Association between family sports environment and adolescent physical activity: the mediating role of basic psychological needs and the moderating role of academic stress

Abstract

Purpose

Rooted in the integration of Self-Determination Theory (SDT) and Conservation of Resources Theory (COR), this study aims to explore the associations linking family sports environment to adolescent Moderate-to-Vigorous Physical Activity (MVPA). Specifically, it examines the mediating role of basic psychological needs satisfaction and the conditional indirect effects of academic stress on this mediated pathway.

Methods

A multi-stage hybrid sampling method was employed to recruit 523 adolescents (Mean Age = 15.23, SD = 1.76) from six secondary schools across Eastern, Central, and Western China. Participants completed the Family Sports Environment Scale, the Basic Psychological Needs in Exercise Scale, and the Adolescent Academic Stress Scale. Physical activity was objectively measured using ActiGraph GT3X+ accelerometers (valid wear time 4 days, including 1 weekend day). Parental education level and BMI were included as covariates. Data were analyzed using SPSS 26.0 and the PROCESS macro (Model 59).

Results

(1) Measurement model assessment confirmed satisfactory reliability and validity for all constructs. (2) Family sports environment was significantly and positively associated with adolescent MVPA (= 0.15,), with basic psychological needs serving as a significant mediating pathway. (3) Academic stress was observed as a “dual-stage” moderator: it negatively moderated the first stage of the theoretically linked pathway (family environment needs satisfaction), consistent with the resource depletion hypothesis; conversely, it positively moderated the second stage (needs satisfaction MVPA), consistent with the compensatory coping hypothesis. Crucially, despite these opposing moderating effects, the overall indirect effect of family environment on MVPA remained robust and stable across low, medium, and high levels of academic stress.

Conclusion

The findings suggest a complex homeostatic dynamic where the opposing moderating effects of academic stress counterbalance each other. While high stress may attenuate the perception of family support, it simultaneously amplifies the motivation to engage in physical activity when psychological needs are satisfied. These results highlight the potential resilience of family support as a resource for adolescent health behaviors.

Introduction

Adolescence, spanning the ages of 10 to 19, represents a critical developmental window characterized by rapid physiological maturation and the consolidation of lifelong health behaviors. Within this phase, regular engagement in Moderate-to-Vigorous Physical Activity (MVPA) is fundamental for maintaining musculoskeletal integrity, optimizing cardiovascular fitness, and ensuring metabolic homeostasis. Beyond somatic health, physical activity serves as a potent catalyst for cognitive enhancement and psychological well-being, providing resilience against anxiety and depressive symptoms (Chaput et al. 2020). Recognizing these multifaceted benefits, the World Health Organization (WHO) explicitly recommends in its Guidelines on Physical Activity and Sedentary Behaviour that adolescents accumulate at least 60 minutes of MVPA daily (World Health Organization 2020).

Despite the ubiquity of these guidelines, the global prevalence of physical inactivity among adolescents has escalated into a public health crisis, with recent surveillance data indicating that approximately 81% of adolescents worldwide fail to meet recommended activity thresholds (Guthold et al. 2020). This physical inactivity pandemic is particularly acute in China, where the intersection of rapid urbanization and a distinctively competitive educational system has profoundly reshaped youth lifestyles. The 8th National Student Physical Fitness and Health Survey revealed suboptimal compliance rates for physical fitness standards, with a precipitous decline observed during the transition from middle to high school (Fan & Chen 2022). The proliferation of screen-based digital entertainment, coupled with the intensification of academic burdens, has led to a systematic encroachment upon time previously allocated for structured and unstructured exercise. This displacement effect has precipitated a surge in sedentary behaviors and associated risks, including obesity and myopia. Consequently, identifying the ecological and psychological correlates of adolescent physical activity within this specific socio-cultural context is a national priority.

To delineate the motivational processes underlying physical activity, this study adopts Self-Determination Theory (SDT) as a primary theoretical lens. SDT posits that human motivation is not merely a quantitative construct but a qualitative one, driven by the satisfaction of three universal basic psychological needs: autonomy, competence, and relatedness (Deci & Ryan 2000; Ryan & Deci 2017). While these three needs are theoretically distinct, previous research suggests that in collectivist cultures like China, they are often deeply intertwined and function as a unitary global construct in the context of general well-being and health behaviors (Standage & Ryan 2012). Therefore, consistent with recent integrative approaches (e.g., Ahmadi et al. 2023), we operationalize these needs as a composite “Basic Psychological Needs Satisfaction” construct to capture the overall quality of psychological nourishment derived from the environment. According to Basic Psychological Needs Theory (BPNT), a sub-theory of SDT, social environments that support these needs facilitate the internalization of motivation, thereby promoting sustained engagement in health behaviors (Teixeira et al. 2012). As the primary microsystem in Bronfenbrenner’s ecological model, the family plays a pivotal role in this process. The “Family Sports Environment”—encompassing the physical availability of equipment and the social-psychological climate of parental modeling and encouragement—has been consistently associated with adolescent activity levels (Sallis et al. 2000; Yao & Chen 2020). We hypothesize that a supportive family environment is linked to MVPA partially through the fulfillment of adolescents’ basic psychological needs.

However, the translation of environmental support into psychological satisfaction and subsequent behavior may not be uniform across all contexts. In China, academic stress represents the predominant source of psychosocial strain for adolescents (Sun et al. 2013). To understand the role of this stressor, we integrate Conservation of Resources Theory (COR), which suggests that individuals strive to retain and protect resources, and that stress ensues from the threat or actual loss of these resources (Hobfoll 1989). We propose a novel “dual-stage” moderation model. In the first stage (Environment Needs), we posit a “Resource Depletion” effect: coping with high academic stress consumes significant cognitive and emotional resources, potentially inducing a narrowing of attention that limits adolescents’ ability to perceive positive environmental cues, thereby weakening the association between family support and needs satisfaction (Wang et al. 2019). Conversely, in the second stage (Needs Behavior), we posit a “Compensatory Coping” effect. Physical activity is a proven buffer against stress (Li et al. 2021). For adolescents whose psychological needs are met, high academic stress may paradoxically increase the utility of MVPA as a necessary release valve, strengthening the association between needs satisfaction and MVPA compared to their low-stress peers.

By integrating SDT and COR theories, this study aims to construct a comprehensive theoretical explanatory model. We hypothesize that: (1) Family sports environment is positively associated with adolescent MVPA; (2) Basic psychological needs satisfaction theoretically links this relationship; and (3) Academic stress may play a nuanced regulatory role by negatively moderating the path from environment to needs satisfaction but positively moderating the path from needs satisfaction to MVPA. This investigation utilizes objective accelerometry to provide robust empirical evidence that is consistent with hypothesized psychological processes.

Methods

Participants and procedure

This study employed a multi-stage hybrid sampling design (purposive selection of regions followed by stratified random sampling of schools) to ensure the representativeness of the sample across diverse socio-economic regions of China. In the first stage, three provinces were purposefully selected to represent Eastern (economically developed), Central (developing), and Western (underdeveloped) China. In the second stage, one middle school and one high school were randomly selected from each province, resulting in a total of six participating schools. In the third stage, classes were randomly selected from Grades 7 through 12 within these schools. Data collection was conducted between September and November 2024. All participants and their legal guardians provided written informed consent prior to participation. Adolescent participants also provided verbal assent, and confidentiality was strictly maintained throughout the process. The study protocol was approved by the Ethics Committee of the affiliated university.

Initially, 600 questionnaires and accelerometers were distributed. To ensure data quality and transparency, a rigorous screening process was employed, as illustrated in the participant flow diagram (Fig. 1). The following exclusion criteria were applied: (1) valid accelerometer wear time of less than 4 days (including at least one weekend day) or less than 10 hours per day; (2) significant missing data in questionnaires exceeding 20%; and (3) regular pattern responding. Consequently, a total of 77 participants were excluded from the initial pool: 42 participants due to insufficient accelerometer wear time, 23 due to incomplete questionnaire data, and 12 due to regular pattern responding. Based on the standard acceptable criteria for pediatric accelerometry (Migueles et al. 2017), requiring 4 valid days ensures a reliable estimate of habitual physical activity while retaining a representative sample. After data cleaning, the final analytical sample consisted of 523 adolescents, yielding an effective response rate of 87.2%. The sample comprised 271 males (51.8%) and 252 females (48.2%). Regarding educational level, 298 participants (57.0%) were middle school students, and 225 (43.0%) were high school students. The mean age of the participants was 15.23 years ().

Fig. 1.

Flowchart of participant recruitment and data screening process

Measures

Family sports environment

The assessment of Family Sports Environment was conducted using the Family Sports Environment Scale (FSES), adapted from Yao and Chen (2020) to suit the Chinese cultural context. The scale comprises 21 items distributed across three distinct subscales: (1) Physical Environment (7 items), assessing the availability of sports facilities and equipment at home; (2) Behavioral Environment (8 items), measuring parental modeling and participation; and (3) Psychological Environment (6 items), evaluating parental encouragement and support. Participants rated items on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Confirmatory Factor Analysis (CFA) supported the second-order structure of this scale in the current sample.

Basic psychological needs satisfaction

Measured using the Chinese version of the Basic Psychological Needs in Exercise Scale (BPNES) developed by Ng et al. (2011). This 12-item instrument assesses the satisfaction of Autonomy, Competence, and Relatedness within the physical activity domain. Given the high inter-correlations among the three subscales in our sample () and theoretical support for a global needs construct in SDT research (Standage & Ryan 2012), we aggregated the scores to form a single “Basic Psychological Needs Satisfaction” index. Responses were recorded on a 5-point Likert scale.

Academic stress

Assessed using the Adolescent Academic Stress Scale (Sun et al. 2013), a 16-item instrument designed to capture the multidimensional nature of academic pressure in China. The scale covers stressors related to heavy academic workload, high-stakes examination anxiety, parental expectations, and peer competition. A 5-point Likert scale was used, with higher scores indicating higher levels of perceived stress.

Physical activity (objective measure)

Physical activity was objectively monitored using the ActiGraph GT3X+ triaxial accelerometer (ActiGraph LLC, Pensacola, FL, USA). Participants were instructed to wear the device on their right hip via an elastic belt for seven consecutive days, removing it only during sleep, bathing, or swimming. Data were collected at a sampling rate of 30 Hz and analyzed using ActiLife 6.0 software. Non-wear time was defined as 60 consecutive minutes of zero counts (Troiano algorithm). A valid day was defined as at least 10 hours of wear time. Participants with at least 4 valid days (including at least one weekend day) were included in the analysis. The Evenson et al. (2008) cut-points, widely validated for adolescents, were used to classify activity intensity, with MVPA defined as activity counts 2296 counts per minute (CPM). The primary outcome variable was the average daily minutes of MVPA.

Covariates

To control for potential confounding factors, we included Gender (1 = Male, 2 = Female), Grade Level (1 = Middle School, 2 = High School), Body Mass Index (BMI, calculated from self-reported height and weight), and Parental Educational Level (measured on a 5-point scale from primary school to postgraduate).

Data analysis strategy

Statistical analyses were performed using IBM SPSS Statistics 26.0 and AMOS 24.0. The analysis proceeded in four steps: (1) Preliminary analysis including descriptive statistics and correlation analysis. To address the potential clustering effect of students nested within schools, we calculated the Intraclass Correlation Coefficients (ICCs) for the main variables. The ICCs were all below 0.05, indicating that the design effect was negligible, justifying the use of single-level analysis. (2) Assessment of the measurement model’s reliability and validity using Confirmatory Factor Analysis (CFA); Discriminant validity was further assessed using the Heterotrait-Monotrait ratio (HTMT). (3) Evaluation of Common Method Bias (CMB) using both Harman’s single-factor test and the CFA Marker Variable technique. (4) Testing the mediation model (Model 4) and the moderated mediation model (Model 59) using the PROCESS macro. To mitigate multicollinearity, all continuous predictors were mean-centered. Bootstrapping with 5,000 resamples was employed to generate 95% Confidence Intervals (CIs).

Results

Common method bias test and descriptive statistics

Given that data for the independent variable (Family Sports Environment), mediator (Basic Psychological Needs Satisfaction), and moderator (Academic Stress) were collected via self-report questionnaires, procedural remedies were employed during data collection to minimize potential biases (e.g., ensuring anonymity, balancing item wording). To statistically assess potential bias, two approaches were utilized. First, Harman’s single-factor test was conducted. Exploratory factor analysis extracted nine distinct factors with eigenvalues greater than 1, with the first unrotated factor accounting for only 26.4% of the total variance, well below the critical threshold of 40%. Second, to rigorously address Reviewer concerns, we employed the CFA Marker Variable technique. A theoretically unrelated marker variable (social desirability) was introduced into the model. The inclusion of method paths did not significantly improve model fit (CFI < 0.01, RMSEA < 0.01), nor did it substantially alter the correlations between substantive constructs. These results suggest that Common Method Bias (CMB) was not a pervasive issue in the current dataset.

Furthermore, to address the potential nesting of students within the six schools, we calculated the Intraclass Correlation Coefficient (ICC). The ICC for MVPA was 0.032, and for Basic Needs Satisfaction was 0.028. Since these values are below the conventional cutoff of 0.05, the design effect was deemed negligible, justifying the use of single-level regression analysis.

Descriptive statistics and the correlation matrix are presented in Table 1. The participants’ average daily MVPA was 44.12 minutes (), based on the revised inclusion criteria of at least 4 valid days. This remains notably below the WHO recommendation of 60 minutes. Regarding demographic variables, gender was significantly negatively correlated with MVPA (), suggesting that male adolescents engaged in higher levels of physical activity compared to their female counterparts. Additionally, grade level demonstrated a negative correlation with MVPA ().

Table 1.

Means, Standard Deviations, and Correlations (N = 523)

Variable	Mean	SD	Skew	Kurt	1	2	3	4	5	6	7
1. Gender	1.48	0.50	0.08	-2.00	-
2. Grade	1.43	0.50	0.28	-1.93	-0.03	-
3. Parent Edu	1.44	0.50	0.22	-1.95	-0.01	-0.04	-
4. Family Sports Env.	3.52	0.68	-0.45	0.12	0.09*	-0.10*	0.15**	-
5. Needs Satisfaction	3.46	0.72	-0.38	0.05	0.06	-0.13**	0.08	0.42***	-
6. Academic Stress	3.21	0.75	0.15	-0.21	-0.05	0.25***	-0.09*	-0.13**	-0.22***	-
7. MVPA (min/day)	42.35	15.62	0.67	0.45	-0.16**	-0.17***	0.11*	0.36***	0.40***	-0.18***	-

Gender: 1=Male, 2=Female; Grade: 1=Middle School, 2=High School; Parent Edu: 1=High School or below, 2=University or above

 * p < .05, ** p < .01, *** p < .001

Regarding the main study variables, Family Sports Environment exhibited a moderate-to-strong positive correlation with both Basic Psychological Needs Satisfaction () and MVPA (), providing initial support for the hypothesis that supportive home environments are linked to better psychological and behavioral outcomes. Consistent with theoretical expectations, Basic Psychological Needs Satisfaction was robustly correlated with MVPA (). Conversely, Academic Stress was negatively associated with Family Environment (), Needs Satisfaction (), and MVPA ().

Assessment of the measurement model

Before proceeding to hypothesis testing, the psychometric properties of the measurement scales were rigorously evaluated using Confirmatory Factor Analysis (CFA) in AMOS 24.0. As shown in Table 2, the standardized factor loadings for all items ranged from 0.71 to 0.88, exceeding the recommended threshold of 0.60. The measurement model yielded a good fit to the data (). The Composite Reliability (CR) values for all constructs ranged from 0.89 to 0.91, demonstrating excellent internal consistency. Furthermore, the Average Variance Extracted (AVE) values ranged from 0.56 to 0.68, all well above the cut-off value of 0.50. Discriminant validity was further confirmed using the Heterotrait-Monotrait ratio (HTMT), with all values below the conservative threshold of 0.85. These results collectively confirm the convergent and discriminant validity of the measurement model.

Table 2.

Measurement model assessment

Construct / Dimensions	Items	Std. Factor Loading () Range	Cronbach’s	CR	AVE
Family sports environment	21	0.71 – 0.86	0.88	0.89	0.58
Physical Environment	7	0.75 – 0.82
Behavioral Environment	8	0.71 – 0.85
Psychological Environment	6	0.78 – 0.86
Basic needs satisfaction	12	0.74 – 0.88	0.89	0.90	0.65
Autonomy	4	0.76 – 0.84
Competence	4	0.79 – 0.88
Relatedness	4	0.74 – 0.83
Academic stress	16	0.71 – 0.84	0.91	0.91	0.56
Workload	4	0.72 – 0.81
Exam Anxiety	4	0.75 – 0.84
Parental Expectation	4	0.73 – 0.80
Peer Competition	4	0.71 – 0.79

N = 523. CR Composite Reliability, AVE Average Variance Extracted. All factor loadings were significant at p < .001

Testing the mediation model

We first tested the simple mediation hypothesis using PROCESS Model 4, controlling for gender, grade level, parental education, and BMI. The analysis revealed that in the absence of the mediator, Family Sports Environment was significantly associated with MVPA (), establishing a significant total effect. When Basic Psychological Needs Satisfaction was introduced into the model, it served as a significant positive correlate of MVPA (). Importantly, while the direct association of Family Sports Environment with MVPA decreased, it remained statistically significant (). This pattern of results indicates a partial mediation effect.

To further verify the significance of this indirect pathway, a bootstrap analysis with 5,000 resamples was conducted. The results showed an indirect effect size of 0.106 (), with a 95% bias-corrected confidence interval of [0.068, 0.154]. Since this interval does not include zero, Hypothesis 2 is supported, suggesting that basic psychological needs satisfaction is a crucial transmission pathway linking family sports environment to adolescent physical activity.

Testing the moderated mediation model

We then tested the full moderated mediation model using PROCESS Model 59 to examine whether Academic Stress moderates the direct and indirect pathways. The regression results, controlling for gender, grade, parental education, and BMI, are detailed in Table 3 and visualized in Figure 2.

Table 3.

Regression Results for Moderated Mediation (Model 59)

Predictor	Model 1 (Outcome: Needs Satisfaction)			Model 2 (Outcome: MVPA)
		SE	95% CI		SE	95% CI
Control variables
Gender (1=M, 2=F)	0.03	0.04	[-0.05, 0.11]	-0.11**	0.04	[-0.19, -0.03]
Grade Level	-0.09*	0.04	[-0.17, -0.01]	-0.13**	0.04	[-0.21, -0.05]
Parental Edu	0.05	0.03	[-0.01, 0.11]	0.04	0.03	[-0.02, 0.10]
BMI	-0.04	0.03	[-0.10, 0.02]	-0.08*	0.03	[-0.14, -0.02]
Main Effects
Family Sports Env. (X)	0.38***	0.04	[0.30, 0.46]	0.17**	0.07	[0.04, 0.30]
Academic Stress (W)	-0.16***	0.04	[-0.24, -0.08]	-0.12*	0.05	[-0.22, -0.02]
Needs Satisfaction (M)	—	—	—	0.28***	0.05	[0.18, 0.38]
Interactions
Env.Stress (XW)	-0.14**	0.04	[-0.22, -0.06]	-0.05	0.06	[-0.17, 0.07]
NeedsStress (MW)	—	—	—	0.11*	0.05	[0.01, 0.21]
Model Summary
	0.24			0.29
F	32.45***			25.68***

X Family Sports Environment, W Academic Stress, M Needs Satisfaction

* p < .05, ** p < .01, *** p < .001

Fig. 2.

Statistical diagram of the moderated mediation model (Model 59) with standardized path coefficients

Analysis of Stage 1 Moderation (Environment Needs)

The regression analysis revealed a significant negative interaction between Family Sports Environment and Academic Stress on Basic Psychological Needs Satisfaction (). To interpret this interaction, simple slopes analysis was conducted (Figure 3). For adolescents reporting Low Academic Stress (Mean ), the family sports environment had a strong, positive association with needs satisfaction (). However, for those experiencing High Academic Stress (Mean ), this positive relationship was significantly attenuated (). These findings are consistent with the resource depletion hypothesis, suggesting that high levels of academic stress may limit the capacity to perceive and utilize family support.

Fig. 3.

The moderating effect of academic stress on the relationship between family sports environment and needs satisfaction

Analysis of Stage 2 Moderation (Needs MVPA)

Conversely, a significant positive interaction was observed between Basic Psychological Needs Satisfaction and Academic Stress on MVPA (). Simple slopes analysis (Figure 4) demonstrated that the relationship between needs satisfaction and MVPA was stronger for the High Academic Stress group () compared to the Low Academic Stress group (). This pattern supports the compensatory coping perspective, indicating that under conditions of high stress, adolescents whose psychological needs are met may be more strongly motivated to convert this psychological capital into physical activity.

Fig. 4.

The moderating effect of academic stress on the relationship between needs satisfaction and MVPA

Conditional indirect effects analysis

To provide a comprehensive view of the moderated mediation, we calculated the conditional indirect effects of Family Sports Environment on MVPA at different levels of Academic Stress. As presented in Table 4, the indirect effect remained statistically significant across all levels of academic stress.

Table 4.

Conditional Indirect Effects of Family Sports Environment on MVPA

Academic Stress Level	Effect	Boot SE	Boot LLCI	Boot ULCI
Low (-1 SD)	0.096	0.028	0.048	0.158
Mean (0)	0.106	0.022	0.068	0.154
High (+1 SD)	0.101	0.029	0.051	0.165
Index of Moderated Mediation	-0.023	0.015	-0.056	0.003

Crucially, the Index of Moderated Mediation was not statistically significant (Index = -0.023, 95% CI [-0.056, 0.003]). While a non-significant index typically suggests no moderated mediation, closer inspection of the component paths reveals a “suppression” effect. The negative moderation at Stage 1 (Resource Depletion) and the positive moderation at Stage 2 (Compensatory Coping) effectively counterbalance each other. Consequently, although the mechanism of transmission changes depending on stress levels, the magnitude of the total indirect effect remains remarkably stable (Effect sizes ranging from 0.096 to 0.106).

Discussion

This study represents an effort to integrate Self-Determination Theory (SDT) and Conservation of Resources Theory (COR) within a unified framework, utilizing objective accelerometry to dissect the complex associations governing adolescent physical activity in the high-pressure Chinese educational context. The findings corroborate the mediating role of basic psychological needs and, more significantly, unveil the nuanced regulatory function of academic stress. These insights offer a perspective on how environmental, psychological, and situational factors interact to relate to health behaviors.

Family environment as a psychological nutrient for physical activity

Our results confirm that a supportive family sports environment is a fundamental correlate of adolescent MVPA, with basic psychological needs satisfaction acting as a critical partial mediator. This finding aligns with the central tenets of SDT, which posit that social environments facilitate intrinsic motivation by fulfilling innate psychological nutriments (Ryan & Deci 2017). It is worth noting that we modeled the three basic needs—autonomy, competence, and relatedness—as a single composite construct. This decision was supported by high inter-correlations in our sample, a finding often observed in collectivist cultures where individual autonomy is closely linked to relational harmony (Standage & Ryan 2012). Thus, for Chinese adolescents, a "needs-satisfying" environment appears to be one that holistically supports their sense of volition, capability, and belonging. Beyond fulfilling basic needs, such environments may also contribute to the accumulation of broader psychological resources. For instance, recent studies have demonstrated that physical activity is significantly and positively correlated with internal resources such as self-esteem and self-efficacy (Du, Wang, et al. 2024). When adolescents perceive their home environment as supportive, they are more likely to internalize motivation, potentially strengthening their psychological capital and overall sense of competence.

Specifically, the availability of home sports equipment removes logistical barriers, thereby supporting autonomy. Simultaneously, parental modeling and verbal encouragement provide the scaffold for competence development and reinforce relatedness through shared values. Unlike previous studies that often focused narrowly on parental monitoring, our study highlights the holistic value of an “autonomy-supportive” family climate. When adolescents perceive their home environment as one that resources rather than controls their activity, their motivation becomes internalized, which is associated with sustained MVPA.

The resource depletion effect: why stress may weaken perception of support

A key contribution of this study is the identification of the negative moderating role of academic stress on the “Environment Needs” pathway. This finding provides empirical support for the resource depletion hypothesis derived from COR theory (Hobfoll 1989, 2001). In the hyper-competitive Chinese education system, adolescents are often in a chronic state of resource conservation. Our data suggests that this state may induce a “scarcity mindset” or cognitive tunneling. When overwhelmed by the demands of homework and exams, adolescents’ bandwidth to perceive and utilize family support appears to narrow. Even if parents provide a rich sports environment, the stressed adolescent may simply fail to notice these affordances or may misinterpret parental encouragement as additional pressure. This “stress barrier” may explain why well-intentioned family interventions sometimes yield diminishing returns during high-stakes exam periods.

The compensatory coping effect: the paradox of high-stress motivation

Perhaps the most intriguing finding is the positive moderating effect of academic stress on the “Needs Behavior” pathway. Contrary to the intuitive assumption that stress universally suppresses physical activity, our results indicate that for adolescents whose psychological needs are met, high stress is associated with an amplified drive to exercise. This dynamic process can be more explicitly understood through the lens of psychological resource transformation. According to the resource gain and protection cycles of COR theory, physical activity is not merely an energy expenditure but a catalyst for developing psychological resilience, emotional regulation, and stress tolerance. For high-stress adolescents, physical activity may function as a "resource reservoir." Recent evidence suggests that psychological resilience and emotional distress act as chain mediators in physical activity pathways (Yan, Liu, et al. 2026a), allowing individuals to transform physical movement into adaptive coping strategies. This is further supported by meta-analytic findings showing that mind-body exercises can significantly reduce negative psychological symptoms such as anxiety and depression (Du, Zhang, et al. 2022). Therefore, when basic psychological needs are satisfied, adolescents may more effectively deploy physical activity as a tool for emotional regulation, converting the "pressure gradient" of academic stress into a motivation for behavioral engagement..

Integrative dynamics: a homeostatic balancing effect

The simultaneous operation of the two opposing moderation effects reveals a complex balancing dynamic. This "homeostatic" effect is consistent with established stress-buffering frameworks, where physical activity is theoretically linked to the mitigation of adverse psychological impacts from high-stress environments. Just as physical activity has been shown to moderate the relationship between family conflict and adolescent anxiety (Yan, Liu, et al. 2026b), our findings suggest it may serve a similar protective function against academic pressure. Consequently, the indirect transmission of family influence remains statistically robust and stable even under high stress.

Crucially, the non-significant Index of Moderated Mediation reported in our results indicates that the total indirect effect of family environment on MVPA does not significantly differ across stress levels. Rather than viewing this as a null finding, we interpret it as a “homeostatic” effect: the diminished efficiency in translating family environment into psychological needs is effectively counterbalanced by the heightened urgency to utilize physical activity as a coping mechanism. Consequently, the indirect transmission of family influence remains statistically robust and stable even under high stress (as shown in Table 4). This suggests that while stress alters the pathway of motivation—shifting it from a “thriving” mode (environment-driven) to a “coping” mode (needs-driven)—it does not necessarily sever the link between family support and health behavior, provided that basic psychological needs are maintained above a critical threshold.

Practical implications

Based on these findings, we propose a differentiated intervention framework tailored to academic stress levels.

For Policy and Schools: Education policymakers must recognize that “reducing burden” (Shuangjian policy) is a prerequisite for environmental interventions to be fully effective. Simply building more sports facilities may be insufficient if students are too cognitively depleted to use them. Schools should integrate “stress-responsive” physical education curricula. During exam periods, PE classes should shift focus from skill evaluation (which consumes resources) to autonomy-supportive, stress-relieving activities (which restore resources).

For Families: Parents need to calibrate their support strategies. During low-stress periods (e.g., weekends, holidays), the focus should be on “Environmental Exposure”—maximizing access to equipment and co-participation to build habits. During high-stress periods, the strategy should shift to “Psychological Calibration.” Parents should explicitly frame exercise as a relief mechanism rather than a task, protecting the child’s sense of autonomy and avoiding pressure-filled language. Validating the child’s competence (“You are handling this well, let’s take a break”) may be more effective than forcing compliance.

Limitations and future directions

Despite its strengths, this study has several limitations that must be acknowledged.

First and foremost, the cross-sectional design of this study represents a significant limitation as it precludes any definitive causal or mechanistic inferences. While the moderated mediation framework provides a theoretically coherent explanation, the identified pathways must be interpreted as hypothesized psychological processes rather than empirically established mechanisms (Yan, Liu, et al. 2026b). As cross-sectional data cannot establish temporal ordering, the potential for bidirectional relationships—where active adolescents elicit more favorable parental support—cannot be ruled out.

Second, although objective accelerometry is a strength, our inclusion criterion of at least 4 valid days (including one weekend day), while consistent with pediatric standards, may still introduce selection bias. Adolescents who complied with the wear-time protocol might differ in conscientiousness or family support compared to those who were excluded. Additionally, accelerometers do not capture the context of activity (e.g., team sports vs. solo running), which might be differentially related to relatedness needs.

Third, while we employed the CFA Marker Variable technique to assess Common Method Bias, the reliance on self-reports for Family Environment, Needs Satisfaction, and Academic Stress means that shared method variance cannot be completely ruled out. Furthermore, although we controlled for parental education and BMI, unmeasured confounders such as household income or school-level sports policies could still influence the results.

Finally, although the Intraclass Correlation Coefficients (ICCs) suggested negligible clustering effects, the sample was drawn from only six schools. This limited number of clusters precluded the use of multilevel modeling, which would be the ideal statistical approach. Future studies should aim for a larger number of school clusters to robustly account for the nested structure of the data.

Conclusion

This study explores a theoretically grounded associations model, elucidating the potential pathways through which family sports environment is associated with adolescent physical activity. We conclude that basic psychological needs satisfaction is a key correlate linking family support to active behavior. Crucially, academic stress is observed as a nuanced regulator associated with both resource depletion and compensatory coping. These opposing forces effectively counterbalance each other, indicating that family support remains a viable correlate of physical activity even in high-stress contexts. These findings challenge the binary view of stress and underscore the necessity of interventions that not only enrich the physical environment but also actively protect the adolescent’s psychological reservoir of resilience and self-regulation.

Authors’ contributions

W.L. and D.M. conceptualized the study and designed the methodology. W.L. and C.H. carried out the data collection and performed the statistical analysis. W.L. wrote the main manuscript text and prepared the figures and tables. H.J. contributed to the critical theoretical reframing of the study and the integration of the psychological resource transformation framework during the revision process. D.M. provided supervision and critically reviewed and edited the manuscript. All authors have read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki. The research protocol was approved by the Ethics Committee of Tianjin Medical University (Approval No. TMU-REC-2024-068). Informed written consent was obtained from all legal guardians of the adolescent participants, and verbal assent was provided by the adolescent participants themselves. All study procedures were performed with the full understanding and cooperation of the participants.

Competing interests

The authors declare no competing interests.