As sports socializing is becoming a dominant lifestyle that integrates physical health with social interaction in China, understanding the underlying drivers of participation is crucial. However, traditional research predominantly relies on a “variable-centered” paradigm, which assumes population homogeneity and focuses on linear relationships between single motives and behaviors. This approach often fails to capture the complexity of how multiple motivations are configured within individuals (heterogeneity), and how these internal configurations are associated with external behavioral choices. To address this gap, this study employed a novel hybrid methodological framework combining Latent Profile Analysis (LPA) and Random Forest (RF) modeling. Based on data from 1,104 adults, LPA was first used to identify distinct motivational subgroups. Subsequently, RF algorithms, utilizing feature importance ranking and “One-vs-Rest” strategies, were applied to identify the associative patterns between these motivational profiles and key behavioral indicators, including sports types, media usage, and economic investment. The analysis identified four distinct motivational profiles: (1) Psychologically Introverted (3.6%), prioritizing internal psychological rewards over social status; (2) Physiologically Oriented (44.1%), the largest group, driven primarily by physical health needs; (3) Balanced (39.0%), exhibiting moderate levels across all motivational dimensions; and (4) High-Motivation/Comprehensively Oriented (13.3%), showing high intensity in both internal and external rewards. The RF model achieved a training accuracy of 99.9% and identified that Sports Type (specifically large-ball games), Media Channels (particularly Douyin/Rednote), and Annual Spending were the top three salient behavioral markers distinguishing these profiles. Notably, the High-Motivation group was characterized by heavy reliance on visual social media for social display. Participation in sports socializing among Chinese residents is not characterized by a singular, homogeneous motivation but features a clear internal stratification structure. The specific pattern of motivational combinations (i.e., the type) systematically maps onto external behavioral choices, where the sociocultural attributes of the sport and the media characteristics of digital social platforms constitute the key predictive markers of behavioral differentiation. The establishment of this “Motivation Type—Behavioral Signal” integrated framework promotes a theoretical shift in the sports socializing research paradigm from “homogeneity” to “heterogeneity” and deepens the understanding of the complex manifestations of Self-Determination Theory and Social Capital Theory in a sports context. It also provides precise user profiles and behavioral insights for sports social platforms, commercial clubs, and public sports service departments. Exploring service customization and policy adjustments based on different motivation-behavior patterns could potentially enhance user engagement and satisfaction, suggesting a possible direction for the development of the sports socializing industry.
Tian et al. (Wed,) studied this question.