Abstract
Background
Future anxiety may be relevant to academic burnout among university students, but its short-term temporally ordered association with burnout and the cognitive processes involved remain unclear. This study tested whether brooding rumination mediated this association and whether physical activity habit automaticity moderated the first stage.
Methods
A short-term three-wave longitudinal questionnaire survey was conducted among 1271 university students recruited from five provinces in China. Future anxiety and physical activity habit automaticity were measured at Time 1, brooding rumination at Time 2, and academic burnout at Time 3. Pearson correlations, time-lagged mediation, first-stage moderated mediation, simple slope tests, and sensitivity analyses were conducted using SPSS 26.0 and PROCESS 4.2.
Results
Future anxiety was positively associated with later brooding rumination (B = 0.275, p < 0.001), and brooding rumination was positively associated with later academic burnout (B = 0.432, p < 0.001). The indirect association from future anxiety to academic burnout through brooding rumination was significant (effect = 0.119, 95% CI [0.092, 0.147]). Physical activity habit automaticity moderated the association between future anxiety and brooding rumination (B = -0.072, p = 0.019), but the increase in explained variance was very small (ΔR² = 0.0037). The index of moderated mediation was significant (index = -0.031, 95% CI [-0.055, -0.007]).
Conclusions
Brooding rumination statistically accounted for part of the short-term time-lagged association between future anxiety and academic burnout. Physical activity habit automaticity showed a statistically significant but very small moderating pattern. Because the study was observational, the follow-up period was short, and autoregressive controls for brooding rumination and academic burnout were unavailable, the findings should be interpreted as time-lagged statistical associations rather than causal or intervention effects.
Keywords: Future anxiety, Brooding rumination, Academic burnout, Physical activity habit automaticity, Short-term longitudinal study, Moderated mediation, University students
Background
Future anxiety and academic burnout
University students are in a developmental period characterized by academic transition, career planning, interpersonal adjustment, and increasing responsibility for future life choices. Mental health problems are common in this population, and stress-related symptoms often cluster with academic and functional impairment [1]. Studies of college students have also shown that anxiety, depression, and perceived stress are closely tied to academic and social stressors [2]. Large international college-student initiatives further suggest that academic, financial, and future-oriented stressors are central correlates of student mental health [3]. Against this background, future anxiety is a relevant construct for understanding why some students show sustained cognitive distress and academic exhaustion over time.
Future anxiety refers to apprehensive and negatively toned anticipation of personal or broader future events. The construct was originally introduced to capture uncertainty and anxiety directed toward possible future outcomes rather than immediate threats [4]. The Dark Future Scale later provided a concise instrument for assessing this tendency in survey-based studies. Anticipatory stress can continue to activate psychological and physiological systems before a concrete stressor occurs [5]. Conceptually, future anxiety should be distinguished from general anxiety because it is defined by its anticipatory orientation, uncertainty, and cognitive elaboration of possible negative future scenarios. Whereas general anxiety may involve diffuse threat sensitivity and physiological arousal, future anxiety specifically concerns imagined future outcomes that are personally consequential but difficult to resolve in the present. This anticipatory and ambiguous nature may make future anxiety particularly likely to elicit repetitive negative thinking, because students may repeatedly attempt to mentally simulate, predict, or control uncertain academic and career outcomes. Recent longitudinal evidence suggests that future-oriented self-efficacy develops in connection with family and school support [6]. Related evidence links future time perspective and emotional problems with adolescent academic burnout [7]. Symptom-network and digital-behavior studies further indicate that internalizing and future-oriented cognitions are relevant in Chinese adolescent samples [8, 9].
The mediating role of brooding rumination
Brooding rumination may explain why future anxiety becomes academically consequential. Response styles theory argues that repetitive attention to the causes and consequences of distress can prolong negative mood and interfere with adaptive problem solving [10]. The brooding component of rumination captures passive comparison between one’s current situation and an unattained standard, making it especially relevant to future-oriented concerns. Broader reviews of repetitive thought also suggest that perseverative negative cognition can impair problem solving and emotional recovery [11]. Students who repeatedly dwell on uncertain futures may therefore find it more difficult to disengage from threat-related thoughts. Recent network evidence among adolescents with traumatic experiences also suggests that brooding is a maladaptive cognitive pattern that can be linked to poorer resilience-related processes [12].
Empirical findings have extended rumination beyond depression-specific contexts. Rumination has been identified as a transdiagnostic process associated with both depressive and anxiety symptoms [13]. Repetitive negative thinking has also been proposed as a shared cognitive process across emotional disorders [14]. Factor-analytic evidence supports the distinction between brooding and reflection, showing that brooding is especially linked to maladaptive psychological states [15]. These findings suggest that brooding rumination may operate as a cognitive bridge between future-oriented threat anticipation and subsequent burnout-related academic difficulties.
Academic burnout and short-term temporally ordered associations
Academic burnout reflects exhaustion, cynicism toward schoolwork, and feelings of inadequacy in academic contexts. Burnout theory emphasizes that prolonged demands can deplete psychological resources and lead to sustained disengagement [16]. The School Burnout Inventory conceptualizes school-related burnout as a multidimensional academic syndrome rather than a momentary mood state. Research among university students has shown that burnout-related processes are closely related to academic engagement and cross-cultural student adjustment [17]. Life stress has also been linked to academic burnout, supporting the view that sustained stress exposure can become embedded in students’ academic functioning [18]. Within a demands-resources framework, school burnout and engagement may develop as opposite responses to academic demands and available resources [19].
From a time-lagged perspective, brooding rumination is a plausible mediator because it may translate future-oriented stress into later academic depletion. Research on medical and university students has shown that academic stress and psychological distress can accumulate into burnout-relevant outcomes [20]. Conservation of resources theory provides the main theoretical basis for this pathway: students who repeatedly invest cognitive resources in anticipated threats may have fewer resources available for recovery, learning engagement, and adaptive coping [21, 22]. Contemporary resource-loss perspectives further suggest that resource loss can spiral when individuals have difficulty interrupting stress-related processes. This interpretation is also consistent with the longitudinal pattern discussed above, in which future-oriented concerns and emotional problems were related to adolescent academic burnout.
The moderating role of physical activity habit automaticity
Within this focused resource-based framework, physical activity habit automaticity may function as a boundary condition rather than a separate explanatory theory. The Self-Report Behavioural Automaticity Index measures the automaticity component of habit and has been validated as a concise habit measure. Earlier habit research showed that repeated behavior can become automatically cued by context and less dependent on effortful intention [23]. Habit formation studies further indicate that automaticity develops gradually through repeated performance in stable contexts [24]. In the physical activity domain, automatic habits may help students maintain activity even when academic pressure or future uncertainty reduces motivation.
Habit-based accounts are particularly relevant because students often know that physical activity is beneficial but do not always act consistently. Reviews of health behavior habit suggest that automaticity may be an active ingredient linking past behavior with future behavior [25]. Broader analyses of habit in health behavior also caution that habit should not be reduced to frequency alone, because automatic cue-response tendencies are conceptually distinct from how often a behavior occurs [26]. Work on non-conscious processes in health behavior further supports the idea that automatic processes can shape activity patterns independently of deliberate intentions [27]. This distinction is important in the present study because physical activity frequency was controlled as a covariate, allowing the model to focus on the automaticity of physical activity habits. Strong physical activity habit automaticity may attenuate the transition from future anxiety to brooding rumination by reducing the need for effortful decision-making, providing a predictable daily rhythm, and shifting attention from abstract threat simulation to embodied routine-based action. Therefore, automaticity may be more relevant to the cognitive escalation from future anxiety to brooding than to the direct reduction of burnout itself.
Physical activity is also linked to mental health and student adjustment. Meta-analytic evidence shows that intention does not fully translate into physical activity behavior, underscoring the need to examine automatic processes in addition to conscious motivation [28]. Current global guidelines emphasize regular physical activity as a core health behavior across populations [29]. Reviews among young people indicate that physical activity is associated with better mental health outcomes [30]. Meta-meta-analytic work also suggests that physical activity interventions can have beneficial effects on depression and anxiety outcomes [31]. Prospective meta-analytic evidence has further linked higher physical activity with lower risk of later depression [32]. Physical activity has also been discussed as a potentially relevant behavior for anxiety-related outcomes [33]. Recent multinational longitudinal evidence further indicates that physical activity shows longitudinal associations with adolescent well-being, supporting the developmental relevance of activity-related routines [34].
The current study
The present study was guided by two primary theories: conservation of resources theory and response styles theory. Conservation of resources theory was used as the main framework for explaining why anticipatory future-related anxiety may deplete students’ cognitive and emotional resources and relate to later academic burnout. Response styles theory was used to explain why brooding rumination may carry this association over time, because passive and repetitive attention to distress can prolong negative affect and impair adaptive problem solving. To avoid an overly diffuse theoretical framework, physical activity habit automaticity was not introduced as a separate broad theory; instead, it was treated as a habit-based boundary condition within the resource framework. Specifically, stronger automaticity may help students maintain routinized activity under stress, thereby weakening the association between future anxiety and brooding. Accordingly, we proposed that future anxiety would show a short-term temporally ordered association with subsequent academic burnout through brooding rumination. We further proposed that physical activity habit automaticity would moderate the first stage of this indirect association, such that the association between future anxiety and brooding rumination would be weaker among students with stronger automatic physical activity habits.
Methods
Study design and participants
This study employed a short-term three-wave longitudinal questionnaire design. Participants were recruited from five provinces in China, although the sampling frame was not designed to be nationally representative. To preserve institutional anonymity during the review process, the specific province-level and institutional identifiers are not disclosed in the manuscript. Future anxiety and physical activity habit automaticity were assessed at Time 1, brooding rumination was assessed at Time 2, and academic burnout was assessed at Time 3. A self-report questionnaire survey was conducted in 2026. The three waves were administered sequentially, with Time 2 approximately three weeks after Time 1 and Time 3 approximately six to seven weeks after Time 2. The overall observation interval was approximately 10 weeks. Therefore, the design allowed temporal ordering among the study variables, but it may not fully capture the longer developmental process of academic burnout. This temporal ordering was used because longitudinal mediation models are more appropriate than cross-sectional mediation models when the research question concerns indirect associations over time [35]. The final analytic sample included 1271 students aged 18 to 26 years (M = 20.23, SD = 1.53); 583 were male (45.9%) and 688 were female (54.1%). Grade distribution was as follows: first year, 438 (34.5%); second year, 384 (30.2%); third year, 301 (23.7%); fourth year, 100 (7.9%); and master’s students or above, 48 (3.8%). The study was approved by the Ethics Committee of Xinjiang Normal University (approval number: XJNU2026LLSC53), and all participants provided informed consent before completing the survey.
Quality control and matching
Data screening was conducted before wave matching. Responses were excluded if participants did not provide consent, failed the attention-check item, completed the questionnaire faster than the minimum duration threshold, or submitted duplicate responses. The valid attention-check responses were 4 at Time 1, 3 at Time 2, and 5 at Time 3. The minimum duration thresholds were 82 s at Time 1, 53 s at Time 2, and 51 s at Time 3. For duplicate submissions, the last valid submission was retained. After quality control and anonymous matching across the three waves, 1271 students were included in the final analytic sample.
Measures
All core psychometric instruments used in this study were based on previously published scales rather than newly developed for the present study. Specifically, the Dark Future Scale, the Self-Report Behavioural Automaticity Index, the brooding subscale of the Ruminative Responses Scale, and the School Burnout Inventory were previously published instruments and are cited below. The physical-activity-specific wording of the Self-Report Behavioural Automaticity Index was adapted for the present study, but no new multi-item psychometric scale was developed. Physical activity frequency and employment/graduate-study stress were assessed as single-item covariates rather than newly developed scales.
Future anxiety was measured at Time 1 using the five-item Dark Future Scale, which was previously published and validated for assessing anxiety directed toward possible negative future outcomes [36]. The scale was not developed for the present study. The Chinese survey wording was prepared from the published items and checked by the research team before administration. Items were scored on a 7-point response scale, and mean scores were computed, with higher scores indicating greater future anxiety. No item reversal was applied in the present scoring. In the present sample, Cronbach’s alpha was 0.803.
Physical activity habit automaticity was measured at Time 1 using a four-item physical activity adaptation of the Self-Report Behavioural Automaticity Index, a previously published brief measure of habit automaticity [37]. The scale was not newly developed for the present study; the item wording was adapted to refer specifically to physical activity, but no new multi-item psychometric scale was developed. The Chinese survey wording was prepared from the published items and checked by the research team for clarity and content relevance before administration; no formal back-translation procedure was conducted. Items were scored on a 7-point response scale, and mean scores were computed, with higher scores indicating stronger automatic physical activity habits. No item reversal was applied in the present scoring. Cronbach’s alpha was 0.819. No additional pre-testing or independent validation of this adapted version was conducted before the main survey, which is acknowledged as a limitation.
Brooding rumination was measured at Time 2 using the five-item brooding subscale derived from the previously published Ruminative Responses Scale [38]. The subscale was not developed for the present study. The Chinese survey wording was prepared from the published items and checked by the research team before administration. Items were scored on a 4-point response scale, and mean scores were computed, with higher scores indicating stronger brooding rumination. No item reversal was applied in the present scoring. Cronbach’s alpha was 0.860.
Academic burnout was measured at Time 3 using the nine-item School Burnout Inventory, which was previously published and validated for assessing exhaustion, cynicism, and inadequacy in school contexts [39]. The scale was not developed for the present study. The Chinese survey wording was prepared from the published items and checked by the research team before administration. Items were scored on a 6-point response scale, and mean scores were computed, with higher scores indicating greater academic burnout. No item reversal was applied in the present scoring. Cronbach’s alpha was 0.904.
Gender, grade, and physical activity frequency were included as covariates in the main models. Age was examined in sensitivity analyses because age and grade were highly correlated (r = 0.855, p < 0.001), and grade was considered the more direct indicator of academic stage. Employment/graduate-study stress was also used in a sensitivity analysis. Physical activity frequency and employment/graduate-study stress were single-item covariates rather than multi-item psychometric scales. No new multi-item scale was developed for this study.
Statistical analysis
Descriptive statistics, reliability coefficients, attrition comparisons, Pearson correlations, short-term time-lagged mediation analyses, moderated mediation analyses, and sensitivity analyses were conducted using IBM SPSS Statistics 26.0 and PROCESS macro version 4.2. Mediation was tested using PROCESS Model 4, and first-stage moderated mediation was tested using PROCESS Model 7. Bias in cross-sectional mediation analysis was considered when interpreting the temporally ordered design [40]. Additional guidance on longitudinal mediation informed the cautious interpretation of short-term indirect associations over time [41]. First-stage moderated mediation was tested according to moderated-mediation recommendations [42]. The index of moderated mediation was interpreted following Hayes [43]. PROCESS-based regression models were implemented according to Hayes’ conditional process framework [44]. Simple slopes were probed at one standard deviation below and above the moderator mean following standard interaction procedures [45]. Regression coefficients were interpreted using conventional applied regression guidance [46]. Predictors involved in the interaction term were mean-centered before the product term was computed. Bootstrap confidence intervals were based on 5000 resamples. Because the interaction between future anxiety and physical activity habit automaticity was specified a priori as the only hypothesized interaction test, no formal multiple-comparison correction was applied. Because brooding rumination and academic burnout were not repeatedly measured at all waves, autoregressive effects of the mediator and outcome could not be controlled. Therefore, the models were designed to test short-term time-lagged statistical associations rather than strict causal longitudinal mediation.
Common method bias was assessed using Harman’s single-factor test and a single-factor confirmatory model. The interpretation of common method bias followed behavioral research recommendations [47]. Additional guidance on method bias in social science research was also considered [48]. Missing-data patterns were interpreted with reference to missing-data methodology [49]. Complete-item sensitivity analysis was considered in light of applied missing-data guidance [50]. Broader missing-data decisions were guided by applied missing-data analysis texts [51]. Confirmatory factor analysis was estimated using maximum-likelihood estimation in Python 3.11 and interpreted with reference to structural equation modeling guidance [52]. Additional CFA interpretation followed applied CFA guidance [53]. Reliability was interpreted in relation to general test theory [54]. Multivariate assumptions were evaluated using standard applied multivariate guidance [55].
Results
Sample flow and attrition analysis
As shown in Table 1, 2113 responses were obtained at Time 1. After quality control and wave matching, 1271 students were retained for the final analysis. Compared with participants lost after Time 1 cleaning, retained participants did not differ significantly in age, gender, grade, future anxiety, physical activity habit automaticity, physical activity frequency, or employment/graduate-study stress. These results suggested that baseline attrition bias was not evident in the measured variables.
Table 1.
Sample flow and attrition analysis
| Item | Result |
|---|---|
| T1 raw responses | 2113 |
| T1 valid after consent, attention check, duration threshold, and deduplication | 1988 |
| T2 valid after consent, attention check, duration threshold, and deduplication | 1734 |
| T3 valid after consent, attention check, duration threshold, and deduplication | 1356 |
| Three-wave matched analytic sample | 1271 |
| Age, years | M = 20.23, SD = 1.53; range = 18–26 |
| Gender | Male: 583 (45.9%); female: 688 (54.1%) |
| Grade | Year 1: 438 (34.5%); year 2: 384 (30.2%); year 3: 301 (23.7%); year 4: 100 (7.9%); master’s students or above: 48 (3.8%) |
| Attrition from cleaned T1 | 717 participants (36.1%) |
| Baseline continuous-variable attrition tests | Age p = 0.084; FA p = 0.240; PAH p = 0.515; PA frequency p = 0.120; stress p = 0.464 |
| Baseline categorical-variable attrition tests | Gender χ² = 0.016, p = 0.900; grade χ² = 2.564, p = 0.633 |
FA future anxiety, PAH physical activity habit automaticity
Measurement quality, common method bias, and descriptive statistics
The four study scales showed acceptable to excellent internal consistency. Harman’s single-factor test indicated that the first unrotated component accounted for 29.61% of the variance, below the commonly used 40% reference value. A four-factor CFA was estimated using item-complete cases (n = 1254) with maximum-likelihood estimation. The four-factor model showed good fit and fitted the data better than the three-factor, two-factor, and single-factor alternatives. These results did not indicate severe common method bias and provided supportive measurement evidence for the four focal constructs. The reliability coefficients and measurement-model results are presented in Table 2. Descriptive statistics and correlations are presented in Table 3.
Table 2.
Reliability and CFA evidence
| Measure/model | n | α or χ²/df | CFI | TLI | RMSEA | SRMR |
|---|---|---|---|---|---|---|
| Future anxiety (DFS) | 1271 | α = 0.803 | — | — | — | — |
| Physical activity habit automaticity (SRBAI-PA) | 1271 | α = 0.819 | — | — | — | — |
| Brooding rumination (BR) | 1271 | α = 0.860 | — | — | — | — |
| Academic burnout (SBI) | 1254 | α = 0.904 | — | — | — | — |
| Harman single-factor PCA | 1254 | First factor = 29.61%; eigenvalues > 1 = 4 | — | — | — | — |
| Four-factor model | 1254 | 1.180 | 0.997 | 0.996 | 0.012 | 0.022 |
| 3-factor: FA + BR combined | 1254 | 7.722 | 0.870 | 0.855 | 0.073 | 0.078 |
| 2-factor: FA + BR + AB combined | 1254 | 16.238 | 0.703 | 0.672 | 0.110 | 0.114 |
| Single-factor model | 1254 | 22.624 | 0.576 | 0.534 | 0.131 | 0.134 |
CFA and reliability for academic burnout were estimated using item-complete cases (n = 1254), whereas the main regression models used available-item mean scoring and a complete-item sensitivity analysis was also conducted. CFA was estimated using maximum-likelihood estimation in Python 3.11
CFA confirmatory factor analysis, RMSEA root mean square error of approximation, SRMR standardized root mean square residual
Table 3.
Descriptive statistics and correlations
| Variable | M | SD | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. Gender | 1.54 | 0.50 | — | ||||||||
| 2. Age | 20.23 | 1.53 | 0.017 | — | |||||||
| 3. Grade | 2.16 | 1.10 | 0.019 | 0.855*** | — | ||||||
| 4. PA freq | 2.76 | 0.95 | -0.037 | 0.035 | 0.023 | — | |||||
| 5. Stress | 2.84 | 0.84 | 0.035 | 0.107*** | 0.102*** | -0.046 | — | ||||
| 6. FA | 4.06 | 0.73 | 0.060* | 0.044 | 0.040 | -0.035 | 0.468*** | — | |||
| 7. PAH | 3.80 | 0.81 | -0.089** | 0.031 | 0.051 | 0.444*** | -0.030 | -0.050 | — | ||
| 8. BR | 2.52 | 0.71 | 0.039 | -0.017 | -0.029 | -0.139*** | 0.153*** | 0.286*** | -0.298*** | — | |
| 9. AB | 3.40 | 0.81 | 0.025 | -0.006 | -0.029 | -0.095*** | 0.115*** | 0.249*** | -0.197*** | 0.424*** | — |
Stress = employment/graduate-study stress. Gender was coded 1 = male, 2 = female. Grade was coded 1 = year 1, 2 = year 2, 3 = year 3, 4 = year 4, 5 = master’s students or above
PAH physical activity habit automaticity
* p < 0.05, ** p < 0.01, *** p < 0.001
Short-term time-lagged mediation analysis
The short-term time-lagged mediation analysis results are presented in Table 4. Future anxiety at Time 1 was significantly and positively associated with brooding rumination at Time 2 (B = 0.275, p < 0.001). Brooding rumination at Time 2 showed a short-term time-lagged association with academic burnout at Time 3 after controlling for future anxiety and covariates (B = 0.432, p < 0.001). The direct association between future anxiety and academic burnout remained significant (B = 0.157, p < 0.001), indicating partial mediation. The bootstrap indirect association was significant (effect = 0.119, 95% CI [0.092, 0.147]).
Table 4.
Time-lagged mediation analysis
| Equation/effect | Predictor/path | B/effect | SE/BootSE | β | t | p | 95% CI |
|---|---|---|---|---|---|---|---|
| M equation: BR | Future anxiety | 0.275 | 0.026 | 0.282 | 10.546 | < 0.001 | [0.223, 0.326] |
| Y equation: AB | Future anxiety | 0.157 | 0.029 | 0.141 | 5.353 | < 0.001 | [0.099, 0.214] |
| Y equation: AB | Brooding rumination | 0.432 | 0.030 | 0.378 | 14.268 | < 0.001 | [0.373, 0.491] |
| Total effect | Future anxiety | 0.275 | 0.030 | 0.248 | 9.110 | < 0.001 | [0.216, 0.334] |
| Indirect effect | FA → BR → AB | 0.119 | 0.014 | - | - | - | [0.092, 0.147] |
Covariates were gender, grade, and physical activity frequency. β = standardized coefficient
Moderated mediation analysis
The first-stage moderated mediation results are shown in Table 5. Physical activity habit automaticity was significantly and negatively associated with brooding rumination (B = -0.243, p < 0.001). The interaction between future anxiety and physical activity habit automaticity was also significant (B = -0.072, p = 0.019), indicating that physical activity habit automaticity moderated the association between future anxiety and brooding rumination. However, the increase in explained variance was small (ΔR² = 0.0037), equivalent to approximately 0.37% additional variance explained in brooding rumination, so the moderation effect should be interpreted cautiously.
Table 5.
Moderated mediation analysis
| Equation | Predictor | B | SE | β | t | p | 95% CI |
|---|---|---|---|---|---|---|---|
| BR equation | Future anxiety | 0.265 | 0.025 | 0.272 | 10.573 | < 0.001 | [0.216, 0.314] |
| BR equation | PAH | -0.243 | 0.025 | -0.278 | -9.660 | < 0.001 | [-0.293, -0.194] |
| BR equation | Future anxiety × PAH | -0.072 | 0.031 | -0.060 | -2.353 | 0.019 | [-0.132, -0.012] |
| BR equation | Gender | -0.002 | 0.037 | -0.002 | -0.068 | 0.946 | [-0.074, 0.069] |
| BR equation | Grade | -0.016 | 0.017 | -0.026 | -0.991 | 0.322 | [-0.049, 0.016] |
| BR equation | PA frequency | -0.004 | 0.021 | -0.006 | -0.206 | 0.837 | [-0.046, 0.037] |
| Model change | Interaction ΔR² | 0.0037 | - | - | 5.538 | 0.019 | - |
Predictors in the interaction were centered before product-term construction
PAH physical activity habit automaticity
The conditional indirect effects and the index of moderated mediation are presented in Table 6. The indirect association from future anxiety to academic burnout through brooding rumination was significant at low, mean, and high levels of physical activity habit automaticity. The indirect effect decreased from low PAH (effect = 0.140, 95% CI [0.106, 0.175]) to mean PAH (effect = 0.115, 95% CI [0.089, 0.141]) and high PAH (effect = 0.089, 95% CI [0.060, 0.120]). The index of moderated mediation was significant (index = -0.031, 95% CI [-0.055, -0.007]).
Table 6.
Conditional indirect effects
| Effect | PAH level/raw value | FA → BR slope | Conditional indirect effect/index | BootSE | 95% CI |
|---|---|---|---|---|---|
| Conditional indirect effect | Low (-1 SD) (2.995) | 0.324 | 0.140 | 0.018 | [0.106, 0.175] |
| Conditional indirect effect | Mean (3.804) | 0.265 | 0.115 | 0.013 | [0.089, 0.141] |
| Conditional indirect effect | High (+ 1 SD) (4.612) | 0.207 | 0.089 | 0.016 | [0.060, 0.120] |
| Index of moderated mediation | - | - | -0.031 | 0.012 | [-0.055, -0.007] |
Low and high PAH correspond to one standard deviation below and above the mean, respectively
PAH physical activity habit automaticity
The hypothesized three-wave conditional process model is presented in Fig. 1.
Fig. 1.
Short-term conditional process model. Coefficients are unstandardized estimates from PROCESS Model 7. The interaction term represents first-stage moderation of the path from future anxiety to brooding rumination. * p < 0.05, *** p < 0.001
The simple slope plot is shown in Fig. 2. Future anxiety was positively associated with brooding rumination at low PAH (B = 0.324, p < 0.001), mean PAH (B = 0.265, p < 0.001), and high PAH (B = 0.207, p < 0.001). The slope was weaker at higher automaticity, which was consistent with the negative interaction term in the moderated mediation model.
Fig. 2.
Simple slopes for future anxiety and brooding rumination. Lines show predicted brooding rumination at low and high physical activity habit automaticity. Simple slope tests were significant at low PAH (B = 0.324, p < 0.001), mean PAH (B = 0.265, p < 0.001), and high PAH (B = 0.207, p < 0.001), with a weaker slope at high PAH
Sensitivity analyses
Sensitivity analyses supported the robustness of the main findings. As shown in Table 7, the interaction remained significant when age was added to the main model, when employment/graduate-study stress was added, when physical activity habit automaticity was included as a direct predictor of academic burnout, and when the item-complete sample was used. Variance inflation factors were below 4.0, suggesting that multicollinearity was not severe.
Table 7.
Sensitivity analyses
| Model | N | FA × PAH B | p | ΔR² | Indirect effect | Indirect 95% CI | Index | Index 95% CI |
|---|---|---|---|---|---|---|---|---|
| Main model | 1271 | -0.072 | 0.019 | 0.0037 | 0.119 | [0.092, 0.147] | -0.031 | [-0.055, -0.007] |
| + age | 1271 | -0.072 | 0.019 | 0.0036 | 0.118 | [0.093, 0.146] | -0.031 | [-0.056, -0.007] |
| + stress | 1271 | -0.071 | 0.021 | 0.0035 | 0.114 | [0.086, 0.145] | -0.031 | [-0.055, -0.005] |
| + PAH in outcome | 1271 | -0.072 | 0.019 | 0.0037 | 0.112 | [0.087, 0.140] | -0.030 | [-0.053, -0.005] |
| Complete-item sample | 1254 | -0.071 | 0.022 | 0.0035 | 0.119 | [0.093, 0.148] | -0.031 | [-0.057, -0.005] |
Stress= employmenr/graduate-study stress. The main model adjusted for gender, grade, and physical activity frequency
PAH physical activity habit automaticity, PA_freq physical activity frequency
Discussion
This study examined the short-term time-lagged association between future anxiety and academic burnout among university students recruited from five provinces in China, as well as the mediating role of brooding rumination and the moderating role of physical activity habit automaticity. The findings supported a partial mediation model. Students with higher future anxiety at Time 1 reported higher brooding rumination at Time 2, and higher brooding rumination showed a short-term time-lagged association with higher academic burnout at Time 3. The direct association between future anxiety and academic burnout remained significant after brooding rumination was included, indicating that brooding rumination statistically accounted for part, but not all, of this short-term temporally ordered association.
The mediating role of brooding rumination is consistent with the view that future-oriented uncertainty may become more detrimental when students repeatedly focus on its causes, consequences, and possible negative outcomes. Future anxiety may keep students’ attention oriented toward uncertain academic and career scenarios. When this apprehension is accompanied by brooding rumination, students may become trapped in passive comparison and repetitive self-focused thinking. Such a cognitive pattern may consume psychological resources that would otherwise be available for academic recovery, planning, and problem solving. Therefore, brooding rumination can be understood as a plausible short-term temporally ordered cognitive association, rather than a confirmed causal mechanism, linking future-oriented uncertainty with subsequent academic burnout. This interpretation is also consistent with the aforementioned evidence identifying brooding as a maladaptive cognitive pattern in adolescent adjustment. The persistence of the direct association also suggests that future anxiety may be related to academic burnout through factors other than brooding rumination, such as sleep disturbance, reduced academic motivation, avoidance, social withdrawal, or digital-context coping processes.
The moderating result further showed that physical activity habit automaticity statistically moderated the association between future anxiety and brooding rumination. Students with stronger automatic physical activity habits showed a weaker positive association between future anxiety and subsequent brooding. This pattern suggests that automatic physical activity routines may provide a relatively stable self-regulatory context when students face future-related uncertainty. However, the effect size of the interaction was very small (ΔR² = 0.0037), indicating that the interaction explained only approximately 0.37% additional variance in brooding rumination. Thus, the result should be interpreted as a statistically detectable but very small boundary condition rather than evidence that physical activity habits substantively alter the cognitive consequences of future anxiety. Given the tiny incremental variance explained, this moderation effect is unlikely to be practically meaningful for intervention development when considered in isolation. Recent moderated-mediation research on resilience factors has similarly shown that buffering effects in adolescent affective-distress pathways should be interpreted cautiously when their magnitude is small [56].
The conditional indirect effects provide a more detailed interpretation of this pattern. The indirect association from future anxiety to academic burnout through brooding rumination was significant at low, mean, and high levels of physical activity habit automaticity, but the magnitude of this indirect association decreased as automaticity increased. This result resembles previous longitudinal moderated-mediation work among Chinese college students in which physical activity was considered as a factor in a psychological-risk association [57]. Comparable three-wave longitudinal research further suggests that mediation and moderation findings should be interpreted with attention to stable between-person dispositions as well as time-lagged associations. In addition, the aforementioned multinational longitudinal evidence suggests that physical activity may be relevant to adolescent well-being over time, although the present study focuses specifically on habit automaticity rather than activity volume.
This study contributes to the student mental health literature by focusing on future anxiety rather than general distress, thereby addressing a form of anticipatory stress that is particularly relevant to students facing academic, employment, and life-transition decisions. Identifying brooding rumination as a short-term temporally ordered mediator also extends work on repetitive negative thinking by showing how future-oriented anxiety may be statistically linked with subsequent academic burnout through a specific ruminative style. The moderation analysis further shifts attention from physical activity frequency alone to the automatic and routinized nature of physical activity behavior. Because physical activity frequency was controlled, the moderation result points to the habitual quality of physical activity rather than simply to how often students were active. Developmental-context research indicates that adolescent and student adjustment can be shaped by family, school, peer, and broader sociocultural contexts. Cross-cultural evidence on parent-adolescent warmth and internalizing symptoms further supports the need to interpret the present association within its institutional and regional setting [58].
Several practical implications can be drawn cautiously. Universities may help students identify repetitive future-oriented thinking and develop strategies for disengaging from unproductive brooding. Programs promoting physical activity may also benefit from habit-oriented components, such as stable cues, routine scheduling, and gradual integration of activity into daily life. However, these implications should be regarded as preliminary and hypothesis-generating. The present findings do not show that strengthening physical activity habit automaticity will necessarily reduce brooding rumination or academic burnout. Given the very small interaction effect, physical activity habit automaticity is unlikely to be a meaningful stand-alone intervention target for burnout prevention. Future intervention studies should test multi-component strategies that combine habit-based physical activity promotion with cognitive reappraisal, mindfulness, academic counseling, or stress-management components.
This study has several limitations. First and most critically, although the three-wave design provides stronger temporal ordering than a cross-sectional design, the data remain observational and cannot establish causal effects. The study did not include baseline measures of brooding rumination or academic burnout at each wave and therefore could not adjust for autoregressive effects. As a result, the findings should be interpreted as short-term time-lagged statistical associations rather than evidence of strict causal longitudinal mediation. Future longitudinal studies should measure the predictor, mediator, and outcome repeatedly across at least four waves or use latent growth, cross-lagged, or intensive longitudinal models to distinguish within-person change from stable between-person differences [59]. Network-based longitudinal designs may also help identify symptom-level dynamics that are not captured by conventional regression models. Second, the total follow-up period was approximately 10 weeks, which may be insufficient to capture the longer developmental process of academic burnout. Academic burnout is typically conceptualized as a syndrome that develops over extended exposure to academic demands, and the present short-term associations may reflect fluctuations in future anxiety, brooding rumination, and exhaustion-related experiences rather than the chronic formation of burnout. Longer follow-up periods are needed to examine how academic burnout develops and stabilizes over time. Third, although participants were recruited from five provinces in China, the sample was not nationally representative, and the specific province-level and institutional identifiers were not disclosed to preserve institutional anonymity during the review process. Only 3.8% of participants were master’s students or above. Therefore, the findings should not be generalized to all Chinese university students without caution, particularly to students from other geographic regions, institutional types, or postgraduate populations. Future research should also test whether the conditional process differs across academic stages, such as first-year versus fourth-year undergraduate students or undergraduate versus postgraduate students. The aforementioned multinational work on physical activity and well-being suggests that developmental adjustment processes may vary across geographic settings. Cross-cultural evidence on family-context differences and internalizing symptoms also supports the need for replication across broader relational and cultural contexts. Fourth, all variables were measured using self-report questionnaires, which may introduce reporting bias. Fifth, although the adapted SRBAI-PA showed acceptable internal consistency and supportive measurement evidence in the present sample, internal consistency alone does not establish content or structural validity. The physical-activity-specific adaptation was not independently validated before hypothesis testing, and future studies should further examine its psychometric properties in Chinese university samples. Finally, unmeasured variables such as sleep, socioeconomic background, clinical symptoms, and other academic stressors may have influenced the observed associations. Intervention or intensive longitudinal designs are needed to test whether strengthening physical activity habit automaticity can reduce brooding rumination or academic burnout over time.
Conclusions
This short-term three-wave longitudinal conditional process study found that brooding rumination statistically accounted for part of the short-term time-lagged association between future anxiety and academic burnout among university students. Physical activity habit automaticity showed a statistically significant but very small moderation of the association between future anxiety and brooding rumination, and the index of moderated mediation was significant. These findings suggest a short-term temporally ordered statistical association in which future anxiety was linked with subsequent academic burnout through brooding rumination, with the first stage of the association varying by physical activity habit automaticity. Because the study was observational, the follow-up period was short, and autoregressive effects for all core constructs were not controlled, the findings should not be interpreted as evidence of causal or intervention effects.
Acknowledgements
The authors thank the participating students for their time and cooperation.
Use of artificial intelligence-assisted tools
ChatGPT was used only for language polishing and formatting support.
Abbreviations
- AB
Academic burnout
- BR
Brooding rumination
- CFA
Confirmatory factor analysis
- CI
Confidence interval
- DFS
Dark Future Scale
- FA
Future anxiety
- PAH
Physical activity habit automaticity
- SBI
School Burnout Inventory
- SRBAI
Self-Report Behavioural Automaticity Index
Authors' contributions
YT contributed to conceptualization, investigation, and manuscript drafting. SL contributed to data management, statistical analysis, and manuscript revision. LZ contributed to supervision, funding acquisition, and critical revision of the manuscript. All authors read and approved the final manuscript.
Funding
This work was supported by the 2025 Autonomous Region Social Science Fund Project “Practical Strategies for Strengthening the Sense of Community for the Chinese Nation through Sports in Xinjiang” (Grant No. 2025BTY153).
Data availability
The research data are not publicly available. They may be obtained from the corresponding author upon reasonable request and with the corresponding author’s approval.
Declarations
Ethics approval and consent to participate
The study was approved by the Ethics Committee of Xinjiang Normal University (approval number: XJNU2026LLSC53). All procedures were conducted in accordance with the Declaration of Helsinki. All participants provided informed consent before participation.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Auerbach RP, Alonso J, Axinn WG, Cuijpers P, Ebert DD, Green JG, et al. Mental disorders among college students in the World Health Organization World Mental Health Surveys. Psychol Med. 2016;46(14):2955–70. 10.1017/S0033291716001665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Beiter R, Nash R, McCrady M, Rhoades D, Linscomb M, Clarahan M, et al. The prevalence and correlates of depression, anxiety, and stress in a sample of college students. J Affect Disord. 2015;173:90–6. 10.1016/j.jad.2014.10.054. [DOI] [PubMed] [Google Scholar]
- 3.Karyotaki E, Cuijpers P, Albor Y, Alonso J, Auerbach RP, Bantjes J, et al. Sources of stress and their associations with mental disorders among college students: results of the World Health Organization World Mental Health Surveys International College Student Initiative. Front Psychol. 2020;11:1759. 10.3389/fpsyg.2020.01759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zaleski Z. Future anxiety: concept, measurement, and preliminary research. Pers Individ Dif. 1996;21(2):165–74. 10.1016/0191-8869(96)00070-0. [Google Scholar]
- 5.Brosschot JF, Gerin W, Thayer JF. The perseverative cognition hypothesis: a review of worry, prolonged stress-related physiological activation, and health. J Psychosom Res. 2006;60(2):113–24. 10.1016/j.jpsychores.2005.06.074. [DOI] [PubMed] [Google Scholar]
- 6.Do MH, Cui Y, Hill NE, Liang B, Perella J. Adolescent self-efficacy and orientation about the future: longitudinal associations with family/school support and sense of purpose. J Res Adolesc. 2025;35(3):e70055. 10.1111/jora.70055. [DOI] [PubMed] [Google Scholar]
- 7.Zhao Y, Liu Y, Jin L, Hao H, Lyu H. Future time perspective and emotional problems mediated the relationship between family functioning and adolescent academic burnout: a two-wave longitudinal study. J Adolesc. 2025;97(8):2236–45. 10.1002/jad.70036. [DOI] [PubMed] [Google Scholar]
- 8.Zhao Y, Liang J, Qu D, He Y, Ren Y, Chi Y. Unraveling depressive symptom networks: a three-year longitudinal study among Chinese junior high school adolescents. J Res Adolesc. 2025;35(1):e13040. 10.1111/jora.13040. [DOI] [PubMed] [Google Scholar]
- 9.Li Y, Yang L, Lin Z, Huang M, Wu Y, Lin Z, et al. A digital Skinner box: the bidirectional longitudinal relationship between anxiety symptoms, delay of gratification, and short-form video addiction among Chinese adolescents. J Res Adolesc. 2025;35(4):e70098. 10.1111/jora.70098. [DOI] [PubMed] [Google Scholar]
- 10. Nolen-Hoeksema S, Wisco BE, Lyubomirsky S. Rethinking rumination. Perspect Psychol Sci. 2008;3(5):400–24. 10.1111/j.1745-6924.2008.00088.x. [DOI] [PubMed] [Google Scholar]
- 11.Watkins ER. Constructive and unconstructive repetitive thought. Psychol Bull. 2008;134(2):163–206. 10.1037/0033-2909.134.2.163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yang L, Liang Y, Huang Q, Wu Y, Guan Z, Ju K, et al. Network structure of rumination and resilience in adolescents with traumatic experiences: a comparison of PTSD and non-PTSD groups. Int J Soc Psychiatry. 2026;72(2):364–75. 10.1177/00207640251361657. [DOI] [PubMed] [Google Scholar]
- 13.McLaughlin KA, Nolen-Hoeksema S. Rumination as a transdiagnostic factor in depression and anxiety. Behav Res Ther. 2011;49(3):186–93. 10.1016/j.brat.2010.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ehring T, Watkins ER. Repetitive negative thinking as a transdiagnostic process. Int J Cogn Ther. 2008;1(3):192–205. 10.1521/ijct.2008.1.3.192. [Google Scholar]
- 15.Whitmer AJ, Gotlib IH. Brooding and reflection reconsidered: a factor analytic examination of rumination in currently depressed, formerly depressed, and never depressed individuals. Cognit Ther Res. 2011;35(2):99–107. 10.1007/s10608-011-9361-3. [Google Scholar]
- 16.Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev Psychol. 2001;52:397–422. 10.1146/annurev.psych.52.1.397. [DOI] [PubMed] [Google Scholar]
- 17.Schaufeli WB, Martinez IM, Pinto AM, Salanova M, Bakker AB. Burnout and engagement in university students: a cross-national study. J Cross Cult Psychol. 2002;33(5):464–81. 10.1177/0022022102033005003. [Google Scholar]
- 18.Lin SH, Huang YC. Life stress and academic burnout. Act Learn High Educ. 2014;15(1):77–90. 10.1177/1469787413514651. [Google Scholar]
- 19.Salmela-Aro K, Upadyaya K. School burnout and engagement in the context of demands-resources model. Br J Educ Psychol. 2014;84(1):137–51. 10.1111/bjep.12018. [DOI] [PubMed] [Google Scholar]
- 20.Dyrbye LN, Thomas MR, Shanafelt TD. Systematic review of depression, anxiety, and other indicators of psychological distress among U.S. and Canadian medical students. Acad Med. 2006;81(4):354–73. 10.1097/00001888-200604000-00009. [DOI] [PubMed] [Google Scholar]
- 21.Hobfoll SE. Conservation of resources: a new attempt at conceptualizing stress. Am Psychol. 1989;44(3):513–24. 10.1037/0003-066X.44.3.513. [DOI] [PubMed] [Google Scholar]
- 22.Hobfoll SE, Halbesleben J, Neveu JP, Westman M. Conservation of resources in the organizational context: the reality of resources and their consequences. Annu Rev Organ Psychol Organ Behav. 2018;5:103–28. 10.1146/annurev-orgpsych-032117-104640. [Google Scholar]
- 23.Verplanken B, Orbell S. Reflections on past behavior: a self-report index of habit strength. J Appl Soc Psychol. 2003;33(6):1313–30. 10.1111/j.1559-1816.2003.tb01951.x. [Google Scholar]
- 24.Lally P, van Jaarsveld CHM, Potts HWW, Wardle J. How are habits formed: modelling habit formation in the real world. Eur J Soc Psychol. 2010;40(6):998–1009. 10.1002/ejsp.674. [Google Scholar]
- 25.Gardner B, Lally P, Wardle J. Making health habitual: the psychology of habit-formation and general practice. Br J Gen Pract. 2012;62(605):664–6. 10.3399/bjgp12X659466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gardner B. A review and analysis of the use of ‘habit’ in understanding, predicting and influencing health-related behaviour. Health Psychol Rev. 2015;9(3):277–95. 10.1080/17437199.2013.876238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rebar AL, Dimmock JA, Jackson B, Rhodes RE, Kates A, Starling J, et al. A systematic review of the effects of non-conscious regulatory processes in physical activity. Health Psychol Rev. 2016;10(4):395–407. 10.1080/17437199.2016.1183505. [DOI] [PubMed] [Google Scholar]
- 28.Rhodes RE, de Bruijn GJ. How big is the physical activity intention-behaviour gap? A meta-analysis using the action control framework. Br J Health Psychol. 2013;18(2):296–309. 10.1111/bjhp.12032. [DOI] [PubMed] [Google Scholar]
- 29.Bull FC, Al-Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54(24):1451–62. 10.1136/bjsports-2020-102955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Biddle SJH, Asare M. Physical activity and mental health in children and adolescents: a review of reviews. Br J Sports Med. 2011;45(11):886–95. 10.1136/bjsports-2011-090185. [DOI] [PubMed] [Google Scholar]
- 31.Rebar AL, Stanton R, Geard D, Short C, Duncan MJ, Vandelanotte C. A meta-meta-analysis of the effect of physical activity on depression and anxiety in non-clinical adult populations. Health Psychol Rev. 2015;9(3):366–78. 10.1080/17437199.2015.1022901. [DOI] [PubMed] [Google Scholar]
- 32.Schuch FB, Vancampfort D, Firth J, Rosenbaum S, Ward PB, Silva ES, et al. Physical activity and incident depression: a meta-analysis of prospective cohort studies. Am J Psychiatry. 2018;175(7):631–48. 10.1176/appi.ajp.2018.17111194. [DOI] [PubMed] [Google Scholar]
- 33.Kandola A, Vancampfort D, Herring M, Rebar A, Hallgren M, Firth J, et al. Moving to beat anxiety: epidemiology and therapeutic issues with physical activity for anxiety. Curr Psychiatry Rep. 2018;20(8):63. 10.1007/s11920-018-0923-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Bertrand C, Steinberg L, Duell N, Di Giunta L, Dodge KA, Gurdal S, et al. Physical activity and two-year change in adolescent well-being in nine countries. J Res Adolesc. 2025;35(2):e70035. 10.1111/jora.70035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Cole DA, Maxwell SE. Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling. J Abnorm Psychol. 2003;112(4):558–77. 10.1037/0021-843X.112.4.558. [DOI] [PubMed] [Google Scholar]
- 36.Zaleski Z, Sobol-Kwapinska M, Przepiorka A, Meisner M. Development and validation of the Dark Future scale. Time Soc. 2019;28(1):107–23. 10.1177/0961463X16678257. [Google Scholar]
- 37.Gardner B, Abraham C, Lally P, de Bruijn GJ. Towards parsimony in habit measurement: testing the convergent and predictive validity of an automaticity subscale of the Self-Report Habit Index. Int J Behav Nutr Phys Act. 2012;9:102. 10.1186/1479-5868-9-102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Treynor W, Gonzalez R, Nolen-Hoeksema S. Rumination reconsidered: a psychometric analysis. Cognit Ther Res. 2003;27(3):247–59. 10.1023/A:1023910315561. [Google Scholar]
- 39.Salmela-Aro K, Kiuru N, Leskinen E, Nurmi JE. School Burnout Inventory (SBI): reliability and validity. Eur J Psychol Assess. 2009;25(1):48–57. 10.1027/1015-5759.25.1.48. [Google Scholar]
- 40. Maxwell SE, Cole DA. Bias in cross-sectional analyses of longitudinal mediation. Psychol Methods. 2007;12(1):23–44. 10.1037/1082-989X.12.1.23. [DOI] [PubMed] [Google Scholar]
- 41.Selig JP, Preacher KJ. Mediation models for longitudinal data in developmental research. Res Hum Dev. 2009;6(2–3):144–64. 10.1080/15427600902911247. [Google Scholar]
- 42.Preacher KJ, Rucker DD, Hayes AF. Addressing moderated mediation hypotheses: theory, methods, and prescriptions. Multivar Behav Res. 2007;42(1):185–227. 10.1080/00273170701341316. [DOI] [PubMed] [Google Scholar]
- 43.Hayes AF. An index and test of linear moderated mediation. Multivar Behav Res. 2015;50(1):1–22. 10.1080/00273171.2014.962683. [DOI] [PubMed] [Google Scholar]
- 44.Hayes AF. Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. 3rd ed. New York: Guilford Press; 2022. [Google Scholar]
- 45.Aiken LS, West SG. Multiple regression: testing and interpreting interactions. Newbury Park: Sage; 1991. [Google Scholar]
- 46.Cohen J, Cohen P, West SG, Aiken LS. Applied multiple regression/correlation analysis for the behavioral sciences. 3rd ed. Mahwah: Lawrence Erlbaum Associates; 2003. [Google Scholar]
- 47.Podsakoff PM, MacKenzie SB, Lee JY, Podsakoff NP. Common method biases in behavioral research: a critical review of the literature and recommended remedies. J Appl Psychol. 2003;88(5):879–903. 10.1037/0021-9010.88.5.879. [DOI] [PubMed] [Google Scholar]
- 48. Podsakoff PM, MacKenzie SB, Podsakoff NP. Sources of method bias in social science research and recommendations on how to control it. Annu Rev Psychol. 2012;63:539–69. 10.1146/annurev-psych-120710-100452. [DOI] [PubMed] [Google Scholar]
- 49.Little RJA. A test of missing completely at random for multivariate data with missing values. J Am Stat Assoc. 1988;83(404):1198–202. 10.1080/01621459.1988.10478722. [Google Scholar]
- 50.Graham JW. Missing data analysis: making it work in the real world. Annu Rev Psychol. 2009;60:549–76. 10.1146/annurev.psych.58.110405.085530. [DOI] [PubMed] [Google Scholar]
- 51.Enders CK. Applied missing data analysis. New York: Guilford Press; 2010. [Google Scholar]
- 52.Kline RB. Principles and practice of structural equation modeling. 5th ed. New York: Guilford Press; 2023. [Google Scholar]
- 53.Brown TA. Confirmatory factor analysis for applied research. 2nd ed. New York: Guilford Press; 2015. [Google Scholar]
- 54.McDonald RP. Test theory: a unified treatment. Mahwah: Lawrence Erlbaum Associates; 1999. [Google Scholar]
- 55.Tabachnick BG, Fidell LS, Ullman JB. Using multivariate statistics. 7th ed. New York: Pearson; 2019. [Google Scholar]
- 56.Zhou K, Zhu X, Lu C, Yang Y, Chen S. Adverse childhood experiences, basic psychological needs, and adolescent affective distress: revisiting the buffering role of resilience factors. J Res Adolesc. 2025;35(4):e70080. 10.1111/jora.70080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Jiang Z, Xu H, Li S, Liu Y, Jin Z, Li R, et al. Childhood maltreatment and depressive symptoms among Chinese college students: a moderated mediation model of adult attachment styles and physical activity. J Affect Disord. 2022;309:63–70. 10.1016/j.jad.2022.04.100. [DOI] [PubMed] [Google Scholar]
- 58.Esposito C, Miranda MC, Rothenberg WA, Skinner AT, Lansford JE, Gurdal S, et al. Parent-adolescent discrepancies in perceptions of parental warmth: cross-cultural differences and longitudinal associations with internalizing symptoms. J Res Adolesc. 2025;35(4):e70093. 10.1111/jora.70093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Curran PJ, Bauer DJ. The disaggregation of within-person and between-person effects in longitudinal models of change. Annu Rev Psychol. 2011;62:583–619. 10.1146/annurev.psych.093008.100356. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The research data are not publicly available. They may be obtained from the corresponding author upon reasonable request and with the corresponding author’s approval.