73)	Coach: N = 34 (3F, 31 M); Athlete: N = 151 (EG = 82, CG = 69)	Athletes in the EG decreased in sport performance trait anxiety (total anxiety score, somatic anxiety, worry, and concentration disruption), whereas the CG showed the opposite.	Utilizing a compact educational workshop yielded effects comparable to those of season-long comprehensive training in terms of athlete performance and trait anxiety.
91	Sullivan (2005)	Grade 6	Coach N = 14 (EG = 7, CG = 7; all M); Athlete N = 125 (63F, 62 M)	No significant differences were found in the EG regarding athletes’ motivation, intentions to play following season, and satisfaction between youth athletes and their coaches.	Although the coaches gained learning experience, these might not have constituted transformative learning.
47	Vella et al. (2021)	12–17 years	Coach: N = 8; Athlete: N = 350	Athletes showed significant improvements in depression, anxiety literacy, formal help-seeking intentions, mental health confidence, resilience, and overall well-being, with no changes in stigma or informal support perceptions.	The intervention involved coaches, caregivers, and youth athletes, providing ecologically thorough evidence on youth athletes’ mental health literacy and resilience. • Provide evidence that a mental health intervention in organized sport could be effective for a sample of nonelite, adolescent male sport participants.
49	Vella, Oades, & Crowe (2013)	12–18 years	Coach: N = 18 (all males; EG = 9, CG = 9); Athlete: N = 280 (EG = 133; CG = 147)	Athletes experienced increased perception of coaches’ transformational leadership behaviors (appropriate role modeling, fostering acceptance of group goals, and intellectual stimulation). And positive developmental experiences (personal and social skills, cognitive skills, goal setting, and initiative) in the EG across time.	The transformational leadership training program showed positive impact on athletes’ developmental experiences as well as their perceived coaches’ transformational leadership behaviors/skills through CET/MAC approach.
96	Wilczyńska et al. (2021)	9–12 years	Coach: N = 8 (EG = 3, CG = 5); Athlete: N = 57 (EG = 29, CG = 28)	Substantial effects of the workshop on athletes were observed, enhancing motivation, reducing state anxiety, improving self-reflection, and decreasing reaction time.	Investigating the effect of coach education interventions on athletes’ well-being using the PERMA framework.
97	Wilczyńska et al. (2022)	9–12 years	Coach: N = 8 (EG = 3, CG = 5); Athlete: N = 57 (EG = 29, CG = 28)	In the Experimental Group, athletes showed less decline in intrinsic motivation. Closeness in athlete-coach relationships moderated reaction time changes, relational commitment influenced identified motivation, and competitive trait anxiety predicted changes in integrated and intrinsic motivation.	i7 W showed a protective effect on the intrinsic motivation of athletes.

Note. In total, 27 studies were included. CET = Coach Effectiveness Training; CG = Control group; ELT = Explicit Learning Theories; EG = Experimental group; F = Female; FDG = Feedback group; MAC = Mastery Approach to Coaching; M = Male; MLM = Multilevel modeling; PERMA = Positive emotions, engagement, relations, meaning, and accomplishments; PYD = Positive youth development; SDT = Self-determination theory; ILT = Implicit Learning Theories

Meta-analyses results

Athlete-level outcomes from 27 studies (total effect size k = 193) were extracted and categorized according to data preparation. Of these, 132 effect sizes reflected individual-level outcomes, including cognitive (k = 13), social (k = 24), emotion (k = 26), identity (k = 13), and values (k = 56), and 61 reflected micro level outcomes. The multilevel meta-analysis revealed significant heterogeneity in effect sizes (Q = 629.36, df = 192, p < .001). The variance components indicated moderate heterogeneity both between studies (σ² = 0.03, τ = 0.17) and within studies (σ² = 0.03, τ = 0.18). Total heterogeneity I² is 61.20% indicates that approximately 61% of observed variance reflected true differences in the effect sizes rather than sampling error. The heterogeneity was relatively evenly distributed between studies (I² = 29.44%) and within studies (I² = 31.76%), supporting the use of a multilevel meta-analytic approach. Egger's regression test indicated potential small-study effects, with a marginally significant intercept (0.64, p = .074) and a significant slope (0.16, p = .004), suggesting possible overrepresentation of smaller studies with larger effects (Supplementary Figure S2).

Model comparisons favored the three-level specification, supported by lower Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) values and by the likelihood ratio test (χ² = 38.29, p < .001). The frequentist hierarchical meta-analysis identified a significant small to moderate overall effect of CEPIs on athlete-level outcomes (SMD = 0.27, 95% CI [0.19, 0.34]). Significant small to moderate effects were also observed across individual-level outcomes overall (SMD = 0.30, 95% CI [0.19, 0.40]), including cognitive (SMD = 0.46, 95% CI [0.26, 0.66]), emotion (SMD = 0.34, 95% CI [0.25, 0.43]), social (SMD = 0.27, 95% CI [0.08, 0.46]), values (SMD = 0.35, 95% CI [0.16, 0.54]), and identity (SMD = 0.43, 95% CI [0.15, 0.71]). Micro level outcomes also showed a significant small to moderate effect (SMD = 0.19, 95% CI [0.11, 0.27]). Full results are presented in Table 3.

Table 3.

Frequentist hierarchical meta-analysis results.

Athlete level outcomes	k	Effect size (SE)	95% CI	t	p
Overall outcomes	193	0.27 (0.04)	[0.19, 0.34]	6.85	< .001
Individual level outcomes	132	0.30 (0.05)	[0.19, 0.40]	5.74	< .001
Cognitive	13	0.46 (0.09)	[0.26, 0.66]	4.92	< .001
Emotion	26	0.34 (0.04)	[0.25, 0.43]	7.86	< .001
Social	24	0.27 (0.09)	[0.08, 0.46]	2.94	.007
Values	56	0.35 (0.09)	[0.16, 0.54]	3.69	< .001
Identity	13	0.43 (0.13)	[0.15, 0.71]	3.34	.006
Microsystem level outcomes	61	0.19 (0.04)	[0.11, 0.27]	4.55	< .001

Notes. In total, 27 studies with 193 effect sizes were included; CI = Confidence Interval; SE = Standard Error

Sensitivity analyses using the three-parameter selection model indicated that all domains, except values, and micro level outcomes were not significantly biased by underrepresentation of non-significant findings (Table 4). Publication bias was detected in values related outcomes at the individual level. Adjusted estimates increased from 0.35 to 0.56 at single cut-point α = .025 while retaining statistical significance, suggesting potential bias from selective reporting. To incorporate prior evidence and estimate posterior distributions of CEPI effects, the frequentist analysis was complemented with a Bayesian approach.⁷⁴

Table 4.

Results adjusted for publication bias.

	Adjusted for publication bias					Likelihood ratio test
Single cut-point (α = .025)	k	Effect size (SE)	95% CI	z	p	χ ²	p
Overall outcomes	193	0.33 (0.04)	[0.26, 0.40]	8.90	< .001	4.29	.038
Individual level outcomes	132	0.37 (0.05)	[0.28, 0.46]	7.97	< .001	3.16	.076
Cognitive	13	0.49 (0.13)	[0.24, 0.75]	3.78	< .001	0.12	.729
Emotion	26	0.32 (0.06)	[0.20, 0.44]	5.36	< .001	0.27	.602
Social	24	0.09 (0.09)	[-0.08, 0.26]	1.01	.314	0.98	.323
Values	56	0.56 (0.09)	[0.38, 0.75]	5.93	< .001	17.42	< .001
Identity	13	0.21 (0.22)	[-0.22, 0.64]	0.96	.337	2.00	.157
Microsystem level outcomes	61	0.22 (0.05)	[0.11, 0.32]	4.12	< .001	0.26	.609
Single cut-point (α = .050)
Overall outcomes	193	0.28 (0.04)	[0.21, 0.35]	7.78	< .001	0.03	.865
Individual level outcomes	132	0.34 (0.05)	[0.25, 0.43]	7.48	< .001	0.72	.395
Cognitive	13	0.47 (0.13)	[0.21, 0.72]	3.59	< .001	0.01	.930
Emotion	26	0.35 (0.06)	[0.24, 0.46]	6.12	< .001	0.06	.808
Social	24	0.06 (0.09)	[-0.12, 0.24]	0.65	.516	1.91	.167
Values	56	0.42 (0.07)	[0.27, 0.57]	5.65	< .001	4.94	.026
Identity	13	0.31 (0.21)	[-0.10, 0.72]	1.49	.136	0.62	.429
Microsystem level outcomes	61	0.14 (0.05)	[0.04, 0.23]	2.88	.004	2.20	.138

Note. In total, 27 studies with 193 effect sizes were included; CI = Confidence Interval; SE = Standard Error

In the Bayesian meta-analysis, hierarchical model estimates for marginal posterior distributions of μ and τ are presented in Table 5. All $\hat{R}$ values were below 1.01, indicating convergence in both overall and subgroup models. Posterior predictive checks confirmed close alignment between the replicated and observed data distributions. Compared with the frequentist results, the estimated overall effect size was slightly smaller (SMD = 0.26, 95% CrI [0.18, 0.34]). The estimated between-study heterogeneity was consistent with prior assumptions (τ = 0.19, 95% CrI [0.14, 0.26]). For both parameters, the 95% CrIs excluded zero, indicating a statistically meaningful effect and the presence of between-study variability. The model explained 33% of the variance in athlete-level outcomes (R² = 0.33, SE = 0.03, 95% CrI [0.27, 0.39]). Graphical analyses showed that the overall effect and the marginal posterior distributions of μ and τ approximated normal curves (Figure 2).

Figure 2.

Bayesian summary plots of overall effect. Note. The left-side plot is the marginal posterior distribution of overall mean effect size. The right-side plot is the marginal posterior distribution of between-studies standard deviation.

Table 5.

Bayesian hierarchical meta-analysis results.

Quantity	k	Mode	Median	Mean	SD	95% CrI	ESS	$\hat{R}$	95% HPDI
Between-studies Standard Deviation (τ)
Overall outcomes	193	0.18	0.18	0.19	0.03	[0.14, 0.26]	2722	1.00	[0.13, 0.25]
Individual level outcomes	132	0.24	0.24	0.25	0.04	[0.18, 0.34]	2952	1.00	[0.17, 0.33]
Cognitive	13	0.28	0.30	0.32	0.11	[0.16, 0.59]	3090	1.00	[0.13, 0.54]
Emotion	26	0.06	0.08	0.09	0.06	[0.01, 0.24]	2477	1.00	[0.00, 0.21]
Social	24	0.27	0.29	0.31	0.09	[0.18, 0.52]	2696	1.00	[0.16, 0.48]
Values	56	0.32	0.34	0.35	0.09	[0.21, 0.57]	2530	1.00	[0.19, 0.53]
Identity	13	0.30	0.33	0.35	0.11	[0.19, 0.63]	2829	1.00	[0.17, 0.58]
Microsystem level outcomes	61	0.14	0.15	0.15	0.04	[0.08, 0.24]	3906	1.00	[0.08, 0.23]
Overall Mean (µ)
Overall outcomes	193	0.26	0.26	0.26	0.04	[0.18, 0.34]	1515	1.00	[0.19, 0.35]
Individual level outcomes	132	0.30	0.29	0.29	0.05	[0.19, 0.40]	2127	1.00	[0.19, 0.40]
Cognitive	13	0.47	0.47	0.47	0.12	[0.22, 0.73]	2679	1.00	[0.24, 0.74]
Emotion	26	0.33	0.34	0.34	0.05	[0.24, 0.45]	2205	1.00	[0.24, 0.44]
Social	24	0.29	0.27	0.28	0.10	[0.08, 0.49]	2695	1.00	[0.09, 0.50]
Values	56	0.36	0.35	0.35	0.10	[0.15, 0.55]	2054	1.00	[0.14, 0.55]
Identity	13	0.43	0.42	0.42	0.14	[0.14, 0.69]	2046	1.00	[0.14, 0.69]
Microsystem level outcomes	61	0.18	0.19	0.19	0.04	[0.10, 0.28]	4662	1.00	[0.11, 0.28]

Note. CrI = Credible interval; HPDI = Highest Posterior Density Interval; SD = Standard Deviation; $\hat{R}$ = $\hat{R}$ convergence; ESS = Effective sample size

The marginal posterior distributions for individual and micro level outcomes were broadly similar, each approximating normal curve (Figure 3). Central tendency measures were closely aligned, indicating consistent location estimates across distributions. For τ, the cognitive, emotional, and identity domains showed slight right skew, suggesting greater heterogeneity in the upper bounds of between-study variance (Figure 4). For μ, moderate to large effects were found for cognitive (SMD = 0.47, 95% CrI [0.22, 0.73]) with an R² of 0.55 (SE = 0.13, 95% CrI [0.25, 0.74]), and identity (SMD = 0.42, 95% CrI [0.14, 0.69]) with an R² of 0.58 (SE = 0.10, 95% CrI [0.35, 0.73]). Small to moderate effects were observed for emotion (SMD = 0.34, 95% CrI [0.24, 0.45]), with an R² of 0.12 (SE = 0.10, 95% CrI [0.00, 0.35]), values (SMD = 0.35, 95% CrI [0.15, 0.55]), with an R² of 0.59 (SE = 0.07, 95% CrI [0.43, 0.71]), and social (SMD = 0.28, 95% CrI [0.08, 0.49]), with an R² of 0.71 (SE = 0.09, 95% CrI [0.50, 0.83]).

Figure 3.

Bayesian summary plots of system level effect. Note. The left-side plots are the marginal posterior distribution of mean effect sizes with respect to each individual and micro level outcomes. The right-side plots are the marginal posterior distribution of between-studies standard deviations.

Figure 4.

Bayesian summary plots of individual levels. Note. The left-side plots are the marginal posterior distribution of mean effect sizes with respect to each individual sublevel outcomes. The right-side plots are the marginal posterior distribution of between-studies standard deviations.

Overall and individual-level outcomes were of similar magnitude across both approaches, whereas micro level effects were smaller in the Bayesian analysis compared with the frequentist results. At the domain level, the Bayesian approach produced larger effects for cognitive, values, and identity domains, but smaller effects for emotion and social outcomes. The results provided strong evidence for CEPI effectiveness. Bayesian posterior probability analysis indicated a 100% probability that the overall effect is positive, with all posterior samples exceeding zero. This quantifies the certainty of intervention benefits based on observed data and prior information, offering definitive statistical evidence for the positive direction of CEPI effects on athlete-level outcomes.

Sensitivity analyses showed consistent posterior distributions across all prior specifications. The student-t prior model performed marginally best (ΔELPD = 0), with the weakly informative and literature-informed models performing similarly (both with ΔELPD = -0.1, SE = 0.2). Even skeptical priors centered at zero performed adequately (ΔELPD = -0.5, SE = 0.3), indicating sufficient data strength to offset prior skepticism. The conservative heterogeneity model performed weakest (ΔELPD = -0.8, SE = 0.2), suggesting the data support moderate between-study variability. All ΔELPD differences were small relative to their standard errors (all SE ≥ 0.2), and posterior estimates were consistent across models. Posterior distributions for the sensitivity analyses are shown in Supplementary Figures S3–4.

Subgroup analyses were conducted to examine potential moderation effects regarding coaching experience, athlete age, and CEPIs delivery format. Coaching experience was a significant moderator on CEPI effects (F _{(3, 190)} = 17.44, p < .001), with the largest effects for early-career coaches (5 years or less coaching experiences), followed by those with more than five years of experience and those with unreported experience. Athlete age was also a significant moderator (F _{(3, 190)} = 16.66, p < .001), with mixed-age groups showing the largest effects, followed by youth athletes aged 12–18 and those younger than 12. Besides, CEPIs delivery format was a significant moderator (F _{(3, 190)} = 14.77, p < .001). Both in-person and hybrid formats showed significant positive effects, while the online-only format, represented by only three effect sizes, showed the largest estimated effect but did not reach statistical significance (p = .087). Full results are presented in Tables 6 –8 and Figures 5–7.

Figure 5.

Subgroups analysis of athlete-level outcomes by coaching experiences.

Figure 6.

Subgroups analysis of athlete-level outcomes by age of athlete.

Figure 7.

Subgroups analysis of athlete-level outcomes by CEPI delivery format.

Table 6.

Subgroup analysis on coaching experience.

	Group level effect					Group comparison
Athlete level outcomes	k	Effect size (SE)	95% CI	t	p	F	p
Overall outcomes						17.44	< .001
Veteran coach	100	0.26 (0.05)	[0.16, 0.37]	5.14	< .001
Early career coach	30	0.40 (0.09)	[0.21, 0.58]	4.18	< .001
Not reported	63	0.20 (0.07)	[0.06, 0.34]	2.90	.004
Individual level outcomes						11.77	< .001
Veteran coach	62	0.33 (0.07)	[0.19, 0.47]	4.63	< .001
Early career coach	27	0.38 (0.12)	[0.14, 0.62]	3.14	.002
Not reported	43	0.19 (0.09)	[0.00, 0.37]	2.00	.047
Microsystem level outcomes						6.42	< .001
Veteran coach	38	0.21 (0.06)	[0.10, 0.32]	3.72	< .001
Early career coach	3	0.27 (0.19)	[-0.10, 0.65]	1.47	.147
Not reported	20	0.14 (0.08)	[-0.01, 0.29]	1.81	.075

Notes. In total, 27 studies with 193 effect sizes were included; CI = Confidence Interval; SE = Standard Error; Veteran coach = Coaching experience equal or greater than five years; Early career coach = Coaching Experience less than five years; Not reported = Coaching experience was not reported in the study

Table 7.

Subgroup analysis on age of athlete.

	Group level effect					Group comparison
Athlete level outcomes	k	Effect size (SE)	95% CI	t	p	F	p
Overall outcomes						16.66	< .001
12- to 18-year-old	73	0.23 (0.06)	[0.10, 0.35]	3.67	< .001
Younger than 12-year-old	39	0.20 (0.09)	[0.02, 0.38]	2.22	.028
Mixed	81	0.33 (0.06)	[0.21, 0.45]	5.62	< .001
Individual level outcomes						10.57	< .001
12- to 18-year-old	34	0.29 (0.09)	[0.11, 0.47]	3.21	.002
Younger than 12-year-old	31	0.22 (0.12)	[-0.02, 0.46]	3.14	.074
Mixed	67	0.33 (0.08)	[0.18, 0.48]	4.26	< .001
Microsystem level outcomes						7.15	< .001
12- to 18-year-old	39	0.17 (0.05)	[0.06, 0.28]	3.09	.003
Younger than 12-year-old	8	0.12 (0.11)	[-0.10, 0.34]	1.08	.283
Mixed	14	0.27 (0.08)	[0.00, 0.43]	3.27	.002

Notes. In total, 27 studies with 193 effect sizes were included; CI = Confidence Interval; SE = Standard Error

Table 8.

Subgroup analysis on type of training.

	Group level effect					Group comparison
Athlete level outcomes	k	Effect size (SE)	95% CI	t	p	F	p
Overall outcomes						14.77	< .001
Hybrid	43	0.27 (0.09)	[0.10, 0.44]	3.14	.002
In-person	143	0.26 (0.05)	[0.17, 0.35]	5.61	< .001
Online	3	0.39 (0.23)	[-0.06, 0.83]	1.72	.087
Individual level outcomes						10.12	< .001
Hybrid	40	0.28 (0.11)	[0.06, 0.51]	2.47	.015
In-person	90	0.30 (0.06)	[0.18, 0.42]	4.79	< .001
Online	2	0.33 (0.29)	[-0.24, 0.90]	1.14	.255
Microsystem level outcomes						8.25	< .001
Hybrid	7	0.08 (0.10)	[-0.12, 0.27]	0.79	.435
In-person	53	0.20 (0.05)	[0.11, 0.29]	4.45	< .001
Online	1	0.51 (0.24)	[0.02, 0.99]	2.09	.041

Notes. In total, 27 studies with 193 effect sizes were included; CI = Confidence Interval; SE = Standard Error; Type of training refers to the delivery format of the CEPIs; Hybrid = Training(s) was conducted through a combination of both in-person and online sessions

Discussion

Youth athlete outcomes from 27 studies were contextualized at individual and micro levels and further categorized into domains using the EASEL taxonomy. Results from both frequentist and Bayesian meta-analyses indicated positive impacts of CEPIs on athlete development, consistent with previous reviews.^6,15 By mapping athletes’ non-academic outcomes, this review extends prior work by bridging coaching and youth sport research with education and developmental psychology, offering a new lens on coaching and sport-based youth development while informing future research. At the same time, the findings revealed improvements alongside differences across domains, with changes observed in athlete beliefs, skills, and competencies at the individual level and in team climate and coach–athlete interactions at the micro level.

Overall, CEPIs demonstrated small to moderate effects in catalyzing sport-based youth development. At the individual level, athletes with trained coaches showed significant gains in cognitive, social, and emotional skills and competencies, including cognitive flexibility, goal setting, emotion and behavior regulation, and prosocial behaviors. These increases can be understood through an indirect pathway in which CEPIs support coach learning and inform coaching practices that shape athletes’ sport experiences.^51,98 Specifically, CEPIs may contribute to changes in coaches’ knowledge, beliefs, or instructional intentions, which are then reflected in practices such as need-supportive behaviors and mastery-oriented motivational climates. Through these practices, coaches are better prepared to support athletes’ developmental outcomes by fostering more engaging and developmentally appropriate sport experiences.^6,32 In the same vein, CEPIs also showed positive effects on athlete identity and values, beliefs and attitudes that guide when and how social, cognitive, and emotional skills are applied, which in turn foster self-determination, self-knowledge, self-efficacy, growth mindset, and self-esteem. These outcomes are linked to psychologically safe sport environments and lower levels of athlete anxiety and burnout.⁹⁷ Consistent with a prior meta-analysis by Li and colleagues,¹⁵ which also included performance-related outcomes, athlete-level outcomes yielded effect sizes comparable to those of coaches trained in CEPIs.

Increased self-efficacy and positive emotions can further enhance self-determined motivation, which in turn nurtures the development of self-management skills such as initiative and goal setting.^99,100 CEPIs showed positive effects on athlete self-management outcomes. These effects may be attributed to training curricula grounded in SDT and AGT, which help coaches become more aware of and skilled in fulfilling athletes’ basic psychological needs, thereby motivating athletes to be more self-determined.⁹² Sport participation, when supported by effective coaching practices, may be associated with stronger social connections, greater team inclusion, and higher levels of perceived support among athletes.⁴⁹ At the same time, it can direct athletes’ focus on skill development and valuing effort within autonomy-supportive and task-oriented environments.^93,101 Within such contexts, athletes have ample opportunities to develop self-discipline, intrinsic motivation, and skills transferable across sport and life domains, including time management, coping, academic orientation, and mental toughness.^3,6,102

Notably, no athlete outcomes related to the perspectives domain, defined as how youth view and approach the world, were identified in the current review. Perspective-related outcomes, such as optimism (e.g., approaching challenges with a positive attitude), gratitude (e.g., expressing appreciation), and openness (e.g., adapting willingly to change), are likely more distal and developmentally cumulative than proximal outcomes such as cognitive engagement and perceived motivational climate. Several of those constructs are closely tied to personality-related characteristics that show meaningful variability and change from childhood to adolescence,^103,104 as well as to anxiety and mental health.¹⁰⁵ Positive perspectives may promote emotion regulation, resilience, and creativity in the face of adversity, with implications for both sport performance and daily life. In addition, emerging evidence links positive perspectives and related psychological capacities to neural plasticity and functional brain connectivity in youth.^105,106 Given the critical developmental window, future CEPI designs and research could explore bidirectional body–brain pathways linking evidence-based coaching practice, brain health, and youth well-being.¹⁰⁷ It is also important to note that such outcomes may take time to emerge and become observable, and that consistent coach support over time is critical.¹⁰⁸

Relatedly, the pooled effect size at the individual level was about twice as large as at the micro level, which included outcomes such as athletes’ perceptions of coach attitudes, coaching behaviors, and team environment. This suggests that CEPIs may be more effective in fostering individual level changes, such as supporting autonomy by providing opportunities for learning and goal attainment. In contrast, building team culture or shifting the broader climate at the micro level may require more time than the duration of many interventions (e.g., four weeks or a single season) before becoming observable or effective. These effects can also be time sensitive. It is reasonable to expect that CEPIs exert rapid effects at the individual level, whereas some micro level outcomes are likely to emerge only later and become observable post-intervention. Moreover, micro level changes may be delayed, as building team culture requires time and may also depend on the length of a coach's tenure with a team and are also shaped by the broader ecological systems in which youth sport participation occurs.⁵⁹ Therefore, future studies would benefit from employing longitudinal designs that track outcomes and confounding factors chronologically and explicitly consider time as a facet.^58,109,110

Although the selected studies primarily focused on athlete outcomes, such outcomes are embedded within broader ecological contexts, including coach–athlete relationships, coach–parent relationships, family involvement, and organizational or community settings, which were less frequently examined. These factors may shape how coach education is implemented and how athlete outcomes emerge and evolve. Dorsch and colleagues⁵⁹ emphasized that youth sport should be understood as a system of interdependent persons and contexts that influence and are influenced by athletes’ behaviors, attitudes, experiences, and outcomes. As such, coaching practices and effectiveness depend on interactions within family and team subsystems and is embedded within organizations, communities, and societies.⁵⁹ Within these systems, controlling or toxic coaching behaviors, win-at-all-costs cultures, and adult-driven priorities may undermine the educative goals of youth sport participation, and CEPIs, by weakening mastery-oriented climates, sacrificing fun and safety, and increasing risks for injury, burnout, and dropout.^26,111–117 Taken together, these ecological factors underscore that the effects of CEPIs depend not only on what coaches learn, but also on how coaching practices are enacted within relationships, organizations, and broader youth sport cultures.

Furthermore, CEPIs are not a one-size-fits-all approach, as effects vary across studies. CEPIs could be more effective when delivered not as isolated single-session workshops but as ongoing conversations and interactions with stakeholders in schools and communities.^118,119 Coaching and coach education are contextual.^17,59 For example, in the United States, youth sport coaching is largely volunteer based.¹²⁰ Although coaches value education that emphasizes holistic athlete development, engagement is often driven by requirements and access to relevant content, with time and cost identified as common barriers.¹²¹ Therefore, continued support for coach professional development requires training that delivers relevant learning opportunities while aligning with the contexts and needs of coaches. Situating coach education within its broader ecological context, strengthening alignment among coach learning, coaching practice, and athlete outcomes, and ensuring ongoing validation of how those outcomes are measured and interpreted are essential for conceptualizing future CEPI design.

Limitations and future direction

Several limitations should be acknowledged. First, as noted by Van Aert and colleagues,¹²² there is currently no method that provides robust publication-bias results when between-study heterogeneity is high. In such cases, publication-bias analyses should be avoided. In this review, heterogeneity was high, so analyses were conducted for each domain in more homogeneous groups. Even so, imbalanced group sizes remain a concern, particularly the limited number of effect sizes for identity, which may not adequately represent that domain. Accordingly, pooled effects varied across studies and should be interpreted with caution. In addition, many CEPI outcomes are latent traits measured with psychometric instruments. Validity evidence needed to support proposed score interpretations and uses was seldom reported across the selected studies and was not examined systematically here. Finally, the use of the EASEL taxonomy is theoretically grounded but has not yet been widely examined as a standalone framework in youth sport contexts, and its application would benefit from further empirical validation. In addition, the current review reflects a limited representation of ecological systems, as no outcomes were reported at the organizational, community, and societal levels; accordingly, the findings should be interpreted with caution when generalizing to broader ecological contexts within the youth sport system.

Future work could strengthen youth sport coach education research by using survival analyses to examine coach and athlete retention, conducting longitudinal designs to measure long-term coach efficacy and athlete outcomes, and employing mixed-methods and participatory approaches led by youth athletes that center their voices. Moreover, sport-based youth development and evidence-informed CEPIs align closely with prevention science.⁶¹ Identifying risk and protective factors and developing evidence-informed strategies may help prevent health and social problems while supporting holistic development. These benefits can extend across ecological systems by fostering individual growth and creating positive effects for families, schools, and communities. Longitudinal evidence indicates that EASEL-aligned competencies measured in kindergarten are associated with young-adult outcomes in education, employment, criminal activity, substance use, and mental health.¹²³ As a favored activity among children for decades, sport presents opportunities to advance public health and the public good by understanding and scaling conditions that support sustained physical activity, mental well-being, and prosocial development.

Despite the recognized importance of youth sport, research in this area remains underfunded and under-resourced.^124,125 Higher education institutions are well positioned to address this gap, given their infrastructure, community partnerships, and role in supporting transition-aged youth.¹²⁶ Emerging evidence suggests that college physical education programs, particularly those using the Sport Education Model, can promote physical activity and buffer mental health challenges,^127,128 highlighting the need to further explore the transition from high school sport to college-based physical activity. Moreover, sport serves as a powerful connector across ecological systems, offering natural social spaces for interaction and development. Coaches play a pivotal role in shaping these environments by fostering social awareness and positive relationships.⁵² Integrating athlete voice in team decision-making and engaging youth in discussions around social issues, ethics, and diverse perspectives can cultivate compassion, empathy, and social responsibility.^37,129,130 These opportunities are increasingly vital in a rapidly evolving society shaped by technological change and population health challenges^131,132 and align with human-centered goals to build healthy, resilient communities.^133,134

Conclusion

Meta-analytic findings indicate that coach education is linked to holistic youth development when viewed through an ecological lens. Investing in coach education supports coaches’ professional learning and may help create developmental climates in which young people build social, cognitive, and emotional skills, as well as the agency to apply them. Future research should examine long-term impacts using rigorous and innovative designs, such as longitudinal and mixed-methods studies, and specify clear interpretive arguments that connect training goals to observable outcomes.

Supplemental Material

sj-png-1-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-png-1-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-png-2-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-png-2-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-png-3-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-png-3-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-png-4-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-png-4-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-docx-5-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-docx-5-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-docx-6-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-docx-6-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-docx-7-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-docx-7-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Supplemental Material

sj-docx-8-spo-10.1177_17479541261461871 - Supplemental material for An ecological meta-analysis of coach education for sport-based youth development

Supplemental material, sj-docx-8-spo-10.1177_17479541261461871 for An ecological meta-analysis of coach education for sport-based youth development by Longxi Li, Irina Tereschenko and Julie McCleery in International Journal of Sports Science & Coaching

Footnotes

ORCID iDs

Longxi Li

Irina Tereschenko

Julie McCleery

Ethical considerations

The review protocol was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO), an online database for systematic review protocols (CRD42023479392).

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

Data will be made available on request.

Supplemental material

Supplemental material for this article is available online.

References

Fleary

Joseph

Pappagianopoulos

. Adolescent health literacy and health behaviors: a systematic review. J Adolesc 2018; 62: 116–127.

Siedentop

Van der Mars

. Introduction to physical education, fitness, and sport. Champaign: Human Kinetics, 2022.

Atkinson

Bates

Anderson-Butcher

, et al. Mapping school-based coach education requirements in the United States. Int Sport Coach J 2023; 10: 276–288.

Kellstedt

Schenkelberg

Von Seggern

, et al. Youth sport participation and physical activity in rural communities. Arch Public Health 2021; 79: 46.

Eather

Miller

Jones

, et al. Evaluating the impact of a coach development intervention for improving coaching practices and player outcomes in netball: the MASTER coaching randomized control trial. Int J Sports Sci Coach 2021; 16: 439–455.

Bengtsson

Stenling

Nygren

, et al. The effects of interpersonal development programmes with sport coaches and parents on youth athlete outcomes: a systematic review and meta-analysis. Psychol Sport Exerc 2024; 70: 102558.

Camiré

. Youth development in North American high school sport: review and recommendations. Quest 2014; 66: 495–511.

Fawver

Beatty

Roman

, et al. The status of youth coach training in the United States: existing programs and room for improvement. Int Sport Coach J 2020; 7: 239–251.

Allan

Côté

. A cross-sectional analysis of coaches’ observed emotion-behavior profiles and adolescent athletes’ self-reported developmental outcomes. J Appl Sport Psychol 2016; 28: 321–337.

10.

Mountjoy

Rhind

Tiivas

, et al. Safeguarding the child athlete in sport: a review, a framework and recommendations for the IOC youth athlete development model. Br J Sports Med 2015; 49: 883–886.

11.

Baghurst

Benham

. Quality sport coaching in action: the application of the national standards for sport coaches in the interscholastic sport context. Strategies 2020; 33: 21–27.

12.

Fraser-Thomas

Côté

. Understanding adolescents’ positive and negative developmental experiences in sport. Sport Psychologist 2009; 23: 3–23.

13.

McCleery

Burton

. How parents and coaches can support positive development. In: Christino

Pluhar

Micheli

(eds) Psychological considerations in the young athlete. Cham: Springer, 2023, pp.193–213.

14.

Côté

Hummel

Strachan

, et al. Coach effectiveness in youth sport. In: Davis

Keegan

Jowett

(eds) Social psychology in sport. 2nd ed. Champaign: Human Kinetics, 2024, pp.263–273.

15.

Olson

Tereschenko

, et al. Impact of coach education on coaching effectiveness in youth sport: a systematic review and meta-analysis. Int J Sports Sci Coach 2025; 20: 340–356.

16.

Jowett

. Coaching effectiveness: the coach–athlete relationship at its heart. Curr Opin Psychol 2017; 16: 154–158.

17.

Côté

Gilbert

. An integrative definition of coaching effectiveness and expertise. Int J Sports Sci Coach 2009; 4: 307–323.

18.

McCleery

Hoffman

Tereschenko

, et al. Ambitious coaching core practices: borrowing from teacher education to inform coach development pedagogy. Int Sport Coach J 2021; 9: 62–73.

19.

Ryan

Deci

. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am Psychol 2000; 55: 68.

20.

Dweck

Leggett

. A social-cognitive approach to motivation and personality. Psychol Rev 1988; 95: 256–273.

21.

Lerner

Brown

Kier

. Adolescence: development, diversity, context, and application. Canadian ed. Toronto: Pearson, 2005.

22.

Smith

Smoll

Cumming

. Effects of a motivational climate intervention for coaches on young athletes’ sport performance anxiety. J Sport Exerc Psychol 2007; 29: 39–59.

23.

Smoll

Smith

Cumming

. Effects of a motivational climate intervention for coaches on changes in young athletes’ achievement goal orientations. J Clin Sport Psychol 2007; 1: 23–46.

24.

Smoll

Smith

Cumming

. Effects of coach and parent training on performance anxiety in young athletes: a systemic approach. J Youth Dev 2007; 2: 19–36.

25.

Jones

Eather

Miller

, et al. Evaluating the impact of a coach development intervention for improving coaching practices and player outcomes in football: the MASTER Coaching randomised control trial. Phys Educ Sport Pedagog 2023; 29: 652–669.

26.

Mahoney

Ntoumanis

Gucciardi

, et al. Implementing an autonomy-supportive intervention to develop mental toughness in adolescent rowers. J Appl Sport Psychol 2016; 28: 199–215.

27.

Camiré

Trudel

Forneris

. Examining how model youth sport coaches learn to facilitate positive youth development. Phys Educ Sport Pedagog 2014; 19: 1–17.

28.

Newman

Santos

Pierce

, et al. Scholars’ perspectives of positive youth development in coach education for high school sports. Phys Educ Sport Pedagog 2024; 29: 206–220.

29.

Newman

Ortega

Lower

, et al. Informing priorities for coaching education: perspectives from youth sport leaders. Int J Sports Sci Coach 2016; 11: 436–445.

30.

Camiré

Kendellen

Rathwell

, et al. Evaluating the Coaching for Life Skills online training program: a randomised controlled trial. Psychol Sport Exerc 2020; 48: 101649.

31.

Holt

Neely

Slater

, et al. A grounded theory of positive youth development through sport based on results from a qualitative meta-study. Int Rev Sport Exerc Psychol 2017; 10: 1–49.

32.

MacDonald

Camiré

Erickson

, et al. Positive youth development related athlete experiences and coach behaviors following a targeted coach education course. Int J Sports Sci Coach 2020; 15: 621–630.

33.

Santos

Camiré

Macdonald

, et al. Process and outcome evaluation of a Positive Youth Development-focused online coach education course. Int Sport Coach J 2019; 6: 1–12.

34.

Gould

Carson

. Life skills development through sport: current status and future directions. Int Rev Sport Exerc Psychol 2008; 1: 58–78.

35.

Bean

Forneris

. Is life skill development a by-product of sport participation: perceptions of youth sport coaches. J Appl Sport Psychol 2017; 29: 234–250.

36.

Matthews

Booth

Rooney

, et al. A pilot study assessing the preliminary efficacy and acceptability of a mental health promotion e-learning module for sports coaches. Child Health Care 2023; 53: 299–321.

37.

Power

Seroczynski

. Promoting character development through coach education. J Character Educ 2015; 11: 87–107.

38.

Falcão

Bloom

Bennie

. Coaches’ experiences learning and applying the content of a humanistic coaching workshop in youth sport settings. Int Sport Coach J 2017; 4: 279–290.

39.

Boudreaux

Dolecheck

. Assessment of high school coaches’ knowledge retention of sport-related concussions. Res Issues Contemp Educ 2020; 5: 25–41.

40.

Glang

Koester

Beaver

, et al. Online training in sports concussion for youth sports coaches. Int J Sports Sci Coach 2010; 5: 1–11.

41.

Newman

Santos

Pierce

, et al. Coach education and coach development within a contemporary social justice society: implications for future research and potential pitfalls. Quest 2022; 74: 234–250.

42.

Eather

Jones

Miller

, et al. Evaluating the impact of a coach development intervention for improving coaching practices in junior football (soccer): the “MASTER” pilot study. J Sports Sci 2020; 38: 1441–1453.

43.

Larsen

Van Hoye

Tjomsland

, et al. Creating a supportive environment among youth football players: a qualitative study of French and Norwegian youth grassroots football coaches. Health Educ 2015; 115: 570–586.

44.

Van Hoye

Larsen

Sovik

, et al. Evaluation of the coaches educators training implementation of the PAPA project: a comparison between Norway and France. Scand J Med Sci Sports 2015; 25: e539–e546.

45.

Bates

Anderson-Butcher

Ute

, et al. Mental health training for high school coaches and athletic directors: community-based participatory research to coach beyond. Int J Sports Sci Coach 2023; 19: 539–550.

46.

Kroshus

Zhou

Ledsky

, et al. Randomized evaluation of Centers for Disease Control and Prevention HEADS UP concussion education materials for youth sport coaches. J Neurotrauma 2023; 40: 1584–1595.

47.

Vella

Swann

Batterham

, et al. An intervention for mental health literacy and resilience in organized sports. Med Sci Sports Exerc 2021; 53: 139–149.

48.

Griffiths

Goodyear

Armour

. Massive open online courses (MOOCs) for professional development: meeting the needs and expectations of physical education teachers and youth sport coaches. Phys Educ Sport Pedagog 2022; 27: 276–290.

49.

Vella

Oades

Crowe

. A pilot test of transformational leadership training for sports coaches: impact on the developmental experiences of adolescent athletes. Int J Sports Sci Coach 2013; 8: 513–530.

50.

Pulido

Sánchez-Oliva

Leo

, et al. Effects of an interpersonal style intervention for coaches on young soccer players’ motivational processes. J Hum Kinet 2017; 59: 107–120.

51.

Langan

Toner

Blake

, et al. Testing the effects of a Self-Determination Theory-based intervention with youth Gaelic football coaches on athlete motivation and burnout. Sport Psychologist 2015; 29: 293–301.

52.

Falcão

Bloom

Sabiston

. The impact of humanistic coach training on youth athletes’ development through sport. Int J Sports Sci Coach 2020; 15: 610–620.

53.

Blom

Watson

II Spadaro

. The impact of a coaching intervention on the coach-athlete dyad and athlete sport experience. Athl Insight J 2011; 2: 187–205.

54.

Coatsworth

Conroy

. Enhancing the self-esteem of youth swimmers through coach training: gender and age effects. Psychol Sport Exerc 2006; 7: 173–192.

55.

Bruner

McLaren

Sutcliffe

, et al. The effect of sport-based interventions on positive youth development: a systematic review and meta-analysis. Int Rev Sport Exerc Psychol 2023; 16: 368–395.

56.

Kahn

Bailey

Jones

. Coaching social & emotional skills in youth sport. Report, The Aspen Institute, March 2019.

57.

Bronfenbrenner

. Toward an experimental ecology of human development. Am Psychol 1977; 32: 513–531.

58.

Bronfenbrenner

. Making human beings human: bioecological perspectives on human development. Thousand Oaks: Sage Publications, 2005.

59.

Dorsch

Smith

Blazo

, et al. Toward an integrated understanding of the youth sport system. Res Q Exerc Sport 2022; 93: 105–119.

60.

Moosbrugger

. Correlations between physical activity participation and the environment in children and adolescents: a systematic review and meta-analysis using ecological frameworks. Int J Environ Res Public Health 2021; 18: 9080.

61.

EASEL Lab. Harvard Graduate School of Education. The Taxonomy Project, http://exploresel.gse.harvard.edu/frameworks/ (2025, accessed 4 June 2026).

62.

Jones

McGarrah

Kahn

. Social and emotional learning: a principled science of human development in context. Educ Psychol 2019; 54: 129–143.

63.

Osher

Kidron

Brackett

, et al. Advancing the science and practice of social and emotional learning: looking back and moving forward. Rev Res Educ 2016; 40: 644–681.

64.

Page

Tetzlaff

Moher

. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Value Health 2021; 23: S312–S313.

65.

Higgins

JPT

Altman

Gøtzsche

, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. Br Med J 2011; 343: d5928.

66.

Schünemann

Vist

Higgins

, et al. Interpreting results and drawing conclusions. In: Higgins

Thomas

(eds) Cochrane handbook for systematic reviews of interventions. Hoboken: John Wiley & Sons, 2019, pp.403–431.

67.

McGuinness

Higgins

. Risk-of-bias VISualization (robvis): an R package and Shiny web app for visualizing risk-of-bias assessments. Res Synth Methods 2021; 12: 55–61.

68.

Moher

Liberati

Tetzlaff

, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009; 151: 264–269.

69.

Singh

. Critical appraisal skills programme. J Pharmacol Pharmacother 2013; 4: 76–76.

70.

Ackley

Zimmerman

Brenowitz

, et al. Effect of reductions in amyloid levels on cognitive change in randomized trials: instrumental variable meta-analysis. Br Med J 2021; 372: n156.

71.

Cheung

MWL

. Modeling dependent effect sizes with three-level meta-analyses: a structural equation modeling approach. Psychol Methods 2014; 19: 211–229.

72.

Dunson

. Commentary: practical advantages of Bayesian analysis of epidemiologic data. Am J Epidemiol 2001; 153: 1222–1226.

73.

Sutton

Abrams

. Bayesian Methods in meta-analysis and evidence synthesis. Stat Methods Med Res 2001; 10: 277–303.

74.

Thompson

Semma

. An alternative approach to frequentist meta-analysis: a demonstration of Bayesian meta-analysis in adolescent development research. J Adolesc 2020; 82: 86–102.

75.

Konstantopoulos

. Fixed effects and variance components estimation in three-level meta-analysis. Res Synth Methods 2011; 2: 61–76.

76.

Viechtbauer

. Conducting meta-analyses in R with the metafor package. J Stat Softw 2010; 36: 1–48.

77.

Cushing

Brannon

Suorsa

, et al. Systematic review and meta-analysis of health promotion interventions for children and adolescents using an ecological framework. J Pediatr Psychol 2014; 39: 949–962.

78.

Higgins

Thompson

Deeks

, et al. Measuring inconsistency in meta-analyses. Br Med J 2003; 327: 557–560.

79.

Quintana

. From pre-registration to publication: a non-technical primer for conducting a meta-analysis to synthesize correlational data. Front Psychol 2015; 6: 1549.

80.

McShane

Böckenholt

Hansen

. Adjusting for publication bias in meta-analysis: an evaluation of selection methods and some cautionary notes. Perspect Psychol Sci 2016; 11: 730–749.

81.

Renkewitz

Keiner

. How to detect publication bias in psychological research. Z Psychol 2019; 227: 261–279.

82.

Lovakov

Agadullina

. Empirically derived guidelines for effect size interpretation in social psychology. Eur J Soc Psychol 2021; 51: 485–504.

83.

Bürkner

. Brms: an R package for Bayesian multilevel models using Stan. J Stat Softw 2017; 80: 1–28.

84.

Williams

Rast

Bürkner

. Bayesian meta-analysis with weakly informative prior distributions. PsyArXiv 2018; 10: 1–19.

85.

Vehtari

Gelman

Gabry

. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Stat Comput 2017; 27: 1413–1432.

86.

Conroy

Coatsworth

. The effects of coach training on fear of failure in youth swimmers: a latent growth curve analysis from a randomized, controlled trial. J Appl Dev Psychol 2004; 25: 193–214.

87.

Denison

Faulkner

Brien

, et al. Effectiveness of an educational intervention targeting homophobic language use by young male athletes: a cluster randomised controlled trial. Br J Sports Med 2023; 57: 515–520.

88.

Pulido

Leo

González-Ponce

, et al. Methodological intervention with soccer coaches to improve athlete-perceived coaching competency, satisfaction with the coach, enjoyment and intention to persist. Int J Sports Sci Coach 2021; 16: 16–26.

89.

Simon

Meeûs

Serpa

, et al. The effects of video feedback on coaches’ behavior and the coach-athlete relationship. J Clin Sport Psychol 2010; 4: 323–340.

90.

Smith

Burns

O’Neill

, et al. An evaluation of the impact of the Irish Rugby Football Union Coach Education Framework on the coach–athlete dyad across the age and stage spectrum in rugby union. Int J Sports Sci Coach 2024; 19: 1927–1939.

91.

Sullivan

. The effects of a coaching education workshop on the self-regulated motivation of 6th grade male and female basketball players. PhD Thesis, The Ohio State University, USA, 2005.

92.

Guagliano

Lonsdale

Kolt

, et al. Increasing girls’ physical activity during a short-term organized youth sport basketball program: a randomized controlled trial. J Sci Med Sport 2015; 18: 412–417.

93.

Langdon

Schlote

Harris

, et al. Effects of a training program to enhance autonomy supportive behaviors among youth soccer coaches. J Hum Sport Exerc 2015; 10: 1–14.

94.

McLaren

Eys

Murray

. A coach-initiated motivational climate intervention and athletes’ perceptions of group cohesion in youth sport. Sport Exerc Perform Psychol 2015; 4: 113–126.

95.

Koh

Komar

Tarkington

, et al. The effects of values and principles in sports coach education course designed to promote values-driven coaching styles. Int J Sports Sci Coach 2024; 19: 2272–2285.

96.

Wilczyńska

Łysak-Radomska

Podczarska-Głowacka

, et al. The effectiveness of psychological workshops for coaches on well-being and psychomotor performance of children practicing football and gymnastics. J Sports Sci Med 2021; 20: 586.

97.

Wilczyńska

Walczak-Kozłowska

Alarcón

, et al. Dimensions of athlete-coach relationship and sport anxiety as predictors of the changes in psychomotor and motivational welfare of child athletes after the implementation of the psychological workshops for coaches. Int J Environ Res Public Health 2022; 19: 3462.

98.

Reynders

Vansteenkiste

Van Puyenbroeck

, et al. Coaching the coach: intervention effects on need-supportive coaching behavior and athlete motivation and engagement. Psychol Sport Exerc 2019; 43: 288–300.

99.

Gould

Collins

Lauer

, et al. Coaching life skills through football: a study of award winning high school coaches. J Appl Sport Psychol 2007; 19: 16–37.

100.

Holt

Deal

Pankow

. Positive youth development through sport. In: Tenenbaum

Eklund

Boiangin

(eds) Handbook of sport psychology: social perspectives, cognition, and applications. 4th ed. Hoboken: John Wiley & Sons, Inc., 2020, pp.429–446.

101.

Legg

Newland

Bigelow

. Somebody’s eyes are watching: the impact of coaching observations on empowering motivational climates and positive youth development. J Park Recreat Admi 2018; 36: 90–106.

102.

Camiré

Trudel

Forneris

. High school athletes’ perspectives on support, communication, negotiation and life skill development. Qual Res Sport Exerc 2009; 1: 72–88.

103.

Klimstra

. Adolescent personality development and identity formation. Child Dev Perspect 2013; 7: 80–84.

104.

Mõttus

Briley

Zheng

, et al. Kids becoming less alike: a behavioral genetic analysis of developmental increases in personality variance from childhood to adolescence. J Pers Soc Psychol 2019; 117: 635.

105.

Suzuki

Fitzpatrick

. Good anxiety: harnessing the power of the most misunderstood emotion. New York: Atria Books, 2021.

106.

Beaty

Kenett

Christensen

, et al. Robust prediction of individual creative ability from brain functional connectivity. Proc Natl Acad Sci U S A 2018; 115: 1087–1092.

107.

Backes

Bonnie

. National Academies of Sciences, Engineering, and Medicine. In: The promise of adolescence: realizing opportunity for all youth. Washington: National Academies Press, 2019.

108.

Bengtsson

Svensson

Hertting

, et al. Long-term effects of basic coach education on ice hockey coaches’ beliefs and use of need-supportive behaviors. Appl Psychol 2026; 75: e70083.

109.

Côté

Vierimaa

. The developmental model of sport participation: 15 years after its first conceptualization. Sci Sport 2014; 29: S63–S69.

110.

Smith

Hamon

. Exploring family theories. 3rd ed. New York: Oxford University Press, 2012.

111.

Alexander

Stafford

Lewis

. The experiences of children participating in organised sport in the UK. Report, NSPCC, UK, 2011.

112.

Bean

Fortier

Post

, et al. Understanding how organized youth sport may be harming individual players within the family unit: a literature review. Int J Environ Res Public Health 2014; 11: 10226–10268.

113.

Normand

Wolfe

Peak

. A review of early sport specialization in relation to the development of a young athlete. Int J Kinesiol Sports Sci 2017; 5: 37–42.

114.

Reeve

. Why teachers adopt a controlling motivating style toward students and how they can become more autonomy supportive. Educ Psychol 2009; 44: 159–175.

115.

Roberts

Sojo

Grant

. Organizational factors and non-accidental violence in sport: a systematic review. Sport Manag Rev 2020; 23: 8–27.

116.

Swigonski

Enneking

Hendrix

. Bullying behavior by athletic coaches. Pediatrics 2014; 133: e273–e275.

117.

Waldron

DeFreese

Pietrosimone

, et al. Exploring early sport specialization: associations with psychosocial outcomes. J Clin Sport Psychol 2019; 14: 182–202.

118.

Super

Verkooijen

Koelen

. The role of community sports coaches in creating optimal social conditions for life skill development and transferability–a salutogenic perspective. Sport Educ Soc 2018; 23: 173–185.

119.

Zeisner

Smith

. How athletic coaches can play a key role in SEL development: a literature review of Social-Emotional Learning (SEL). Appl Res Coach Athl Annu 2022; 37: 177–206.

120.

McCleery

Tereschenko

, et al. Gender differences in coaching behaviors supportive of positive youth sports experience. Women Sport Phys Act J 2023; 31: 63–72.

121.

Tan

LQW

O’Connor

. Coaches’ perspectives of the continuing coach education program in the development of quality coach education in Singapore. Int Sport Coach J 2022; 10: 230–243.

122.

Van Aert

Wicherts

van Assen

. Conducting meta-analyses based on p values: reservations and recommendations for applying p-uniform and p-curve. Perspect Psychol Sci 2016; 11: 713–729.

123.

Jones

Greenberg

Crowley

. Early social-emotional functioning and public health: the relationship between kindergarten social competence and future wellness. Am J Public Health 2015; 105: 2283–2290.

124.

North

Callary

Dieffenbach

, et al. A reflection on the state of sport coaching research, its community, and representation: the 2020 international council for coaching excellence research committee consultation. Int Sport Coach J 2021; 8: 405–413.

125.

Whitley

Fraser

Dudfield

, et al. Insights on the funding landscape for monitoring, evaluation, and research in sport for development. J Sport Devel 2020; 8: 21–35.

126.

Ali

Darnell

Dinunzio

. Bringing the community to the uni—critical reflections on youth recreation partnerships in Toronto. Front Sports Act Living 2022; 4: 876468.

127.

Moosbrugger

Mullin

, et al. Targeting well-being and physical activity through sport education in higher education. Quest 2022; 74: 389–405.

128.

Louis

Moosbrugger

. Cultivating college students’ motivation in physical education through sport education model in COVID-19 era: a prospective cohort study. Phys Educator 2024; 81: 205–230.

129.

Backman

Henriksen

Zettler

. Beyond self-compassion: a scoping review of other-focused compassion in competitive sports environments. Int J Sport Exerc Psychol 2025; June: 1–28.

130.

Cormier

Kowalski

Ferguson

, et al. Self-compassion in sport: a scoping review. Int Rev Sport Exerc Psychol 2025; 18: 154–193.

131.

McGorry

Mei

Dalal

, et al. The Lancet Psychiatry Commission on youth mental health. Lancet Psychiatry 2024; 11: 731–774.

132.

Fleming

Robinson

, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014; 384: 766–781.

133.

Malfacini

. The impacts of companion AI on human relationships: risks, benefits, and design considerations. AI Soc 2025; 40: 5527–5540.

134.

Ozmen Garibay

Winslow

Andolina

, et al. Six human-centered artificial intelligence grand challenges. Int J Hum Comput Interact 2023; 39: 391–437.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.03 MB

0.01 MB

0.17 MB

0.05 MB

0.01 MB

0.03 MB

0.04 MB