Sport is defined as a structured, goal-oriented, competitive and contest-based form of physical activity. A growing body of studies have confirmed that persistent sports participation not only has a positive effect on the skills, physical development and social adaptation of adolescents, it also plays a significant role in reducing the risk of chronic diseases, cancer and obesity and in preventing psychological problems (Lindwall et al., 2014; Morris et al., 2019). In contrast to individuals who do not participate in sports, persistent participants experience less psychological difficulties and higher quality of life related to health (Vella et al., 2014a,b). In addition, exercise in adolescence increases the likelihood of physical activity in adulthood, thereby enabling the maintenance of a lasting and beneficial effect on their physical and mental health (Huotari et al., 2011). Despite these potential benefits, young athletes may be struggling with the decision to quit sports. Studies indicate that ~20–50% of young athletes between the ages of 10 and 17 dropout in sports each year (Balish et al., 2014). In Sweden, 77% of children aged 6–12 participate inorganized sports. However, between the ages of 13 and 25, this proportion dropped to 41% (Eliasson and Johansson, 2021). The portion of this dropout may reflect that the teenagers in the sample are trying to transfer between different sports, or quit sports in order to enter other fields such as music, STEM classes, etc. However, some withdrawal from sports may reflect dissatisfaction or negative experience. Sport dropout among young people has become such a common phenomenon in the world that the research on the behavior of persistent sports participation has been given more attention to by many clubs and researchers (Monteiro et al., 2018a; Eime and Harvey, 2019; Silva et al., 2019). Identifying the influencing factors of persistent sports participation in youth becomes more important.

From a measurement standpoint, dropout was primarily assessed on whether youth participants registered for their sport in subsequent seasons (Balish et al., 2014; Crane and Temple, 2015). Using registration in the subsequent season provides a minimalist view of dropout. Therefore, persistent participation is defined as participants go on their sport in the following season. The duration shall be at least 1 year. Many studies have attempted to explain sport persistence and dropout in relation to participants' underlying psychological characteristics, and various theories have been applied to frame this phenomenon. In previous studies, self-determination theory (SDT), achievement goal theory (AGT) and theory of planned behavior (TPB) are highly appropriate conceptual framework from which to study sport persistence and dropout. For example, Gardner et al. (2016) explored the antecedents of enjoyment and intention to continue in youth sports based on the theory of the AGT. Their study indicated that the social climate profiles were linked with intention to continue through enjoyment and the positive coach relationship quality profiles were relatively higher levels of enjoyment and intention to continue (Gardner et al., 2016). Joesaar and Hein (2011) integrated the AGT and SDT theories to confirm that youth athletes' task-involving peer motivational climate and intrinsic motivation predict sport persistence among the athletes (Joesaar et al., 2011). Gucciardi and Jackson (2015) integrated the theories of planned behavior (TPB) and basic psychological needs (BPN) to identify factors associated with young adults' continuation in organized sport. The results indicated that the satisfaction of basic psychological needs, intention and perceived behavioral control predicted sport continuation (Gucciardi and Jackson, 2015). In addition, researchers explored other factors of persistent participation in youth sports, such as demographic, biological, psychological, cultural, environmental (Boiche and Sarrazin, 2009; Bouffard, 2017; Wendling et al., 2018; Soares et al., 2020). Many cross-sectional and longitudinal studies have also shown that persistent sports participation is associated with higher perceived competence, self-esteem, and better emotional and social adaptation (Duda, 2013; Smith et al., 2016).

Although numerous studies identify the factors of persistent participation in youth sports, few researchers have conducted systematic review and meta-analysis of available articles to my knowledge. Due to the different research perspectives and theories adopted by researchers, the understanding of persistent participation in youth sports also differs, making it difficult to fully reflect the influencing factors of persistent participation. The strength or significance of some factors are also inconsistent in different articles. Therefore, it is essential to analyze the factors associated with persistent participation in youth sports through a systematic review and meta-analysis. The aims of this study are as follows: (a) to systematically review the factors of persistent sports participation in the empirical papers and (b) to assess their respective strengths associated with persistent participation in organized sport through the method of meta-analysis. This study can provide a complete understanding of children and adolescents' decisions to continue or discontinue their participation in organized sports. An overview of these factors can provide information that is useful for sports clubs and policy makers for developing new interventions to increase participation in organized sport among these adolescents.

The present review was register on the international platform of PROSPERO (Booth et al., 2012) (registration number: CRD42021229397) and was reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (Moher et al., 2015).

Although numerous studies identify the factors of persistent participation in youth sports, this review synthesizes the factors of persistent participation by using the methods of meta-analysis. Due to the different research perspectives and theories adopted by researchers, the understanding of persistent participation in youth sports also differs, making it difficult to fully reflect the influencing factors of persistent participation. The strength or significance of some factors are also inconsistent in different articles. In this review we summarized the primary factors of persistent participation and conducted a meta-analysis of available effect sizes to assess their respective strengths of association with persistent participation in youth sport. Our findings showed that sports enjoyment was highly correlated with persistent intention, while persistent intention was highly correlated with persistent behavior. In addition, parental support, coach support, peer support, basic psychological needs and sports competence were the primary factors associated with persistent intention and persistent behavior, respectively.

Consistent with previous studies, sports enjoyment and persistent intention were the two important variables for studying sports behavior (Quested et al., 2013; Gardner et al., 2016). Many researchers have also confirmed that sports enjoyment and intention were the key motivational processes that affected teenagers' persistent participation in sports, and lack of enjoyment was the main reason for dropout (Gardner et al., 2017). In the present study, sports enjoyment was highly correlated with persistent intention. However, among the factors of persistent behavior, the factor of sports enjoyment was not identified. The main reason for this finding was that among the 12 longitudinal studies, only two articles demonstrated the influence of sports enjoyment on persistent behavior. Hence, it did not meet the criteria of selection, which should be mentioned in more than two articles. There was a medium effect between intrinsic motivation and persistent behavior, but the extrinsic motivation including three forms of regulation had no significant effect. Although many researchers confirmed that intrinsic motivation and amotivation were the main factors that affect athletes' persistent participation in sports, the conclusions of the studies on the influence of different forms of extrinsic motivation on participation behavior were inconsistent. For example, Vallerand and Rousseau (2001) found that identity regulation was also an important part of athletes' performance, in addition to intrinsic motivation. Pelletier et al. (2001) also believed that a significant difference in identity regulation between persistent and dropout athletes existed, but no significant difference in introjected regulation.

Many researchers indicated that parental support, coach support and peer support were significantly correlated with persistent intention and behavior (Harwood et al., 2015; Monteiro et al., 2018b). In the current study, parental support weakly correlated with persistent intention and persistent behavior. The reason might be children with parental support tended to experience more enjoyment and intrinsic motivation, and thus, they were more likely to continue to participate (Fredricks and Eccles, 2002). However, as children grow older, parents' support gradually decreased because the learning tasks increased.

The relationship between coach support and persistent intention was moderately correlated along with the behavior of persistent participation. The factor of coach support was a key component of the motivational climate, which was characterized by a high sense of intimacy, commitment, complementarity and common orientation (Jowett and Ntoumanis, 2004). Many psychological outcomes in sports, including motivation and intention to participate, were related to coach support (Riley and Smith, 2011; Rocchi et al., 2017). However, low-quality coach-athlete relationships, including coach conflict, coach's control styles, lack of encouragement, and an overemphasis on victory were usually associated with sports dropout behavior (Gearity and Murray, 2011).

A significantly positive relationship between peer support and intention and behavior of persistent participation was also observed. In sports, young people regarded the perceived peer acceptance and friendship as one of the driving forces for persistent participation (Keegan et al., 2010). Ullrichfrench and Smith (2006) found that peer support was significantly correlated with persistent participation intention, and high-quality friendship could buffer negative results related to acceptance.

The relationship between basic psychological needs and persistent intention was moderately correlated but weakly correlated with persistent behavior. For example, Guillet et al. (2002) confirmed that persistent players perceived themselves as significantly more competent, more autonomous and more related to their team than dropout players. Joesaar et al. (2011) also supported the results in which dropout athletes' satisfaction with autonomy, competence and relatedness needs were lower than that of persistent athletes. According to the self-determination theory, only when the 3 basic psychological needs (competence, autonomy and relatedness) were satisfactory could athletes maintain the motivation of self-determination. On the contrary, uncertain motivation and incentive forms were promoted, which might lead to sports dropout (Sarrazin et al., 2002).

The results of the research on the relationship between sports competence and persistent participation were inconsistent. Athletes with better sports competence were generally believed to be more likely to continue to participate in sports, and the perceived competence was considered to be an important predictor of persistent behavior (Soderstrom et al., 2018). The comparative study of persistent behavior and sports dropout indicated that the score of sports competence of dropout athletes was significantly lower than that of persistent athletes. On the contrary, Bars et al. (2009) and Calvo et al. (2010) believed that sports competence was not an important predictor of persistent participation, and relatedness and autonomy might be more important than sports competence in explaining persistent participation or dropout behavior. The reason for the above inconsistency might be that different methods and tools were used to evaluate sports competence. For example, the tool with only 1 item used in Calvo's study might not be the best way to assess the sports competence of athletes.

In this study, among the included articles, although 2 ones studied years of involvement, meta-analysis was not carried out because they did not meet the criteria. In previous studies, years of involvement were significantly associated with persistent participation. Rottensteiner et al., 2015 suggested that persistent athletes appeared to show a higher level of competition, long years of participation and higher exercise frequency in sports than dropout athletes. However, Guillet et al. (2002) showed that although the years of involvement of persistent athletes were higher than those of dropout athletes, no difference in weekly exercise frequency could be observed between the two groups, and no significant correlation between years of involvement and persistent participation behavior. To sum up, not much evidence was found to prove the causal relationship between the years of involvement and persistent participation. In the future, the relationship between years of involvement and persistent participation behavior should be discussed further by expanding the sample of athletes.

The paper systematically reviews the different viewpoints and conclusions on the influencing factors of persistent participation in youth sports. By utilizing techniques of meta-analysis, this study synthesizes 6 factors related to the intention of persistent participation and seven factors related to the behavior of persistent participation in sports. It can provide a reference for researchers to further understand the relationship between influencing factors and persistent participation in youth sports. In addition, this study strictly follows the principle of meta-analysis to collect and analyse data, making the analysis and results more standardized and reliable. It not only enriches the research content of persistent participation in youth sports but also provides insights into future research in this field.

Although the collected articles come from many important databases, it is still incomplete because the conference database was not retrieved. Second, there are inherent limitations in meta-analysis methods and the software used. For example, the selected factors must be mentioned more than twice in papers, resulting in incomplete factors. In addition, the effect size of some factors is small because few articles and samples related to these factors were found, and the results need to be confirmed further by more articles. Finally, this study only includes the independent variables directly related to the dependent variables as influencing factors, and future research should further consider the effects of mediating or moderating variables and the interaction of independent variables.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

MZ draft almost the manuscript. X-CW did data collection and coding and data analysis. BS contributed to study design, supervision, and paper review. All authors contributed to the article and approved the submitted version.

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

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