The Theory of Planned Behavior (TPB), developed by Fishbein and Ajzen as an extension of the Theory of Reasoned Action, is one of the most extensively applied theoretical models for predicting exercise behavior (Fang, 2011; Feng and Mao, 2014). Its application in sports participation research primarily follows two trajectories. The first trajectory focuses on building and validating predictive or interventional models for general physical activity within specific populations, such as adolescents (Fang and Sun, 2010; Hagger et al., 2007). The second trajectory employs TPB to explain and predict participation in specific sports or leisure activities, including pickleball tournament participation (Wang et al., 2023), ice-snow tourism (Cao and Ge, 2024), and sports lottery purchases (Liu et al., 2014).

Theory of planned behavior posits that behavioral intention, the most proximal predictor of actual behavior, is shaped by three core antecedents: attitude (an individual’s positive or negative evaluation of performing the behavior), subjective norm (perceived social pressure from significant others), and perceived behavioral control (the perceived ease or difficulty of performing the behavior). While the predictive validity of intention is well-established, the relative influence of these antecedents can vary across behavioral contexts. For instance, research on ice-snow sports found subjective norm to be the strongest predictor for adolescents (Liang, 2022), whereas studies on lottery purchase identified attitude as the dominant factor (Liu et al., 2014). As a form of sports participation and leisure activity, the formation of behavioral intention among pickleball participants may share similarities with these findings. Nevertheless, the relative explanatory power of each antecedent within the context of pickleball requires further investigation. Based on the TPB framework, this study proposes the following Hypothesis:

Despite its widespread application, the Theory of Planned Behavior (TPB) has well-documented limitations. Empirical evidence suggests that a significant portion of variance in behavioral intention remains unexplained by its three core antecedents alone (Armitage and Conner, 2001). Furthermore, the framework does not explicitly address the motivational origins of these antecedents, thereby limiting its explanatory depth regarding why individuals form specific behavioral intentions (Hagger and Chatzisarantis, 2009). To address this, scholars have attempted to extend the model by integrating complementary theoretical perspectives such as Self-Determination Theory. Within this integrative effort, the construct of motivation has garnered considerable attention due to its central role in initiating, guiding, and sustaining behavior (Feng and Mao, 2014; Wang et al., 2025). Meta-analytic studies (Hagger and Chatzisarantis, 2009) and empirical research in domains including innovation adoption (Zheng and Li, 2022) have demonstrated that motivation not only significantly influences behavioral intention but also shapes the three TPB antecedents.

Exercise motivation, defined as the psychological drive stimulating participation in physical activity (Chen et al., 2013), is a fundamental factor that may precede and shape the cognitive evaluations within TPB. Individuals with strong exercise motivation are more likely to develop a positive attitude toward a specific activity like pickleball, perceive greater social support or normative expectations, and feel more confident in their ability to participate. This view is supported by studies in related domains, such as those on sports event volunteers and esports users, where exercise motivation significantly predicted attitude (Ma and Wang, 2024; Zhang et al., 2023). Consequently, exercise motivation is posited not only to directly enhance participation intention but also to exert indirect influence through the TPB antecedents. Therefore, this study proposes the following Hypothesis:

Furthermore, prior research integrating motivational constructs with TPB has established an extended theoretical framework wherein motivation influences behavioral intention through the mediation of cognitive and perceptual evaluations, namely the TPB constructs (Feng et al., 2015; Ge et al., 2020). As a form of physical activity, pickleball participation is conceptually situated within this established integrative framework. However, given the distinctive characteristics of pickleball, which include its emerging status, strong recreational and social appeal, and the typically community-driven rather than obligatory nature of participation, the specific mediating pathways through which exercise motivation influences the intention to play pickleball require further empirical examination, with the role of subjective norm being of particular interest. Accordingly, this study posits and aims to test the following mediation Hypothesis:

Accordingly, based on the above discussions, this study proposes a research model to investigate the factors influencing behavioral intentions of recreational pickleball participants, as illustrated in Figure 1.

This study investigates the behavioral intentions of recreational pickleball participants and their influencing factors. Data were collected between January and March 2025 using a mixed-mode survey strategy combining online and offline questionnaires. Offline data were gathered through convenience sampling at four established pickleball venues in Guangzhou and Shenzhen, China, where on-site players were randomly invited to participate. This approach facilitated direct access to active participants, thereby enhancing the relevance of responses—a recognized advantage of convenience sampling in exploratory research contexts (Etikan et al., 2016). Online data collection employed snowball sampling, with the questionnaire distributed through social media and dedicated pickleball enthusiast groups to extend reach to a broader and more varied network of participants, including those engaged in informal and socially oriented play—a method particularly effective for accessing otherwise hard-to-reach populations (Naderifar et al., 2017). While these non-probability sampling approaches may limit the generalizability of the findings, they are well-suited for exploratory investigations focusing on emerging and previously under-researched groups.

A total of 360 questionnaires were distributed. After excluding responses with excessively short completion times or clear patterned answers, 315 valid questionnaires were retained, yielding a valid response rate of 87.5%. The sample consisted of 53.3% male and 46.7% female participants. Regarding age distribution: 2.5% were under 18 years old, 36.3% were 18–25 years old, 18.3% were 26–33 years old, 10.5% were 34–41 years old, 11.7% were 42–49 years old, and 20.6% were 50 years or older.

The development of the questionnaire followed a systematic process. First, based on the research objectives and characteristics of pickleball, established scales from domestic and international literature were reviewed to draft initial items. Second, in-depth interviews were conducted with both casual and experienced pickleball participants, and the items were refined based on their feedback. Finally, six experts in sports science and psychology evaluated the content validity of the questionnaire, all of whom rated it as good or higher.

The final questionnaire comprised two sections: the first collected demographic information, and the second measured the core constructs, including exercise motivation, attitude, subjective norm, perceived behavioral control, and behavioral intention.

Data analysis was performed using SPSS 27.0 for reliability analysis, descriptive statistics, correlation analysis, and hypothesis testing. Amos 28.0 was employed for common method bias (CMB) testing and confirmatory factor analysis (CFA), with the bootstrap method used to examine the proposed mediation model. fsQCA 3.0 software was applied for qualitative comparative analysis of the condition configurations influencing the dependent variable.

Table 3 presents the descriptive statistics and correlations among the latent variables. The mean values of exercise motivation, attitude, subjective norm, perceived behavioral control, and behavioral intention ranged from 4.158 to 4.402, with standard deviations below 0.7, indicating low variability and good model fit. Correlation analysis revealed significant positive relationships among all latent variables. As shown in Table 3, the square roots of the AVE for all constructs were greater than their correlations with other latent variables, demonstrating adequate discriminant validity.

While SEM focuses on the net effects of linear relationships, it does not account for interactions among variables or their combined effects. In contrast, fsQCA is particularly suited for analyzing outcomes arising from multiple concurrent causes (Du and Jia, 2017). Therefore, to complement the SEM findings and enhance the depth and robustness of the results, this study employed fsQCA to examine the configurations of antecedent conditions influencing behavioral intentions toward pickleball participation. Based on theoretical and empirical foundations, four antecedent conditions were selected: exercise motivation, attitude, subjective norm, and perceived behavioral control.

Exercise motivation is a pivotal determinant of behavioral intention in pickleball participation. This finding corroborates existing evidence on the positive link between motivation and exercise intention (Ge et al., 2020). More specifically, the results indicate that within the emerging, low-structure context of pickleball, participation intention is strongly driven by intrinsic exercise motives. Unlike traditional racket sports underpinned by established organizational structures and systematic coaching pathways, pickleball engagement appears to be primarily fueled by autonomous motivations, such as the pursuit of health, enjoyment, social connection, or skill mastery. This outcome aligns with the Self-Determination Theory principle that autonomous motivation more effectively sustains behavioral engagement. It further implies that promoting such a sport requires strategies aimed at activating and nurturing these internal drivers.

Attitude and perceived behavioral control play crucial and fully mediating roles in the mechanism influencing pickleball participation intention, consistent with prior findings (Zhang et al., 2023). This pattern of full mediation indicates that, in this context, exercise motivation must first translate into positive cognitive-affective evaluations and a sense of capability before shaping a concrete intention to participate. Specifically, attitude, which captures an individual’s cognitive and affective evaluation of participating in pickleball, serves as the critical translator. A positive attitude leads individuals to perceive the activity as enjoyable and beneficial, thereby strengthening their intention to engage. Simultaneously, perceived behavioral control, reflecting the individual’s assessment of the resources and abilities needed to perform the behavior, acts as a vital enabler. The accessible nature of pickleball, characterized by simple equipment and easy-to-learn rules, effectively reduces technical anxiety and lowers the initial barrier to entry. This accessibility enhances participants’ sense of self-efficacy and controllability. Consequently, promotion strategies must recognize that merely stimulating motivation is insufficient; it is equally essential to cultivate positive subjective experiences and to design participation environments that reinforce individuals’ confidence in their ability to succeed.

The results indicate that subjective norm did not significantly predict pickleball participation intention, and its mediating role between exercise motivation and behavioral intention was non-significant. This finding differs from some conclusions in prior physical activity research. For instance, a meta-analysis in health contexts found a significant, albeit small-effect, mediating role of self-determined motivation on intention via subjective norm (Hagger and Chatzisarantis, 2009). However, it aligns with observations from other studies applying TPB in contexts emphasizing personal volition, such as research on Chinese university students’ extracurricular physical exercise, where subjective norm similarly showed no significant effect (Xia and Shi, 2025). The attenuated role of subjective norms in this study can be attributed to an interplay of activity-specific, cultural, and demographic factors.

First, regarding activity type, pickleball as an emerging leisure sport is characterized by self-organized community engagement and participation driven primarily by personal interest. Unlike institutionalized or health-normative physical activities, its appeal lies in enjoyment and social interaction, where individual autonomy tends to outweigh perceived social pressure (Wang et al., 2023). Second, in terms of cultural and demographic context, our sample was drawn from metropolitan residents in Guangzhou and Shenzhen—economically developed and culturally open cities in China. This socio-cultural milieu, often associated with higher individualism and greater exposure to novel lifestyle choices, may further diminish the salience of traditional social influences on individual behavioral decisions.

Consequently, the findings suggest that the predictive strength of subjective norm within the TPB framework is not universal but is contingent upon behavioral and contextual factors. This does not refute the TPB but underscores the need for its context-sensitive refinement, acknowledging that the relative importance of its core constructs may vary across behavioral domains and cultural settings. It is therefore pertinent that this study employs fsQCA to further investigate whether and how subjective norm operates conditionally within specific configurations of personal and social factors.

The SEM results of this study show that the traditional TPB model explained 68% of the variance in behavioral intention toward pickleball participation among participants, reaffirming its universal value in the domain of individual behavioral decision-making. After introducing exercise motivation as an antecedent variable, the expanded TPB model explained 82% of the variance in behavioral intention, representing a 14% increase over the traditional TPB. This outcome suggests that while the traditional TPB model can explain part of the mechanism behind intention formation, the inclusion of exercise motivation renders the prediction of behavioral intention more precise and comprehensive, particularly in the context of emerging sports characterized by distinct individualistic features. Theoretically, integrating exercise motivation upgrades the traditional TPB framework to a dynamic system that outlines a pathway from being motivation-driven, through psychological cognition, to intention formation. This systemic perspective aligns well with the strong autonomy and diverse needs inherent in pickleball participation.

While the preceding SEM analysis identified the average strength and mediation pathways within the extended TPB model, the fsQCA results provide a crucial complementary perspective by uncovering multiple, distinct pathways to high behavioral intention.

Specifically, fsQCA identified two sufficient causal recipes. The first path centers on the core combination of high exercise motivation and a positive attitude, representing an internally-driven route. The second path hinges on a positive attitude together with strong subjective norms, supported by perceived behavioral control, illustrating a socially-embedded route.

The theoretical significance of these configurational findings is twofold. First, they robustly establish a positive attitude as a necessary, core condition across all solutions, confirming its paramount role. Second, and more importantly, they reveal the conditional nature of variable influences. While SEM indicated a non-significant average mediating effect for subjective norm, fsQCA demonstrates that it can operate as a core causal element within the specific social-contextual configuration (Path 2). This suggests that the influence of social norms is not universally weak but is activated decisively only when combined with a positive personal attitude and adequate perceived control. Similarly, the presence of different core conditions across paths shows that high intention can be driven either by strong internal motivation or by a supportive social environment, a complexity that linear models typically do not capture.

Thus, the two methods offer a dialectical insight. SEM effectively maps the central tendencies and primary mechanisms. fsQCA, in turn, explains the heterogeneity underlying these averages by showing how the same outcome can be achieved through different logical combinations of conditions. This integration resolves the apparent paradox of subjective norm’s role, moving from a conclusion of “no general effect” to a more precise understanding of its contextual importance. Together, they advance a more complete and contingent theoretical framework for understanding participation in emerging sports.

This study advances the Theory of Planned Behavior (TPB) and its application to emerging, voluntary recreational sports through three key theoretical contributions. First, it extends the TPB by establishing exercise motivation as a critical distal antecedent. This specification addresses a core theoretical limitation by elucidating the motivational origins of the model’s cognitive and social antecedents, thereby significantly enhancing its explanatory power for behaviors driven primarily by personal volition rather than external mandates. Second, it refines the understanding of social influence within TPB by clarifying the conditional role of subjective norm. The findings advance the theoretical perspective from viewing subjective norm as a universally predictive factor to recognizing it as a contextually activated core condition, thereby delineating important boundary conditions for its operation. Third, it moves beyond the conventional linear paradigm of TPB research. While prior studies have predominantly focused on linear relationships among variables, our integrated SEM–fsQCA approach explicitly accounts for configurational effects. This provides a more nuanced understanding of participation decisions and establishes a methodological pathway for capturing causal complexity in behavioral research.

Based on the core findings of this study, which identify exercise motivation, positive attitude, and perceived behavioral control as key psychological drivers of behavioral intention toward pickleball, promotion strategies should focus on systematically stimulating intrinsic motivation, shaping positive initial experiences, and lowering barriers to entry. Specifically, promotional content and activity design should directly address participants’ diverse internal motives by highlighting pickleball’s benefits for health and appearance motivation, such as its role in improving cardiorespiratory function, coordination, and body composition; emphasizing its dynamic and enjoyable gameplay to satisfy enjoyment motivation; leveraging its doubles-dominated format and community atmosphere to meet social motivation; and offering clear skill progression pathways to fulfill competence motivation. In practice, priority should be given to creating low-cost, easily accessible “first-experience” scenarios, such as utilizing existing public spaces through minimal infrastructural modifications. It is essential to ensure that beginners receive guided support to achieve enjoyable and successful initial encounters, thereby rapidly fostering a positive attitude. Concurrently, structured onboarding guidance, positive skill transfer from related sports, and the establishment of peer support networks should be provided to reduce technical anxiety and perceived difficulty, effectively enhancing participants’ sense of control and self-efficacy. Supporting measures, including training community instructors, developing accessible event formats, and integrating information resources, can provide the necessary foundation for these core psychological strategies. Collectively, these measures establish a cohesive framework that supports the progression from motivational impetus to sustained behavioral engagement.

However, this study has several limitations. First, the primary use of cross-sectional survey data limits the ability to definitively establish causal relationships among variables. Future research could employ longitudinal tracking or experimental methods to further validate the findings. Second, the sample size was relatively limited, and data collection was confined to specific regions within China. Variations in policy support, economic levels, cultural context, and geographical climate across different regions might lead to differentiated mechanisms influencing pickleball participation intention. For instance, from a policy dimension, varying levels of government support for emerging sports and investment in public sports facilities directly affect the role of facility accessibility as an external condition within perceived behavioral control. Economically, disparities in disposable income and sports consumption capacity between developed and developing countries might result in regional differences in the strength of exercise motivation’s influence. Therefore, future studies could adopt stratified regional sampling and multi-group comparison analysis to contrast the strength of macro-regional characteristic influences and the generalizability of the findings.