Beyond the Scroll: Unmasking the Simultaneous Role of Psychosomatic and Stressors Behind Social Media Users’ Disempowerment and Disengagement

Beyond the Scroll: Unmasking the Simultaneous Role of Psychosomatic and Stressors Behind Social Media Users’ Disempowerment and Disengagement | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Beyond the Scroll: Unmasking the Simultaneous Role of Psychosomatic and Stressors Behind Social Media Users’ Disempowerment and Disengagement Indra Cahaya Tresna, Sri Hartini, Novalia Rachmah Polytechnic This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8396588/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study examines the mechanisms underlying digital disengagement among social media users by testing an integrated Conservation of Resources (COR) Theory model. It investigates how both intrinsic stressors (social comparison, privacy concerns) and extrinsic stressors (social overload, information overload) activate dual psychological pathways disempowerment and psychosomatic tension leading to platform disengagement. A quantitative, cross-sectional study was conducted with N = 1,500 Indonesian social media users (tenure ≤ 10 years, age 18–35). Data were collected via online survey and analyzed using variance-based structural equation modeling (PLS-SEM) with SmartPLS 4.0. Measurement validity was established through confirmatory factor analysis, and mediation pathways were tested via bootstrapping (5,000 resamples). All eight hypothesized relationships received empirical support (p < .001). Intrinsic and extrinsic stressors demonstrated significant direct effects on both disempowerment and psychosomatic tension. Both mediators independently predicted digital disengagement through simultaneous pathways with nearly equivalent indirect effects (disempowerment = 0.706; psychosomatic tension = 0.712). The integrated model explained 67% of disengagement variance, substantially exceeding typical effect sizes in digital stress literature (R² = 0.35–0.50). This study provides the first empirical validation of COR Theory's cascading-effect mechanism across simultaneous psychological domains in social media contexts. It advances beyond compartmentalized frameworks by demonstrating cross-dimensional effects and establishing that resource depletion operates through parallel, equally potent mechanisms rather than single pathways. Biological sciences/Neuroscience Biological sciences/Psychology Social science/Psychology digital disengagement social media stress conservation of resources simultaneous mediation psychosomatic tension disempowerment Figures Figure 1 INTRODUCTION Approximately 5.07 billion individuals engage daily with social media (Data portal, 2024), demonstrating unprecedented digital connectivity. Extensive research documents that social media facilitates interpersonal communication, information sharing, and community building. However, this widespread engagement paradoxically correlates with increased psychological distress, including emotional exhaustion, depression, and anxiety (Malik et al., 2020 ; Mojtabai, 2024 ; Zhang et al., 2021 ). Social media stressors encompass intrinsic factors such as social comparison reducing self-efficacy and privacy concerns undermining autonomy and extrinsic factors, including social and information overload (Acquisti et al., 2015 ; Malik et al., 2020 ). This interconnection differentiates social media stress from other workplace or environmental stressors. Despite this field complexity, digital disengagement users' intentional withdrawal from platforms remains understudied and under-theorised. Paradoxically, one-third of social media users plan to reduce or cease usage altogether (Data portal, 2024), yet engagement-focused algorithmic systems amplify psychological stressors through information and social overload the 'paradox of connectivity' (Erdem et al., 2025 ; Lo, 2019 ). Emerging technologies, including the Metaverse and AI-driven content, will further increase cognitive and social stressors (Dwivedi et al., 2022 ). A critical empirical gap exists: extant literature compartmentalises intrinsic and extrinsic stressors, thereby bypassing simultaneous pathway activation and cross-dimensional interplay. Mediation conditions in pathway multiplication remain unexplored, limiting intervention frameworks. Furthermore, digital disengagement research remains predominantly Western-centric, restricting theoretical generalisation. Without systematic understanding of stressor mechanisms, capacity to predict technology's impact on stress remains limited. This gap constrains the theoretical basis for comprehensive psychosomatic and disempowerment interventions. Predictive capacity requires empirically understood relational mechanisms, particularly for mental health policy development. Geographic diversification of research is essential for validating the generalisation of stress and disengagement mechanisms beyond Western contexts. This study addresses these gaps by proposing and empirically validating an integrated Conservation of Resources (COR) Theory model through a quantitative, cross-sectional design with Indonesian social media users (N = 1,500; tenure ≤ 10 years; age 18–35). The model examines disempowerment and psychosomatic tension as interrelated mediating pathways, investigating: (1) whether intrinsic stressors (social comparison, privacy concerns) directly activate both pathways; (2) whether extrinsic stressors (social overload, information overload) directly activate both pathways; and (3) whether cross-dimensional interaction effects exist. Data are analysed via variance-based structural equation modelling (SmartPLS 4.0). The central research question is: How do different stressors impact digital disengagement through different mechanisms? Eight research questions guide this inquiry, examining direct effects, mediation pathways, and indirect effects. This study provides the first empirical validation of COR Theory's cascading-effect mechanism across simultaneous psychological domains in social media contexts. It advances beyond compartmentalised frameworks by demonstrating that resource depletion operates through parallel, equally potent mechanisms rather than single pathways. Additionally, the Indonesian context contributes essential geographic representation to digital disengagement research, validating cross-cultural applicability. Findings inform technology designers, developers, clinicians, and policymakers seeking evidence-based interventions for platform-related stress and user withdrawal. LITERATURE REVIEW Conservation of Resources Theory COR Theory (Hobfoll, 1989) posits that people protect, maintain, and acquire resources both tangible (time, money) and psychological (self-efficacy, control). Stress occurs when resources are insufficient, lost, or threatened, triggering cascading effects across psychological systems. For example, privacy invasions threaten control/autonomy (disempowerment pathway) while draining emotional resources (psychosomatic pathway). Social media uniquely combines intrinsic stressors (social comparison, privacy concerns) and extrinsic stressors (information overload, social overload), necessitating complex stress theory. COR outperforms alternative frameworks on four dimensions: (1) recognizes multi-domain resource withdrawal simultaneously, unlike TAM and Social Presence Theory; (2) predicts cascading effects where one stressor activates multiple psychological systems; (3) emphasizes resource appraisal, aligning with stress models; and (4) demonstrates empirical support in occupational stress (Dhir et al., 2019; Malik et al., 2020), establishing applicability to other fields. The convergent threats to autonomy (Acquisti et al., 2015), social evaluation (Yue et al., 2021), and information processing (Dhir et al., 2019) justify COR's application to digital disengagement. A comparison of these theoretical frameworks is presented in Table 1. Table 1. Theoretical Framework Theoretical Framework Pathway Specificity Cross-Domain Effects Mediation Specification Disengagement Prediction TAM ((Davis, 1989) Single pathway No Not applicable No Social Presence Theory (Short et al., 1976) Device-specific Limited Not examined No UGT (Katz et al., 1973) Single pathway No Not specified Indirect only COR Theory (Hobfoll et al., 2018; Holmgreen et al., 2017) (Proposed) Dual simultaneous Yes (eksplisit) Simultaneous specification Yes (direct + indirect) CONCEPTUAL MODEL OVERVIEW The proposed integrated model defines 8 direct relationships and two dual-pathway mediation mechanisms. Two independent variable (IV) categories comprise four (4) stressor dimensions: (1) intrinsic stressors: social comparison and privacy concerns; and (2) extrinsic stressors: social overload and information (cognitive) overload. Two mediating variables (MV) capture distinct resource-depletion manifestations: (1) disempowerment (a loss of control appraisals and decreased perceived autonomy), and (2) psychosomatic tension (which includes fatigue and anxiety, and reflects affective and physiological stress manifestations). One dependent variable (DV) operationalises the behavioural outcome: Digital disengagement reflects users' intentional withdrawal from activities on the platform and is operationalised as disengagement in the model. The model aims to test if both mediating pathways act in parallel, if there are cross-dimensional effects (intrinsic stressors → psychosomatic tension; extrinsic stressors → disempowerment) and if both mediators do so in isolation in predicting disengagement. HYPOTHESIS DEVELOPMENT Direct Effects of Intrinsic Stressors on Disempowerment Social comparison and privacy breaches activate loss-of-control appraisals by diminishing self-efficacy and autonomy (Acquisti et al., 2015; Yue et al., 2021). These autonomy-threatening stressors erode psychological resources central to disempowerment. COR Theory predicts that such resource challenges prompt greater loss-of-control appraisals. (H1): Social comparison and privacy concerns directly increase users' disempowerment. Direct Effects of Intrinsic Stressors on Psychosomatic Tension Social comparison and privacy concerns activate psychosomatic tension through distinct mechanisms: mental exhaustion and rumination from social evaluation, and anxiety from perceived privacy breaches (Acquisti et al., 2015; Van Der Schuur et al., 2019). These mechanisms manifest affectively-physiologically (fatigue, anxiety, sleep disruption) and operate independently from disempowerment's control-appraisal pathway. This differentiation aligns with theoretical frameworks proposing stressor-specific activation pathways (Dhir et al., 2019). (H2): Social comparison and privacy concerns directly increase psychosomatic tension. Direct Effects of Extrinsic Stressors on Psychosomatic Tension Social overload and information overload are extrinsic stressors which primarily threaten cognitive and affective resources. Information overload occurs when incoming information exceeds the individual's ability to process such information, resulting in the need to intensify sustained attention and make decisions (Malik et al., 2020; Pang & Zhang, 2024). The depletion of one’s cognitive resources and attention results in mental fatigue and social overload. Social overload refers to excessive social commitments, interpersonal demands, and maintenance of social relations which draw on emotional resources through empathic and social performance (Malik et al., 2020; Pang & Zhang, 2024). Both mechanisms impacting stress and fatigue affect the emotional and physiological systems. Information overload and social overload are linked to psychosomatic symptoms of stress (Jie et al., 2023; Van Der Schuur et al., 2019). According to COR Theory (Hobfoll, 1989), the depletion of resources manifests in the individual psychosomatically as stress. External stressors predict psychosomatic tension. (H3): Social overload and information overload directly increase psychosomatic tension. Direct Effects of Extrinsic Stressors on Disempowerment Although extrinsic stressors mainly burden the cognitive-emotional dimension, COR Theory's cascading-effect principle predicts cross-dimensional impact. Information overload diminishes decision-making and understanding, and, consequently, users' feelings of mastery and control over the platform's navigation and information handling (Dhir et al., 2019; Malik et al., 2020). Social overload also erodes agency. Excessive social demands surpass the capacity for selective participation and lead to feelings of engagement and control being forced upon (Erdem et al., 2025; Pang & Zhang, 2024). Qualitative investigations of withdrawal from platforms describe users’ sentiments of having been “overwhelmed,” “unable to keep up,” which pertains to appraised resource loss, exhaustion, and control (Dhir et al., 2019; Malik et al., 2020). Thus, disempowerment stemmed from extrinsic stressors is the consequence of cross-dimensional resource loss. (H4): Social overload and information overload directly increase disempowerment. Direct Effects of Mediators on Digital Disengagement Disempowerment and perceptions of reduced control and lack of autonomy evoke a behavioural withdrawal response (Dhir et al., 2019; Hobfoll, 1989; Malik et al., 2020; Solomon & Corbit, 1974). Predictive of disengagement from environments that pose control challenges, frameworks of psychological reactance and learned helplessness describe sustained loss of control as disengagement(Dhir et al., 2019; Malik et al., 2020). The disengagement that stems from disempowerment gives rise to an inability to restore control and agency through participation and a sense of futility to be re-engaged. Empirical evidence overwhelmingly demonstrates that loss of control predicts the intention to stop using a given platform (Dhir et al., 2019; Malik et al., 2020) Hence, disempowerment is the cause of digital disengagement. (H5): Disempowerment directly increases digital disengagement. Psychosomatic tension manifests as stress, fatigue, and anxiety, from which individuals seek to escape through psychosomatic affective states. Negative reinforcement (avoidance) leads to the prediction of the individual avoiding social contexts that combine psychosomatic affect states. There is an escape from stress symptoms and a psychosomatic cost associated with prolonged exposure to social media (Solomon & Corbit, 1974). There is considerable evidence that stress and fatigue impact disengagement from social media platforms (Dhir et al., 2019; Song et al., 2014). This disconnects primary social media platforms from psychosomatic tension. (H6): Psychosomatic tension directly increases digital disengagement. Dual Mediation Pathways All four stressor dimensions (both intrinsic and extrinsic) activate disempowerment via control-loss mechanisms (H1, H4), resulting in disengagement (H5) in turn. COR Theory expects and posits that loss-of-control appraisals are at the very centre of how resource threats translate to behavioural changes. Assessing this mediation pathway captures the psychological control aspect of the stress response. (H7): Disempowerment mediates the relationships between intrinsic (social comparison, privacy concerns) and extrinsic stressors (social and information overload) and digital disengagement. All four dimensions of stressors activate psychosomatic tension via resource depletion (H2, H5), in turn resulting in disengagement through a negative reinforcement cycle (H6), and in doing so this helps to isolate the affective-physiological part of the stress response. Hypothesis 8 (H8): Psychosomatic tension mediates the relationships between intrinsic stressors (social comparison, privacy concerns); extrinsic stressors (social overload, information overload); and digital disengagement. Critical Integration: The Simultaneous Dual-Pathway Model The original contribution of this study rests on the simultaneous testing of both mediation pathways within the confines of a singular structural model. Previous studies only investigate and isolate one mediation path, which restricts the ability to assess: (1) the concurrent operation of both pathways; (2) the differing relative strengths of pathways; and (3) the extent to which cross-dimensional pathways (H2, H4) produce a meaningful contribution to the overarching effect. By estimating both pathways concurrently, the study aims to evaluate COR Theory’s fundamental premise of resource loss spanning multiple psychological domains through cascading effects occurring simultaneously and in the context of disengaging from social media. This approach allows for a richer specification of the integrated complexity of the mechanisms involved than is provided by the literature which has approached the subject in a piecemeal manner. The proposed conceptual model is illustrated in Figure 1. CONCEPTUAL MODEL THEORETICAL CONTRIBUTIONS The proposed model advances digital disengagement understanding in three ways. First, simultaneous multipath specification captures cascading, multidimensional resource depletion pathways simultaneously advancing beyond fragmented, single- pathway literature. Second, cross-dimensional hypotheses determine whether stressors operate in isolation or generally across psychological pathways, advancing theoretical and practical understanding. Third, mechanism complexity recognizes digital disengagement stems from centralized functioning of multiple mechanisms rather than singular pathways, better addressing psychological intricacy documented in platform withdrawal research. METHODOLOGY Research Design and Approach Quantitative cross-sectional correlational design enabling simultaneous variable measurement and complex multivariate relationship examination via SEM. Sample and Sampling Procedure Sample Specification: N = 1,500 active social media users. This total sample size is well above the power threshold for multi mediating SEM models. Considering the rule of thumb of 10 to 20 for the indicator count per the variance based structure of SEM, 1,500 provides more than ample power (> 0.90) to help test the hypothesized relationships for the eight relationships which would be small to moderate effects (Hair et al., 2016). An a priori power analysis (G*Power 3.1) was also conducted to check for sample size sufficiency for the multiple regression (primary effect sizes: β = 0.30 for direct effects, β = 0.25 for indirect effects); α = 0.05; power = 0.90); results indicated N of 147 should be sufficient to detect effects of moderate size (f^2 = 0.15) within regression contexts. For the N = 1,500 was well above the threshold, within, post hoc power estimated to be above 0.95 for all effects hypothesized, indirect effects to be small to moderate (0.15 < β < 0.20). Sampling Method: There was purposive sampling. Participants were specifically selected based on inclusion criteria which matched the study population: active users on social media currently experiencing symptoms related to platform stress. While purposive sampling made it possible for us to focus on theoretically pertinent users (current social media users), it came at a cost of potential selection bias. To measure the extent of bias, the respondents' characteristics (age, sex, platform) were compared to the social media demographic profile of Indonesians on a national level (Datar portal 2024). In addition, further procedures were implemented to deal with selection bias, and were also evaluated. The sample was weighted to correctly reflect the overall population relevant to the study, based on the differences in demographic characteristics. Sensitivity analyses were performed to assess the effect of some possibly unrepresented parts of the sample. The analyses established that the key characteristics were indeed represented in a sample, so that the issues related to selection bias were hardly a concern. Participant Eligibility Criteria: Platform Activity Status: Currently use social media (minimum weekly) User Tenure: Maximum 10 years active use (targeting contemporary users) Age Range: 18-35 years old Geographic Context: Indonesia Exclusion Criteria: More than 10 years tenure, outside 18-35 age range, no current engagement, or failed attention-check items. MEASUREMENT INSTRUMENTS Seven latent constructs were operationalised using multi-item scales. All scales employed 5-point Likert-type response anchors (1 = Strongly Disagree to 5 = Strongly Agree), with 50% of items reverse-coded to mitigate acquiescence bias. Reverse-coding implementation followed a systematic procedure: within each 6-item construct, 3 items were phrased in a direction-congruent manner (e.g., SC2: 'I compare my life to others on social media') while 3 items were phrased in an opposite semantic direction (e.g., SC5-reverse: 'I do not care how my life compares to others online'). Reverse-coded items were recoded prior to analyses such that higher scores consistently indicated higher construct levels. This balancing procedure reduces acquiescence threat (tendency to endorse items regardless of content) estimated in the literature to bias 15-25% of correlations when unidirectional items are used exclusively (Podsakoff et al., 2003). In the Indonesian context, we willingly justify the validity and reliability of the construct and measurement. This involved an elaborate process of adaptations of the scales which included translation, back translation, and verified procedures. Also, piloting the tests with the target population as the focus group helped ascertain reliability and comprehension of the measurement tools. It helped ensure that they achieved the desired cultural adaptation as measurement tools. Intrinsic Stressors - Social Comparison (SC) Over the years, several studies have examined social comparison, specifically the tendency to assess one’s worth through the presumed worth of others, using the Social Comparison Orientation Scale adapted for social media (Yue et al., 2021). Social media comparison presumably encompasses contradictory evaluation processes and affective consequences, as respondents may analyse themselves to a negative extreme and experience feelings of worthlessness, such as when observing others’ seemingly unnecessary advanced achievements (“I often compare my life to others’ lives on social media”). Such formats and evaluation processes were employed in the social media use within the adapted instrument, and three items measuring affective dimensions were reverse coded to mitigate response bias, where respondents with a low affective dimension of worth might exhibit a high-biased acute affective self-evaluation process in their answers. Previous studies have indicated that this adapted process produces a high biased affective dimension response in relation to social media, with a Cronbach’s alpha of .82. Intrinsic Stressors - Privacy Concerns (PC) For the purposes of this study, privacy-related concerns of the users in this study, (i.e., the Internet Users' Information Privacy Concerns scale developed by Smith et al., 1996) though tailored to online social media platforms, were utilised. The scale included six total items addressing privacy concerns (collection, control, and awareness). Three of the items were reverse scored. As per previous validation, the Cronbach's alpha was 0.79. Extrinsic Stressors - Social Overload (SO) Social overload has been defined as high social responsibilities and interaction overload on the platforms and was measured through the Social Overload Subscale of the Perceived Stress in Social Media Scale (Malik et al., 2020; Pang & Zhang, 2024). This five-item instrument assesses pressure to maintain relationships (''I feel pressure to keep in touch with everyone''); complexity of relationships (''Keeping social media relationships drains my energy''); and expectations of social interaction (''I feel obligated to respond to everyone''). Responses to two items were reversed. Cronbach's alpha coefficient was 0.81. Extrinsic Stressors - Information Overload (IO) The Information Overload Subscale of Malik et al. (2020); Dhir et al. (2019) Perceived Stress in Social Media Scale was used to measure the overload of information as cognitive strain due to the demands of processing and exposure to excessive information, to which three respondents did not provide answers. This five-item scale measures cognitive load, fragmentation of attention, and difficulty in filtering. Two items of which had been reverse scored. Cronbach's alpha was .77. Mediator - Disempowerment (DE) Disempowerment defined as loss-of-control appraisals and reduced perceived autonomy was measured using a construct adapted Perceived Control Scale (Dhir et al., 2019) and General Perceived Self-Efficacy Scale (Dhir et al., 2019; Zhang et al., 2021) specific to the use contexts of the platforms. The scale measured over six items coded as loss-of-control perceptions, lowered autonomy, and ineffectiveness. Three items were reverse coded. Previous research indicated the combined scale reliability to be 0.84. Mediator - Psychosomatic Tension (PT) Psychosomatic tension affective and physiological manifestations of stress, including fatigue, anxiety, and sleep disruption was measured using the Perceived Stress Scale - Social Media Version (Dhir et al., 2019). The six-item scale captures affective symptoms, physical manifestations, and sleep disruption. Three items received reverse-coding. Cronbach's alpha in published research = 0.86. Dependent Variable - Digital Disengagement (DD) Digital disengagement users' intentional reduction in platform use, content consumption, or interaction frequency was operationalized using the Platform Disengagement Intention Scale (Dhir et al., 2019; Yue et al., 2021) supplemented with behavioral indicators. The six-item scale measures disengagement intent, actual behavior reduction, and platform avoidance. Three items were reverse-coded. Cronbach's alpha = 0.88. Common Method Bias Mitigation Multiple procedural remedies were implemented to minimize common method bias (CMB): Temporal Separation: Predictor items presented in Section A; outcome variables in Section D (mean separation: ~12 survey items) Scale Mixing: Reverse-coded 50% of items across all constructs to prevent acquiescence patterns Psychological Separation: Item randomization within construct blocks to disrupt response patterns Anonymity Assurance: Explicit statement "Your responses will be completely anonymous and confidential" displayed at survey entry Reduced Cognitive Demand: Mean item completion time 12-18 minutes (adequate, not exhausting) Procedural Randomization: Item order randomized across 5 questionnaire versions distributed to different response cohorts Data Collection Procedure Platform: Google Forms online survey. Distribution: Purposive recruitment via social media communities, snowball sampling, and online forums. Timeline:43-week collection period. Data Quality Screening: Attention-check items, completion-time flags, and missing-data rules. Final N = 1,500 after screening 1,587 responses. Ethical Approval and Informed Consent This research was conducted in full accordance with the Declaration of Helsinki (1964, with subsequent amendments), which establishes fundamental ethical principles for research involving human subjects. All procedures and protocols were reviewed and formally approved by the Research Ethics Committee at Citra Buana Indonesia Institute, Sukabumi, Indonesia (Approval Reference: No.189/ETHICS/ICBI-E/VII/2025; Date of Approval: July 05, 2025). Informed consent was obtained from all 1,500 participants included in the final analyzed sample. At the survey entry, participants received an explicit consent statement confirming: (1) participation was entirely voluntary; (2) they could withdraw at any time without penalty; (3) responses would remain completely anonymous and confidential; (4) data would be stored securely; and (5) data would be used exclusively for research purposes. All participants confirmed their understanding and agreed to these terms before accessing the survey. No personally identifiable information was collected. Analytical Strategy Software: PLS-SEM via SmartPLS 4.0. Significance Threshold: α = 0.05 (two-tailed). Two-Stage Process: (1) Measurement model validation (λ > 0.70, CR > 0.70, AVE > 0.50, HTMT < 0.85); (2) Structural model testing via bootstrapping (5,000 resamples, 95% CI), with 5,000 resamples chosen to ensure statistical stability and precision of the indirect-effect estimates. This robust resampling approach provides greater confidence in the reliability of effect sizes compared to fewer resamples. Robustness Checks: Alternative model specification, multi-group analysis. RESULTS Sample Characteristics and Data Screening Data collection yielded 1,587 survey responses. Following systematic data quality screening, 87 responses were excluded: 15 responses failed embedded attention-check items, 28 responses were completed in 10% missing data. The final analyzed sample comprised N = 1,500 respondents, as detailed in Table 2. Table 2. Respondent Characteristic Variabels Gender Primary Platform Female 812 54.1% Instagram 634 42.3% Male 688 45.9% WhatsApp 472 31.5% Age Group Facebook 267 17.8% 18-24 years 465 31.0% X/Twitter 89 5.9% 25-29 years 612 40.8% TikTok 38 2.5% 30-35 years 423 28.2% User Tenure 1-3 years 312 20.8% 4-6 years 589 39.3% 7-10 years 599 39.9% 1-3 years 312 20.8% The sample comprised slightly more female (54.1%) than male respondents, with the modal age group 25-29 years (40.8%), consistent with the target demographic. Instagram and WhatsApp represented primary platforms for 73.8% of participants, reflecting current Indonesian social media adoption patterns. Mean user tenure was 6.8 years (SD = 2.9), confirming contemporary user experience consistent with inclusion criteria (maximum 10 years). Descriptive Statistics Table 3: Descriptive Statistics for All Study Variables Variable N M SD Min Max Skewness Kurtosis Social Comparison (SC) 1,500 3.28 0.84 1.17 5.00 -0.15 -0.38 Privacy Concerns (PC) 1,500 3.61 0.77 1.33 5.00 -0.32 0.22 Social Overload (SO) 1,500 3.45 0.82 1.20 5.00 -0.19 -0.18 Information Overload (IO) 1,500 3.71 0.79 1.40 5.00 -0.39 0.12 Disempowerment (DE) 1,500 3.18 0.89 1.00 5.00 -0.08 -0.62 Psychosomatic Tension (PT) 1,500 3.35 0.86 1.17 5.00 -0.24 -0.48 Digital Disengagement (DD) 1,500 3.52 0.83 1.17 5.00 -0.31 -0.35 confirming contemporary user experience consistent with inclusion criteria maximum 10 years. The descriptive statistics for all study variables are presented in Table 3. All variables demonstrated approximately normal distributions with skewness and kurtosis values within acceptable ranges (−1.0 to +1.0), supporting the suitability of parametric estimation procedures. Mean values clustered around 3.2-3.7 on the 5-point scale, indicating moderate-to-moderately-high endorsement of stressor experiences and disengagement intentions. Information Overload (M = 3.71, SD = 0.79) demonstrated the highest mean, suggesting this stressor is particularly salient for contemporary users. Social Comparison demonstrated the lowest mean (M = 3.28, SD = 0.84), though still indicating substantial experience. Measurement Model Results Structural equation modeling proceeded in two stages: measurement model evaluation followed by structural model testing. All measurement model criteria were satisfied confirming the valid operationalization of study constructs (see Table 4). Table 4. Measurement Model Construct Indicator Loading CR AVE Social Comparison SC1 0.78 0.89 0.67 SC2 0.81 SC3 0.82 SC4 0.74 SC5 0.79 SC6 0.80 Privacy Concerns PC1 0.79 0.87 0.62 PC2 0.82 PC3 0.81 PC4 0.76 PC5 0.77 PC6 0.73 Social Overload SO1 0.81 0.88 0.64 SO2 0.83 SO3 0.78 SO4 0.77 SO5 0.80 Information Overload IO1 0.82 0.90 0.68 IO2 0.84 IO3 0.81 IO4 0.80 IO5 0.79 Disempowerment DE1 0.83 0.91 0.70 DE2 0.86 DE3 0.81 DE4 0.84 DE5 0.79 DE6 0.82 Psychosomatic Tension PT1 0.85 0.92 0.72 PT2 0.87 PT3 0.83 PT4 0.81 PT5 0.84 PT6 0.80 Digital Disengagement DD1 0.84 0.93 0.73 DD2 0.86 DD3 0.82 DD4 0.83 DD5 0.81 DD6 0.79 Indicator Reliability: All factor loadings exceeded the 0.70 threshold, ranging from 0.73 to 0.87. The majority of loadings exceeded 0.80, indicating strong relationships between observed indicators and their respective latent constructs. No indicators required removal based on inadequate reliability. Internal Consistency Reliability: Composite Reliability (CR) values ranged from 0.87 to 0.93, substantially exceeding the 0.70 criterion. Digital Disengagement and Psychosomatic Tension demonstrated the highest reliability (CR = 0.93 and 0.912, respectively), indicating highly consistent multi-item measurement. Convergent Validity: Average Variance Extracted (AVE) values ranged from 0.62 to 0.73, all exceeding the 0.50 threshold. This indicates that latent constructs explain 62-73% of variance in their respective indicators, confirming that indicators validly measure their intended constructs. Table 5: Discriminant ValidityHeterotrait-Monotrait (HTMT) Correlations Variabel SC PC SO IO DE PT DD Social Comparison - Privacy Concerns 0.61 - Social Overload 0.57 0.60 - Information Overload 0.54 0.63 0.70 - Disempowerment 0.67 0.70 0.68 0.73 - Psychosomatic Tension 0.63 0.67 0.72 0.78 0.76 - Digital Disengagement 0.62 0.60 0.61 0.67 0.80 0.78 - All HTMT values remained below 0.85, the conservative threshold for discriminant validity. The highest HTMT ratio was observed between Information Overload and Psychosomatic Tension (0.78), which remained well below the cutoff. (Henseler et al., 2015) This pattern confirms that all seven constructs measure distinct phenomena rather than representing redundant dimensions. Discriminant validity was conclusively established. HTMT ratio represents heterotrait-monotrait correlation ratio; values <.85 indicate discriminant validity (constructs measure distinct phenomena). All HTMT values in Table 5 range from 0.54 to 0.80, substantially below the threshold, confirming discriminant validity Structural Model Results: Direct Effects Following measurement model validation, the structural model was estimated to test the eight hypothesized relationships. Path coefficients, standard errors, t-statistics, and 95% confidence intervals obtained via bootstrapping (5,000 resamples) are presented in Table 6. Table 6: Structural Model Results Direct Effects (H1-H6) Path Β SE t-value 95% CI p-value H1 SC, PC → DE 0.41 0.038 10.79 [0.34, 0.48] <.001 H2 SC, PC → PT 0.37 0.041 9.02 [0.29, 0.45] <.001 H3 SO, IO → PT 0.52 0.036 14.44 [0.45, 0.59] <.001 H4 SO, IO → DE 0.34 0.039 8.72 [0.27, 0.41] <.001 H5 DE → DD 0.47 0.032 14.69 [0.41, 0.53] <.001 H6 PT → DD 0.40 0.035 11.43 [0.33, 0.47] <.001 All direct hypotheses were supported at p < .001 significance level. The pattern of effect sizes reveals theoretically meaningful relationships, interpreted based on the framework in Table 7. H1-H2 (Intrinsic Stressors): Social comparison and privacy concerns significantly affected disempowerment (β = 0.41, p < .001) and psychosomatic tension (β = 0.37, p < .001), supporting cross-dimensional effects. The stronger disempowerment effect aligns with theory; however, substantial psychosomatic tension effects (β = 0.37) demonstrate affective stress manifestations. H3-H4 (Extrinsic Stressors): Social and information overload most strongly affected psychosomatic tension (β = 0.52, p < .001) versus disempowerment (β = 0.34, p < .001). Significant disempowerment effects (β = 0.34) confirm cross-dimensional prediction: information overload reduces perceived mastery, generating loss-of-control appraisals. H5-H6 (Mediators on Disengagement): Disempowerment (β = 0.47, p < .001) and psychosomatic tension (β = 0.40, p < .001) both significantly predicted disengagement. Disempowerment's stronger effect suggests loss-of-control appraisals constitute the more potent disengagement driver, though both mechanisms operate meaningfully. Table 7. Effect Size Interpretation Framework Effect Size (β or f²) Verbal Interpretation Research Context 0.01-0.05 Small Small but meaningful psychological effects 0.05-0.15 Small-to-Medium Typical psychological intervention effects 0.15-0.25 Medium Substantial psychological impact 0.25+ Large Large practical significance H1 (SC, PC→DE): β=0.41 large - intrinsic stressors substantial direct predictor of disempowerment H3 (SO, IO→PT): β=0.52 large - extrinsic stressors strongest predictor of psychosomatic tension H5 (DE→DD): β=0.47 large - disempowerment potent predictor of disengagement Indirect (Disempowerment pathway): 0.706 large - substantial mediation magnitude Structural Model Results: Mediation Effects Indirect effects were calculated to test whether disempowerment (H7) and psychosomatic tension (H8) mediate relationships between stressors and disengagement. Mediation was evaluated via bias-corrected bootstrapping with 5,000 resamples. The results for disempowerment and psychosomatic tension are summarized in Table 8 and Table 9, respectively. Effects were considered significant if 95% confidence intervals excluded zero and p < .05. Table 8: Mediation Effects via Disempowerment (H7) Stressor → Disempowerment → Disengagement Indirect Effect 95% CI Significance Social Comparison → Disempowerment → Digital Disengagement 0.193 0.141, 0.250 P <. 001 Privacy Concerns → Disempowerment → Digital Disengagement 0.193 0.141, 0.250 P <. 001 Social Overload → Disempowerment → Digital Disengagement 0.160 0.111, 0.214 P <. 001 Information Overload → Disempowerment → Digital Disengagement 0.160 0.111, 0.214 P <. 001 Total Mediation via Disempowerment 0.706 0.598, 0.820 P <. 001 Table 9: Mediation Effects via Psychosomatic Tension (H8) Stressor → Psychosomatic Tension → Disengagement Indirect Effect 95% CI Significance Social Comparison → Psychosomatic Tension → Digital Disengagement 0.148 0.104, 0.197 p < .001 Privacy Concerns → Psychosomatic Tension → Digital Disengagement 0.148 0.104, 0.197 p < .001 Social Overload → Psychosomatic Tension → Digital Disengagement 0.208 0.158, 0.263 p < .001 Information Overload → Psychosomatic Tension → Digital Disengagement 0.208 0.158, 0.263 p < .001 Total Mediation via Psychosomatic Tension 0.712 0.591, 0.840 p < .001 Both mediation pathways were fully supported. Notably, disempowerment mediation (total indirect effect = 0.706) and psychosomatic tension mediation (total indirect effect = 0.712) demonstrated comparable magnitude, supporting the theoretical prediction of simultaneous dual-pathway operation. Neither pathway overwhelmingly dominated; rather, both contributed substantially to understanding disengagement etiology. The near-equivalence (difference = 0.9%) provides strong evidence for COR Theory's cascading-effect mechanism. Model Predictive Accuracy Table 10: Model R² Values and Effect Sizes Endogenous Variable R² Interpretation Disempowerment 0.46 Moderate-to-strong; stressors explain 46% of DE variance Psychosomatic Tension 0.51 Strong; stressors explain 51% of PT variance Digital Disengagement 0.67 Strong-to-very-strong; stressors + mediators explain 67% of DD variance The structural model demonstrates substantial predictive accuracy, as detailed in Table 10. Stressors collectively explain 46% of disempowerment variance and 51% of psychosomatic tension variance, indicating that stress mechanisms are substantially, though not exclusively, determined by identified stressor categories. Critically, when both mediating pathways are included, the model accounts for 67% of the variance in digital disengagement. This suggests that nearly two-thirds of the reasons why young Indonesians disengage from social media platforms can be traced back to these stressors, providing a robust specification of disengagement etiology. This R² value substantially exceeds typical effect sizes in psychological disengagement research (typically R² = 0.35-0.50), suggesting theoretical and empirical advancement. Model Validation and Robustness Harman's test yielded 30.8% variance (< 50% threshold), confirming negligible common method variance. The dual-pathway model achieved superior fit (R² = 0.67 vs 0.52 single-pathway; +29% explained variance), validating psychosomatic tension's distinct contribution. Multi-group comparisons across age, gender, and platforms showed no significant between-group differences (permutation tests p > .05), confirming model generalizability. These convergent robustness checks establish validity and stability across demographic subgroups. DISCUSSION All eight hypothesized relationships received empirical support (N = 1,500 Indonesian users; tenure ≤10 years, age 18-35). Intrinsic and extrinsic stressors demonstrated significant direct effects on both disempowerment and psychosomatic tension. Both mediators independently predicted digital disengagement through simultaneous dual pathways with comparable magnitude (disempowerment indirect = 0.706; psychosomatic tension = 0.712). The integrated model explained 67% of disengagement variance, substantially exceeding literature standards (R² typically 0.35-0.50) COR Theory Validation and Theoretical Contributions The central theoretical contribution of this study resides in correlational validation of COR Theory's cascading-effect mechanism predictions. Cross-sectional evidence demonstrates associations consistent with cascading predictions whereby stressor exposure predicts simultaneously manifested disempowerment and psychosomatic strain. While these patterns align with COR Theory predictions regarding simultaneous multidimensional resource depletion, cross-sectional design precludes definitive causal inference. Longitudinal and experimental designs in future research will strengthen confidence regarding temporal causal mechanisms underlying these correlations. Prior research examining social media stress has operated under compartmentalized frameworks assuming unidirectional pathways. The present findings challenge this assumption through three mechanisms: Simultaneous Dual-Pathway Operation The nearly equivalent pathway magnitudes (disempowerment = 0.706, psychosomatic tension = 0.712, difference = 0.9%) provide robust evidence for COR Theory's prediction of simultaneous, parallel resource-depletion mechanisms rather than dominant single pathways. Importantly, the comparable magnitude (50.1% disempowerment vs 49.9% psychosomatic) indicates both mechanisms contribute equally to disengagement ethology neither can be dismissed as secondary or weaker driver. Cross-Dimensional Resource Cascade Effects Differential pathway strengths reveal meaningful stressor-mediator specificity within simultaneous operation: Intrinsic Stressors (Social Comparison, Privacy Concerns): Slightly stronger effects on disempowerment (β = 0.41) than psychosomatic tension (β = 0.37), with 10.8% differential. This pattern reflects theoretical expectation that autonomy-threatening stressors primarily activate loss-of-control appraisals. However, the substantial cross-dimensional effect on psychosomatic tension demonstrates cascading: threats to self-evaluation and autonomy resources consume emotional resources through sustained vigilance and rumination. Extrinsic Stressors (Social Overload, Information Overload): Substantially stronger effects on psychosomatic tension (β = 0.52) than disempowerment (β = 0.34), with 52.9% differential. Information and social overload directly deplete cognitive-emotional capacity, manifesting as fatigue. Yet the significant cross-dimensional effect on disempowerment (β = 0.34) demonstrates that overwhelming information environments reduce perceived mastery, and excessive social obligations overwhelm volitional control, generating loss-of-control appraisals. This differential-yet-simultaneous pattern precisely matches COR Theory's prediction of resource cascade mechanisms: threatened resources in one domain necessitate consumption of alternative resources for restoration attempts, creating multidimensional manifestations. Contemporary User Relevance The sample characteristics (mean tenure = 6.8 years, age modal = 25-29 years) capture users experiencing contemporary platform characteristics with maximum relevance. Unlike earlier studies including long-tenure early adopters, this specification ensures findings reflect current algorithmic and interface conditions shaping disengagement in the digital landscape that contemporary users actually experience. The R² = 0.67 for disengagement using contemporary user data suggests findings are maximally relevant to understanding current digital phenomena. COMPARISON WITH LITERATURE AND EMPIRICAL ADVANCEMENT Addressing Prior Research Limitations: Prior literature has operated under three critical limitations: Fragmented Stressor Examination: Existing research typically examines single or paired stressor categories. This study's comprehensive inclusion of four stressor dimensions enables specification of stressor interactions and relative contributions to disengagement mechanisms. Unidirectional Mechanism Assumption: Predominant frameworks assume stress operates through single pathways. The dual-pathway simultaneous mediation model demonstrates that disengagement emerges from parallel activation of multiple mechanisms. Missing Cross-Dimensional Effects: No published research has examined whether intrinsic stressors activate extrinsic-stressor-typical manifestations or vice versa. This study provides the first direct evidence that resource depletion operates across psychological domains. Empirical Advancement Over Meta-Analyses: Recent meta-analytical reviews synthesizing 150+ studies confirm associations between social media use and psychological distress, yet report wide effect size heterogeneity (r = 0.18 to 0.64). This heterogeneity likely derives from unmeasured mechanism complexity. The present study's integrated measurement of both mechanisms within a single large sample (N = 1,500) clarifies that heterogeneity reflects genuine mechanism complexity: both pathways operate substantially and simultaneously. PRACTICAL IMPLICATIONS FOR STAKEHOLDER ENGAGEMENT For Social Media Developers: Information overload demonstrates the strongest effect on psychosomatic tension (β = 0.52). Platform redesign should prioritize content filtering mechanisms allowing substantive user control over information volume rather than maximizing information presentation. Social overload similarly demands implementation of absence signals and asynchronous messaging reducing perceived responsiveness obligations. Privacy concerns require genuine transparency enabling perceived control. Social comparison mitigation demands algorithmic reconsideration of curated content amplification. For Mental Health Service Providers: The identification of simultaneous disempowerment (β=0.47→DD) and psychosomatic tension (β=0.40→DD) pathways requires clinically-integrated intervention: Disempowerment pathway intervention : Cognitive Reframing: Help clients distinguish between 'platform algorithms controlling information' vs. 'personal lack of control’ reduces internalized helplessness Media Literacy: Educate regarding algorithmic curation (posts are filtered, not representative reality) restores perceived agency through informed understanding Platform Literacy: Teach use of privacy settings, feed control features, notification management restores perceived mastery/autonomy - Implementation: Group cognitive-behavioral intervention addressing platform-specific helplessness cognitions (6 sessions, 1.5 hr each) psychosomatic pathway intervention: Stress Management: Standard CBT techniques for anxiety/fatigue management (progressive muscle relaxation, deep breathing during platform use) - Sleep Hygiene: Address screen-time's sleep disruption (blue light, stimulating content) through bedroom tech boundaries Affective Monitoring: Emotion tracking during/after platform use to identify triggers producing psychosomatic symptoms Gradual Exposure: Systematic desensitization where clients use platforms in controlled, time-limited manner while monitoring stress responses Implementation: Individual behavioral intervention (8-10 sessions) targeting platform-related stress symptom reduction combined approach: Simultaneous pathway activation implies single-pathway interventions risk incompleteness. Most effective clinical approach combines cognitive reframing + stress management, each targeting pathway-specific mechanisms. Efficacy metrics should assess both outcomes: (1) disempowerment reduction (sense of control restoration) + (2) psychosomatic tension reduction (stress symptom alleviation).. For Policymakers and Regulators: Evidence that simultaneous stress pathways drive disengagement at population level (R² = 0.67) suggests substantial policy-relevant public health implications. Regulatory frameworks should elevate algorithmic transparency to equivalent priority as data protection. Platforms should disclose known psychological effects comparable to pharmaceutical labeling. Duty-of-care obligations should address high-use populations at elevated risk. VALIDITY OF MEASUREMENT AND ANALYTICAL APPROACH Measurement Model Validation: All measurement criteria were satisfied. Indicator loadings exceeded 0.75, Composite Reliability ranged 0.87-0.93, and Discriminant Validity was established via HTMT <0.85. This rigor ensures reported associations reflect true construct relationships rather than measurement artifact. Analytical Appropriateness: Variance-based PLS-SEM represents optimal approach for this investigation: (1) handles complex multivariate models with multiple mediators; (2) prioritizes predictive accuracy enabling robust mediation testing; and (3) accommodates theoretical focus on specific postulated mechanisms. Two-stage analysis (measurement validation followed by structural testing) ensures observed relationships reflect true construct associations. Robustness and Generalizability: Multi-group analysis across demographic subgroups (age cohorts, gender, platforms) revealed no significant between-group structural differences, supporting generalizability across demographic diversity within the 18-35 age specification. Common method bias assessment via Harman's test (30.8% variance, below 50% threshold) confirms that reported associations are not artificially inflated. LIMITATIONS AND BOUNDARY CONDITIONS Methodological Limitations: Cross-Sectional Design: Precludes definitive causal inference. Longitudinal and experimental designs would strengthen confidence. Self-Report Measurement: Introduces bias despite procedural mitigation. Behavioral indicators (API data, real-time assessment) would provide convergent validity. Geographic Specification: Indonesia-specific recruitment limits generalizability. Cross-cultural replication is critical. Age/Tenure Window: 18-35 age and ≤10-year tenure excludes other demographic segments. Platform-specific moderation analyses warrant investigation. Theoretical Boundary Conditions: Platform-Specific Effects: This study examined generalized social media stress. Platform-specific mechanisms (Instagram visual comparison vs. X text-focused) may differ substantially. User Motivation Moderation: Individual differences in use motivations may moderate stress mechanisms. Personality Differences: Trait neuroticism and baseline perceived control likely moderate relationships. FUTURE RESEARCH DIRECTIONS Priority 1: Longitudinal and Experimental Designs Future research should employ longitudinal designs enabling causal inference. Panel studies tracking participants across 6-12 months would examine temporal stress-disengagement sequences and whether disengagement produces stress reduction. Experimental manipulation studies randomly assigning users to algorithmic exposure conditions would provide causal evidence complementing observational data. Priority 2: Platform-Specific and Moderation Analyses Multi-platform studies could isolate which stressor types are most salient on specific platforms. Moderation analyses examining personality, motivations, and usage patterns could identify user profiles most vulnerable to documented mechanisms, enabling targeted intervention design. Priority 3: Intervention Development and Testing Randomized controlled trials should test interventions targeting identified stress mechanisms. Intervention conditions might include algorithmic transparency enhancement, cognitive reframing, guided disengagement protocols, and combination interventions. Efficacy evaluation would enable evidence-based recommendations regarding highest-impact modalities. Priority 4: Cross-Cultural Replication Replication in culturally-diverse contexts would clarify which findings generalize universally versus reflect cultural specificity. CONCLUSION An empirical study of 1,500 Indonesian social media users (aged 18–35, tenure ≤ 10 years) validated a Conservation of Resources Theory model with R² = 0.67, substantially exceeding digital stress literature standards (0.35–0.50). Digital disengagement stems from two equally potent, independent mechanisms: loss-of-control appraisals (disempowerment: 0.706) and psychosomatic stress manifestations (0.712) both triggered by intrinsic and extrinsic stressors. This dual-pathway model demonstrates that resource depletion operates simultaneously across multiple psychological domains rather than activating isolated responses. Three stakeholders require action: platform developers must redesign algorithmic systems to reduce psychological burden; mental health practitioners must implement dual-pathway interventions targeting both disempowerment and psychosomatic manifestations; policymakers must establish regulatory frameworks prioritising algorithmic transparency alongside data protection. Methodological limitations include cross-sectional design, self-reported measures, Indonesia-specific recruitment, and restricted age/tenure parameters. Future research demands longitudinal designs, cross-cultural replication, and neurobiological investigation. As artificial intelligence exponentially amplifies user demands, empirically mapping stress mechanisms becomes urgently necessary for protecting psychological wellbeing at population scale. Declarations Data Availability Statement: The data supporting these findings are available from the corresponding author upon reasonable request. Funding Statement: This research was funded by the Jampang Pratama Foundation through a Research Grant for Lecturers Studying at the Polytechnic istikom bina citra informatika in 2025 with the number 207/K01.YBP/SK/III/2025. Conflict of Interest Statement: The authors declare that there are no financial or non-financial conflicts of interest that could influence the results or interpretation of this research. Author Contributions ICT: Conceptualization, Formal analysis, Methodology, Writing – original draft, Data curation, Validation, Funding acquisition, Writing – review & editing. SH: Conceptualization, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing, Data curation. NR: Conceptualization, Supervision, Validation, Writing – review & editing, Methodology. References Acquisti, A., Brandimarte, L., & Loewenstein, G. (2015). Privacy and human behavior in the age of information. Science , 347 (6221), 509–514. https://doi.org/10.1126/science.aaa1465 Davis, F. D. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Quarterly , 13 (3), 319–340. https://doi.org/10.2307/249008 Dhir, A., Kaur, P., Chen, S., & Pallesen, S. (2019). Antecedents and consequences of social media fatigue. Int. J. Inf. Manag. , 48 , 193–202. https://doi.org/10.1016/j.ijinfomgt.2019.05.021 Dwivedi, Y. K., Hughes, L., Baabdullah, A. M., Ribeiro-Navarrete, S., Giannakis, M., Al-Debei, M. M., Dennehy, D., Metri, B., Buhalis, D., Cheung, C. M. K., Conboy, K., Doyle, R., Dubey, R., Dutot, V., Felix, R., Goyal, D. P., Gustafsson, A., Hinsch, C., Jebabli, I., … Wamba, S. F. (2022). Metaverse beyond the hype: Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. International Journal of Information Management , 66 , 102542. https://doi.org/10.1016/j.ijinfomgt.2022.102542 Erdem, H. D., Herrero, J., & Urueña, A. (2025). The Bidirectional Relationships between Social Pressure in Digital Contexts, Depression, and Social Support over Time. Psychosocial Intervention , 34 , 189–200. https://doi.org/10.5093/pi2025a15 Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist , 44 (3), 513–524. https://doi.org/10.1037/0003-066X.44.3.513 Hobfoll, S. E., Halbesleben, J., Neveu, J.-P., & Westman, M. (2018). Conservation of Resources in the Organizational Context: The Reality of Resources and Their Consequences. Annual Review of Organizational Psychology and Organizational Behavior , 5 (1), 103–128. https://doi.org/10.1146/annurev-orgpsych-032117-104640 Holmgreen, L., Tirone, V., Gerhart, J., & Hobfoll, S. E. (2017). Conservation of Resources Theory. In The Handbook of Stress and Health (pp. 443–457). Wiley. https://doi.org/10.1002/9781118993811.ch27 Jie, Z., Mou, Q.-F., Zheng, T., Gao, F., Zhong, Y., Lu, Q., Gao, Y., & Zhao, M. (2023). A serial mediation model of social media addiction and college students’ academic engagement: the role of sleep quality and fatigue. BMC Psychiatry , 23 . https://doi.org/10.1186/s12888-023-04799-5 Katz, E., Blumler, J. G., & Gurevitch, M. (1973). Uses and Gratifications Research. The Public Opinion Quarterly , 37 (4), 509–523. http://www.jstor.org/stable/2747854 Lo, J. (2019). Exploring the buffer effect of receiving social support on lonely and emotionally unstable social networking users. Comput. Hum. Behav. , 90 , 103–116. https://doi.org/10.1016/j.chb.2018.08.052 Malik, A., Dhir, A., Kaur, P., & Johri, A. (2020). Correlates of social media fatigue and academic performance decrement: A large cross-sectional study. Inf. Technol. People , 34 , 557–580. https://doi.org/10.1108/itp-06-2019-0289 Mojtabai, R. (2024). Problematic social media use and psychological symptoms in adolescents. Social Psychiatry and Psychiatric Epidemiology , 59 , 2271–2278. https://doi.org/10.1007/s00127-024-02657-7 Pang, H., & Zhang, K. (2024). Determining multidimensional influences of network heterogeneity on university students’ psychological and academic well-being: The mediating role of social network exhaustion. Heliyon , 10 . https://doi.org/10.1016/j.heliyon.2024.e32328 Short, John., Williams, Ederyn., & Christie, Bruce. (1976). The social psychology of telecommunications . Wiley. Solomon, R. L., & Corbit, J. D. (1974). An opponent-process theory of motivation: I. Temporal dynamics of affect. Psychological Review , 81 (2), 119–145. https://doi.org/10.1037/h0036128 Song, H., Zmyslinski-Seelig, A., Kim, J., Drent, A., Victor, A., Omori, K., & Allen, M. (2014). Does Facebook make you lonely?: A meta analysis. Computers in Human Behavior , 36 , 446–452. https://doi.org/10.1016/j.chb.2014.04.011 Van Der Schuur, W., Baumgartner, S., & Sumter, S. (2019). Social Media Use, Social Media Stress, and Sleep: Examining Cross-Sectional and Longitudinal Relationships in Adolescents. Health Communication , 34 , 552–559. https://doi.org/10.1080/10410236.2017.1422101 Yue, Z., Zhang, R., & Xiao, J. (2021). Passive social media use and psychological well-being during the COVID-19 pandemic: The role of social comparison and emotion regulation. Computers in Human Behavior , 127 , 107050. https://doi.org/10.1016/j.chb.2021.107050 Zhang, S., Shen, Y., Xin, T., Sun, H., Wang, Y., Zhang, X., & Ren, S. (2021). The development and validation of a social media fatigue scale: From a cognitive-behavioral-emotional perspective. PLoS ONE , 16 . https://doi.org/10.1371/journal.pone.0245464 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8396588","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":588775959,"identity":"4812b5d5-9ac5-4803-8f9f-67628926dd32","order_by":0,"name":"Indra Cahaya Tresna","email":"data:image/png;base64,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","orcid":"","institution":"Airlangga University","correspondingAuthor":true,"prefix":"","firstName":"Indra","middleName":"Cahaya","lastName":"Tresna","suffix":""},{"id":588775960,"identity":"277cdafc-5d84-4c4e-b702-621b1a3023ff","order_by":1,"name":"Sri Hartini","email":"","orcid":"","institution":"Airlangga University","correspondingAuthor":false,"prefix":"","firstName":"Sri","middleName":"","lastName":"Hartini","suffix":""},{"id":588775961,"identity":"bc1eed75-c28b-4242-a94b-310fa3edd53e","order_by":2,"name":"Novalia Rachmah Polytechnic","email":"","orcid":"","institution":"Polytechnic Istikom Bina Citra Informatika","correspondingAuthor":false,"prefix":"","firstName":"Novalia","middleName":"Rachmah","lastName":"Polytechnic","suffix":""}],"badges":[],"createdAt":"2025-12-18 14:38:51","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8396588/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8396588/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102543008,"identity":"b34bca74-ec30-4941-97d3-6e1dcf6fc844","added_by":"auto","created_at":"2026-02-12 19:41:00","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":107451,"visible":true,"origin":"","legend":"\u003cp\u003eConceptual Model\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8396588/v1/a72da53e6b131a6b60b1d806.png"},{"id":103056303,"identity":"c7d24d79-368d-419a-9712-b159ad0198df","added_by":"auto","created_at":"2026-02-20 09:05:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":790778,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8396588/v1/fb7f0b10-3682-4694-8315-c3ac785b2cd5.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Beyond the Scroll: Unmasking the Simultaneous Role of Psychosomatic and Stressors Behind Social Media Users’ Disempowerment and Disengagement","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eApproximately 5.07\u0026nbsp;billion individuals engage daily with social media (Data portal, 2024), demonstrating unprecedented digital connectivity. Extensive research documents that social media facilitates interpersonal communication, information sharing, and community building. However, this widespread engagement paradoxically correlates with increased psychological distress, including emotional exhaustion, depression, and anxiety (Malik et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Mojtabai, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Social media stressors encompass intrinsic factors such as social comparison reducing self-efficacy and privacy concerns undermining autonomy and extrinsic factors, including social and information overload (Acquisti et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Malik et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This interconnection differentiates social media stress from other workplace or environmental stressors.\u003c/p\u003e \u003cp\u003eDespite this field complexity, digital disengagement users' intentional withdrawal from platforms remains understudied and under-theorised. Paradoxically, one-third of social media users plan to reduce or cease usage altogether (Data portal, 2024), yet engagement-focused algorithmic systems amplify psychological stressors through information and social overload the 'paradox of connectivity' (Erdem et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Lo, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Emerging technologies, including the Metaverse and AI-driven content, will further increase cognitive and social stressors (Dwivedi et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA critical empirical gap exists: extant literature compartmentalises intrinsic and extrinsic stressors, thereby bypassing simultaneous pathway activation and cross-dimensional interplay. Mediation conditions in pathway multiplication remain unexplored, limiting intervention frameworks. Furthermore, digital disengagement research remains predominantly Western-centric, restricting theoretical generalisation.\u003c/p\u003e \u003cp\u003eWithout systematic understanding of stressor mechanisms, capacity to predict technology's impact on stress remains limited. This gap constrains the theoretical basis for comprehensive psychosomatic and disempowerment interventions. Predictive capacity requires empirically understood relational mechanisms, particularly for mental health policy development. Geographic diversification of research is essential for validating the generalisation of stress and disengagement mechanisms beyond Western contexts.\u003c/p\u003e \u003cp\u003eThis study addresses these gaps by proposing and empirically validating an integrated Conservation of Resources (COR) Theory model through a quantitative, cross-sectional design with Indonesian social media users (N\u0026thinsp;=\u0026thinsp;1,500; tenure\u0026thinsp;\u0026le;\u0026thinsp;10 years; age 18\u0026ndash;35). The model examines disempowerment and psychosomatic tension as interrelated mediating pathways, investigating: (1) whether intrinsic stressors (social comparison, privacy concerns) directly activate both pathways; (2) whether extrinsic stressors (social overload, information overload) directly activate both pathways; and (3) whether cross-dimensional interaction effects exist. Data are analysed via variance-based structural equation modelling (SmartPLS 4.0).\u003c/p\u003e \u003cp\u003eThe central research question is: How do different stressors impact digital disengagement through different mechanisms? Eight research questions guide this inquiry, examining direct effects, mediation pathways, and indirect effects.\u003c/p\u003e \u003cp\u003eThis study provides the first empirical validation of COR Theory's cascading-effect mechanism across simultaneous psychological domains in social media contexts. It advances beyond compartmentalised frameworks by demonstrating that resource depletion operates through parallel, equally potent mechanisms rather than single pathways. Additionally, the Indonesian context contributes essential geographic representation to digital disengagement research, validating cross-cultural applicability. Findings inform technology designers, developers, clinicians, and policymakers seeking evidence-based interventions for platform-related stress and user withdrawal.\u003c/p\u003e"},{"header":"LITERATURE REVIEW","content":"\u003cp\u003e\u003cem\u003eConservation of Resources Theory\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCOR Theory (Hobfoll, 1989) posits that people protect, maintain, and acquire resources both tangible (time, money) and psychological (self-efficacy, control). Stress occurs when resources are insufficient, lost, or threatened, triggering cascading effects across psychological systems. For example, privacy invasions threaten control/autonomy (disempowerment pathway) while draining emotional resources (psychosomatic pathway). Social media uniquely combines intrinsic stressors (social comparison, privacy concerns) and extrinsic stressors (information overload, social overload), necessitating complex stress theory. COR outperforms alternative frameworks on four dimensions: (1) recognizes multi-domain resource withdrawal simultaneously, unlike TAM and Social Presence Theory; (2) predicts cascading effects where one stressor activates multiple psychological systems; (3) emphasizes resource appraisal, aligning with stress models; and (4) demonstrates empirical support in occupational stress\u0026nbsp;(Dhir et al., 2019; Malik et al., 2020), establishing applicability to other fields. The convergent threats to autonomy\u0026nbsp;(Acquisti et al., 2015), social evaluation\u0026nbsp;(Yue et al., 2021), and information processing\u0026nbsp;(Dhir et al., 2019)\u0026nbsp;justify COR\u0026apos;s application to digital disengagement. A comparison of these theoretical frameworks is presented in Table 1.\u003c/p\u003e\n\u003cp\u003eTable 1. Theoretical Framework\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eTheoretical Framework\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003ePathway Specificity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eCross-Domain Effects\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eMediation Specification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003eDisengagement Prediction\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eTAM ((Davis, 1989)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSingle pathway\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eNot applicable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003eNo\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eSocial Presence Theory\u0026nbsp;(Short et al., 1976)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eDevice-specific\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eLimited\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eNot examined\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003eNo\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eUGT\u0026nbsp;(Katz et al., 1973)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSingle pathway\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eNo\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eNot specified\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003eIndirect only\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eCOR Theory\u0026nbsp;(Hobfoll et al., 2018; Holmgreen et al., 2017)\u0026nbsp;(Proposed)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eDual simultaneous\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eYes (eksplisit)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003eSimultaneous specification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003eYes (direct + indirect)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eCONCEPTUAL MODEL OVERVIEW\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe proposed integrated model defines 8 direct relationships and two dual-pathway mediation mechanisms. Two independent variable (IV) categories comprise four (4) stressor dimensions: (1) intrinsic stressors: social comparison and privacy concerns; and (2) extrinsic stressors: social overload and information (cognitive) overload. Two mediating variables (MV) capture distinct resource-depletion manifestations: (1) disempowerment (a loss of control appraisals and decreased perceived autonomy), and (2) psychosomatic tension (which includes fatigue and anxiety, and reflects affective and physiological stress manifestations). One dependent variable (DV) operationalises the behavioural outcome: Digital disengagement reflects users\u0026apos; intentional withdrawal from activities on the platform and is operationalised as disengagement in the model. The model aims to test if both mediating pathways act in parallel, if there are cross-dimensional effects (intrinsic stressors \u0026rarr; psychosomatic tension; extrinsic stressors \u0026rarr; disempowerment) and if both mediators do so in isolation in predicting disengagement.\u003c/p\u003e\n\u003cp\u003eHYPOTHESIS DEVELOPMENT\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDirect Effects of Intrinsic Stressors on Disempowerment\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSocial comparison and privacy breaches activate loss-of-control appraisals by diminishing self-efficacy and autonomy (Acquisti et al., 2015; Yue et al., 2021). These autonomy-threatening stressors erode psychological resources central to disempowerment. COR Theory predicts that such resource challenges prompt greater loss-of-control appraisals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(H1): Social comparison and privacy concerns directly increase users\u0026apos; disempowerment.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDirect Effects of Intrinsic Stressors on Psychosomatic Tension\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSocial comparison and privacy concerns activate psychosomatic tension through distinct mechanisms: mental exhaustion and rumination from social evaluation, and anxiety from perceived privacy breaches (Acquisti et al., 2015; Van Der Schuur et al., 2019). These mechanisms manifest affectively-physiologically (fatigue, anxiety, sleep disruption) and operate independently from disempowerment\u0026apos;s control-appraisal pathway. This differentiation aligns with theoretical frameworks proposing stressor-specific activation pathways (Dhir et al., 2019).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(H2): Social comparison and privacy concerns directly increase psychosomatic tension.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDirect Effects of Extrinsic Stressors on Psychosomatic Tension\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSocial overload and information overload are extrinsic stressors which primarily threaten cognitive and affective resources. Information overload occurs when incoming information exceeds the individual\u0026apos;s ability to process such information, resulting in the need to intensify sustained attention and make decisions\u0026nbsp;(Malik et al., 2020; Pang \u0026amp; Zhang, 2024). The depletion of one\u0026rsquo;s cognitive resources and attention results in mental fatigue and social overload. Social overload refers to excessive social commitments, interpersonal demands, and maintenance of social relations which draw on emotional resources through empathic and social performance\u0026nbsp;(Malik et al., 2020; Pang \u0026amp; Zhang, 2024).\u0026nbsp;Both mechanisms impacting stress and fatigue affect the emotional and physiological systems. Information overload and social overload are linked to psychosomatic symptoms of stress\u0026nbsp;(Jie et al., 2023; Van Der Schuur et al., 2019). According to COR Theory\u0026nbsp;(Hobfoll, 1989), the depletion of resources manifests in the individual psychosomatically as stress. External stressors predict psychosomatic tension.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;(H3): Social overload and information overload directly increase psychosomatic tension.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDirect Effects of Extrinsic Stressors on Disempowerment\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAlthough extrinsic stressors mainly burden the cognitive-emotional dimension, COR Theory\u0026apos;s cascading-effect principle predicts cross-dimensional impact. Information overload diminishes decision-making and understanding, and, consequently, users\u0026apos; feelings of mastery and control over the platform\u0026apos;s navigation and information handling\u0026nbsp;(Dhir et al., 2019; Malik et al., 2020). Social overload also erodes agency. Excessive social demands surpass the capacity for selective participation and lead to feelings of engagement and control being forced upon\u0026nbsp;(Erdem et al., 2025; Pang \u0026amp; Zhang, 2024). Qualitative investigations of withdrawal from platforms describe users\u0026rsquo; sentiments of having been \u0026ldquo;overwhelmed,\u0026rdquo; \u0026ldquo;unable to keep up,\u0026rdquo; which pertains to appraised resource loss, exhaustion, and control\u0026nbsp;(Dhir et al., 2019; Malik et al., 2020). Thus, disempowerment stemmed from extrinsic stressors is the consequence of cross-dimensional resource loss.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;(H4): Social overload and information overload directly increase disempowerment.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDirect Effects of Mediators on Digital Disengagement\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDisempowerment and perceptions of reduced control and lack of autonomy evoke a behavioural withdrawal response\u0026nbsp;(Dhir et al., 2019; Hobfoll, 1989; Malik et al., 2020; Solomon \u0026amp; Corbit, 1974). Predictive of disengagement from environments that pose control challenges, frameworks of psychological reactance and learned helplessness describe sustained loss of control as disengagement(Dhir et al., 2019; Malik et al., 2020). The disengagement that stems from disempowerment gives rise to an inability to restore control and agency through participation and a sense of futility to be re-engaged. Empirical evidence overwhelmingly demonstrates that loss of control predicts the intention to stop using a given platform\u0026nbsp;(Dhir et al., 2019; Malik et al., 2020)\u0026nbsp;Hence, disempowerment is the cause of digital disengagement.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;(H5): Disempowerment directly increases digital disengagement.\u003c/p\u003e\n\u003cp\u003ePsychosomatic tension manifests as stress, fatigue, and anxiety, from which individuals seek to escape through psychosomatic affective states. Negative reinforcement (avoidance) leads to the prediction of the individual avoiding social contexts that combine psychosomatic affect states. There is an escape from stress symptoms and a psychosomatic cost associated with prolonged exposure to social media\u0026nbsp;(Solomon \u0026amp; Corbit, 1974).\u0026nbsp;There is considerable evidence that stress and fatigue impact disengagement from social media platforms\u0026nbsp;(Dhir et al., 2019; Song et al., 2014). This disconnects primary social media platforms from psychosomatic tension.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;(H6): Psychosomatic tension directly increases digital disengagement.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDual Mediation Pathways\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAll four stressor dimensions (both intrinsic and extrinsic) activate disempowerment via control-loss mechanisms (H1, H4), resulting in disengagement (H5) in turn. COR Theory expects and posits that loss-of-control appraisals are at the very centre of how resource threats translate to behavioural changes. Assessing this mediation pathway captures the psychological control aspect of the stress response.\u003c/p\u003e\n\u003cp\u003e(H7): Disempowerment mediates the relationships between intrinsic (social comparison, privacy concerns) and extrinsic stressors (social and information overload) and digital disengagement.\u003c/p\u003e\n\u003cp\u003eAll four dimensions of stressors activate psychosomatic tension via resource depletion (H2, H5), in turn resulting in disengagement through a negative reinforcement cycle (H6), and in doing so this helps to isolate the affective-physiological part of the stress response.\u003c/p\u003e\n\u003cp\u003eHypothesis 8 (H8): Psychosomatic tension mediates the relationships between intrinsic stressors (social comparison, privacy concerns); extrinsic stressors (social overload, information overload); and digital disengagement.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCritical Integration: The Simultaneous Dual-Pathway Model\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe original contribution of this study rests on the simultaneous testing of both mediation pathways within the confines of a singular structural model. Previous studies only investigate and isolate one mediation path, which restricts the ability to assess: (1) the concurrent operation of both pathways; (2) the differing relative strengths of pathways; and (3) the extent to which cross-dimensional pathways (H2, H4) produce a meaningful contribution to the overarching effect. By estimating both pathways concurrently, the study aims to evaluate COR Theory\u0026rsquo;s fundamental premise of resource loss spanning multiple psychological domains through cascading effects occurring simultaneously and in the context of disengaging from social media. This approach allows for a richer specification of the integrated complexity of the mechanisms involved than is provided by the literature which has approached the subject in a piecemeal manner. The proposed conceptual model is illustrated in Figure 1.\u003c/p\u003e\n\u003cp\u003eCONCEPTUAL MODEL\u003c/p\u003e\n\u003cp\u003eTHEORETICAL CONTRIBUTIONS\u003c/p\u003e\n\u003cp\u003eThe proposed model advances digital disengagement understanding in three ways. First, simultaneous multipath specification captures cascading, multidimensional resource depletion pathways simultaneously advancing beyond fragmented, single- pathway literature. Second, cross-dimensional hypotheses determine whether stressors operate in isolation or generally across psychological pathways, advancing theoretical and practical understanding. Third, mechanism complexity recognizes digital disengagement stems from centralized functioning of multiple mechanisms rather than singular pathways, better addressing psychological intricacy documented in platform withdrawal research.\u003c/p\u003e"},{"header":"METHODOLOGY","content":"\u003cp\u003e\u003cem\u003eResearch Design and Approach\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eQuantitative cross-sectional correlational design enabling simultaneous variable measurement and complex multivariate relationship examination via SEM.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSample and Sampling Procedure\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSample Specification: N = 1,500 active social media users. This total sample size is well above the power threshold for multi mediating SEM models. Considering the rule of thumb of 10 to 20 for the indicator count per the variance based structure of SEM, 1,500 provides more than ample power (\u0026gt; 0.90) to help test the hypothesized relationships for the eight relationships which would be small to moderate effects (Hair et al., 2016). An a priori power analysis (G*Power 3.1) was also conducted to check for sample size sufficiency for the multiple regression (primary effect sizes: \u0026beta; = 0.30 for direct effects, \u0026beta; = 0.25 for indirect effects); \u0026alpha; = 0.05; power = 0.90); results indicated N of 147 should be sufficient to detect effects of moderate size (f^2 = 0.15) within regression contexts. For the N = 1,500 was well above the threshold, within, post hoc power estimated to be above 0.95 for all effects hypothesized, indirect effects to be small to moderate (0.15 \u0026lt; \u0026beta; \u0026lt; 0.20).\u003c/p\u003e\n\u003cp\u003eSampling Method: There was purposive sampling. Participants were specifically selected based on inclusion criteria which matched the study population: active users on social media currently experiencing symptoms related to platform stress. While purposive sampling made it possible for us to focus on theoretically pertinent users (current social media users), it came at a cost of potential selection bias. To measure the extent of bias, the respondents\u0026apos; characteristics (age, sex, platform) were compared to the social media demographic profile of Indonesians on a national level (Datar portal 2024). In addition, further procedures were implemented to deal with selection bias, and were also evaluated. The sample was weighted to correctly reflect the overall population relevant to the study, based on the differences in demographic characteristics. Sensitivity analyses were performed to assess the effect of some possibly unrepresented parts of the sample. The analyses established that the key characteristics were indeed represented in a sample, so that the issues related to selection bias were hardly a concern.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eParticipant Eligibility Criteria:\u003c/em\u003e\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003ePlatform Activity Status: Currently use social media (minimum weekly)\u003c/li\u003e\n \u003cli\u003eUser Tenure: Maximum 10 years active use (targeting contemporary users)\u003c/li\u003e\n \u003cli\u003eAge Range: 18-35 years old\u003c/li\u003e\n \u003cli\u003eGeographic Context: Indonesia\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eExclusion Criteria:\u0026nbsp;More than 10 years tenure, outside 18-35 age range, no current engagement, or failed attention-check items.\u003c/p\u003e\n\u003cp\u003eMEASUREMENT INSTRUMENTS\u003c/p\u003e\n\u003cp\u003eSeven latent constructs were operationalised using multi-item scales. All scales employed 5-point Likert-type response anchors (1 = Strongly Disagree to 5 = Strongly Agree), with 50% of items reverse-coded to mitigate acquiescence bias. Reverse-coding implementation followed a systematic procedure: within each 6-item construct, 3 items were phrased in a direction-congruent manner (e.g., SC2: \u0026apos;I compare my life to others on social media\u0026apos;) while 3 items were phrased in an opposite semantic direction (e.g., SC5-reverse: \u0026apos;I do not care how my life compares to others online\u0026apos;). Reverse-coded items were recoded prior to analyses such that higher scores consistently indicated higher construct levels. This balancing procedure reduces acquiescence threat (tendency to endorse items regardless of content) estimated in the literature to bias 15-25% of correlations when unidirectional items are used exclusively (Podsakoff et al., 2003).\u003c/p\u003e\n\u003cp\u003eIn the Indonesian context, we willingly justify the validity and reliability of the construct and measurement. This involved an elaborate process of adaptations of the scales which included translation, back translation, and verified procedures. Also, piloting the tests with the target population as the focus group helped ascertain reliability and comprehension of the measurement tools. It helped ensure that they achieved the desired cultural adaptation as measurement tools.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIntrinsic Stressors - Social Comparison (SC)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOver the years, several studies have examined social comparison, specifically the tendency to assess one\u0026rsquo;s worth through the presumed worth of others, using the Social Comparison Orientation Scale adapted for social media\u0026nbsp;(Yue et al., 2021). Social media comparison presumably encompasses contradictory evaluation processes and affective consequences, as respondents may analyse themselves to a negative extreme and experience feelings of worthlessness, such as when observing others\u0026rsquo; seemingly unnecessary advanced achievements (\u0026ldquo;I often compare my life to others\u0026rsquo; lives on social media\u0026rdquo;). Such formats and evaluation processes were employed in the social media use within the adapted instrument, and three items measuring affective dimensions were reverse coded to mitigate response bias, where respondents with a low affective dimension of worth might exhibit a high-biased acute affective self-evaluation process in their answers. Previous studies have indicated that this adapted process produces a high biased affective dimension response in relation to social media, with a Cronbach\u0026rsquo;s alpha of .82.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIntrinsic Stressors - Privacy Concerns (PC)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFor the purposes of this study, privacy-related concerns of the users in this study, (i.e., the Internet Users\u0026apos; Information Privacy Concerns scale developed by Smith et al., 1996) though tailored to online social media platforms, were utilised. The scale included six total items addressing privacy concerns (collection, control, and awareness). Three of the items were reverse scored. As per previous validation, the Cronbach\u0026apos;s alpha was 0.79.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExtrinsic Stressors - Social Overload (SO)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSocial overload has been defined as high social responsibilities and interaction overload on the platforms and was measured through the Social Overload Subscale of the Perceived Stress in Social Media Scale\u0026nbsp;(Malik et al., 2020; Pang \u0026amp; Zhang, 2024). This five-item instrument assesses pressure to maintain relationships (\u0026apos;\u0026apos;I feel pressure to keep in touch with everyone\u0026apos;\u0026apos;); complexity of relationships (\u0026apos;\u0026apos;Keeping social media relationships drains my energy\u0026apos;\u0026apos;); and expectations of social interaction (\u0026apos;\u0026apos;I feel obligated to respond to everyone\u0026apos;\u0026apos;). Responses to two items were reversed. Cronbach\u0026apos;s alpha coefficient was 0.81.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExtrinsic Stressors - Information Overload (IO)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe Information Overload Subscale of Malik et al. (2020); Dhir et al. (2019) Perceived Stress in Social Media Scale was used to measure the overload of information as cognitive strain due to the demands of processing and exposure to excessive information, to which three respondents did not provide answers. This five-item scale measures cognitive load, fragmentation of attention, and difficulty in filtering. Two items of which had been reverse scored. Cronbach\u0026apos;s alpha was .77.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMediator - Disempowerment (DE)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDisempowerment defined as loss-of-control appraisals and reduced perceived autonomy was measured using a construct adapted Perceived Control Scale\u0026nbsp;(Dhir et al., 2019)\u0026nbsp;and General Perceived Self-Efficacy Scale\u0026nbsp;(Dhir et al., 2019; Zhang et al., 2021)\u0026nbsp;specific to the use contexts of the platforms. The scale measured over six items coded as loss-of-control perceptions, lowered autonomy, and ineffectiveness. Three items were reverse coded. Previous research indicated the combined scale reliability to be 0.84.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMediator - Psychosomatic Tension (PT)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003ePsychosomatic tension affective and physiological manifestations of stress, including fatigue, anxiety, and sleep disruption was measured using the Perceived Stress Scale - Social Media Version\u0026nbsp;(Dhir et al., 2019). The six-item scale captures affective symptoms, physical manifestations, and sleep disruption. Three items received reverse-coding. Cronbach\u0026apos;s alpha in published research = 0.86.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDependent Variable - Digital Disengagement (DD)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDigital disengagement users\u0026apos; intentional reduction in platform use, content consumption, or interaction frequency was operationalized using the Platform Disengagement Intention Scale\u0026nbsp;(Dhir et al., 2019; Yue et al., 2021)\u0026nbsp;supplemented with behavioral indicators. The six-item scale measures disengagement intent, actual behavior reduction, and platform avoidance. Three items were reverse-coded. Cronbach\u0026apos;s alpha = 0.88.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCommon Method Bias Mitigation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMultiple procedural remedies were implemented to minimize common method bias (CMB):\u0026nbsp;\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eTemporal Separation: Predictor items presented in Section A; outcome variables in Section D (mean separation: ~12 survey items)\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eScale Mixing: Reverse-coded 50% of items across all constructs to prevent acquiescence patterns\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePsychological Separation: Item randomization within construct blocks to disrupt response patterns\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eAnonymity Assurance: Explicit statement \u0026quot;Your responses will be completely anonymous and confidential\u0026quot; displayed at survey entry\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eReduced Cognitive Demand: Mean item completion time 12-18 minutes (adequate, not exhausting)\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eProcedural Randomization: Item order randomized across 5 questionnaire versions distributed to different response cohorts\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cem\u003eData Collection Procedure\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003ePlatform: Google Forms online survey. Distribution: Purposive recruitment via social media communities, snowball sampling, and online forums. Timeline:43-week collection period. Data Quality Screening: Attention-check items, completion-time flags, and missing-data rules. Final N = 1,500 after screening 1,587 responses.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEthical Approval and Informed Consent\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis research was conducted in full accordance with the Declaration of Helsinki (1964, with subsequent amendments), which establishes fundamental ethical principles for research involving human subjects. All procedures and protocols were reviewed and formally approved by the Research Ethics Committee at Citra Buana Indonesia Institute, Sukabumi, Indonesia (Approval Reference: No.189/ETHICS/ICBI-E/VII/2025; Date of Approval: July 05, 2025).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Informed consent was obtained from all 1,500 participants included in the final analyzed sample. At the survey entry, participants received an explicit consent statement confirming: (1) participation was entirely voluntary; (2) they could withdraw at any time without penalty; (3) responses would remain completely anonymous and confidential; (4) data would be stored securely; and (5) data would be used exclusively for research purposes. All participants confirmed their understanding and agreed to these terms before accessing the survey. No personally identifiable information was collected.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAnalytical Strategy\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSoftware: PLS-SEM via SmartPLS 4.0. Significance Threshold: \u0026alpha; = 0.05 (two-tailed). Two-Stage Process: (1) Measurement model validation (\u0026lambda; \u0026gt; 0.70, CR \u0026gt; 0.70, AVE \u0026gt; 0.50, HTMT \u0026lt; 0.85); (2) Structural model testing via bootstrapping (5,000 resamples, 95% CI), with 5,000 resamples chosen to ensure statistical stability and precision of the indirect-effect estimates. This robust resampling approach provides greater confidence in the reliability of effect sizes compared to fewer resamples. Robustness Checks: Alternative model specification, multi-group analysis.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cem\u003eSample Characteristics and Data Screening\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eData collection yielded 1,587 survey responses. Following systematic data quality screening, 87 responses were excluded: 15 responses failed embedded attention-check items, 28 responses were completed in \u0026lt;4 minutes (insufficient engagement), 22 responses exceeded 45 minutes of completion time (inattention), and 22 responses contained \u0026gt;10% missing data. The final analyzed sample comprised N = 1,500 respondents, as detailed in Table 2.\u003c/p\u003e\n\u003cp\u003eTable 2. Respondent Characteristic\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 616px;\"\u003e\n \u003cp\u003eVariabels\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 313px;\"\u003e\n \u003cp\u003e\u003cem\u003eGender\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 303px;\"\u003e\n \u003cp\u003e\u003cem\u003ePrimary Platform\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e812\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e54.1%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003eInstagram\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e634\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e42.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e688\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e45.9%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003eWhatsApp\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e472\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e31.5%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cem\u003eAge Group\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003eFacebook\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e267\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e17.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e18-24 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e465\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e31.0%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003eX/Twitter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e5.9%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e25-29 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e612\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e40.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003eTikTok\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e2.5%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e30-35 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e423\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e28.2%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 303px;\"\u003e\n \u003cp\u003e\u003cem\u003eUser Tenure\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003e1-3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e20.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003e4-6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e589\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e39.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003e7-10 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e599\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e39.9%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 99px;\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 118px;\"\u003e\n \u003cp\u003e1-3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e20.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe sample comprised slightly more female (54.1%) than male respondents, with the modal age group 25-29 years (40.8%), consistent with the target demographic. Instagram and WhatsApp represented primary platforms for 73.8% of participants, reflecting current Indonesian social media adoption patterns. Mean user tenure was 6.8 years (SD = 2.9), confirming contemporary user experience consistent with inclusion criteria (maximum 10 years).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDescriptive Statistics\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 3: Descriptive Statistics for All Study Variables\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eMin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003eMax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003eSkewness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eKurtosis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003eSocial Comparison (SC)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e-0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003ePrivacy Concerns (PC)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003eSocial Overload (SO)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e-0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003eInformation Overload (IO)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003eDisempowerment (DE)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e-0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003ePsychosomatic Tension (PT)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e-0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 187px;\"\u003e\n \u003cp\u003eDigital Disengagement (DD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e1.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e5.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e-0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003econfirming contemporary user experience consistent with inclusion criteria maximum 10 years. The descriptive statistics for all study variables are presented in Table 3. All variables demonstrated approximately normal distributions with skewness and kurtosis values within acceptable ranges (\u0026minus;1.0 to +1.0), supporting the suitability of parametric estimation procedures. Mean values clustered around 3.2-3.7 on the 5-point scale, indicating moderate-to-moderately-high endorsement of stressor experiences and disengagement intentions. Information Overload (M = 3.71, SD = 0.79) demonstrated the highest mean, suggesting this stressor is particularly salient for contemporary users. Social Comparison demonstrated the lowest mean (M = 3.28, SD = 0.84), though still indicating substantial experience.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMeasurement Model Results\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eStructural equation modeling proceeded in two stages: measurement model evaluation followed by structural model testing. All measurement model criteria were satisfied\u0026nbsp;confirming the valid operationalization of study constructs (see Table 4).\u003c/p\u003e\n\u003cp\u003eTable 4. Measurement Model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 168px;\"\u003e\n \u003cp\u003eConstruct\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eIndicator\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003eLoading\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eAVE\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" style=\"width: 168px;\"\u003e\n \u003cp\u003eSocial Comparison\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSC1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSC2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSC3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSC4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSC5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSC6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" style=\"width: 168px;\"\u003e\n \u003cp\u003ePrivacy Concerns\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePC1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePC2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePC3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePC4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePC5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePC6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" style=\"width: 168px;\"\u003e\n \u003cp\u003eSocial Overload\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSO1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSO2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSO3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSO4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eSO5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" style=\"width: 168px;\"\u003e\n \u003cp\u003eInformation Overload\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eIO1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"5\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eIO2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eIO3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eIO4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eIO5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" style=\"width: 168px;\"\u003e\n \u003cp\u003eDisempowerment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDE1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDE2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDE3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDE4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDE5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDE6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" style=\"width: 168px;\"\u003e\n \u003cp\u003ePsychosomatic Tension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePT1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePT2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.87\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePT5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ePT6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" style=\"width: 168px;\"\u003e\n \u003cp\u003eDigital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDD1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"6\" style=\"width: 123px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDD2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDD3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDD4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDD5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eDD6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eIndicator Reliability: All factor loadings exceeded the 0.70 threshold, ranging from 0.73 to 0.87. The majority of loadings exceeded 0.80, indicating strong relationships between observed indicators and their respective latent constructs. No indicators required removal based on inadequate reliability.\u003c/p\u003e\n\u003cp\u003eInternal Consistency Reliability: Composite Reliability (CR) values ranged from 0.87 to 0.93, substantially exceeding the 0.70 criterion. Digital Disengagement and Psychosomatic Tension demonstrated the highest reliability (CR = 0.93 and 0.912, respectively), indicating highly consistent multi-item measurement.\u003c/p\u003e\n\u003cp\u003eConvergent Validity: Average Variance Extracted (AVE) values ranged from 0.62 to 0.73, all exceeding the 0.50 threshold. This indicates that latent constructs explain 62-73% of variance in their respective indicators, confirming that indicators validly measure their intended constructs.\u003c/p\u003e\n\u003cp\u003eTable 5: Discriminant ValidityHeterotrait-Monotrait (HTMT) Correlations\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003eVariabel\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eSC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003ePC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eSO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003eIO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eDE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003eDD\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003eSocial Comparison\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003ePrivacy Concerns\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003eSocial Overload\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003eInformation Overload\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003eDisempowerment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003ePsychosomatic Tension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 262px;\"\u003e\n \u003cp\u003eDigital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAll HTMT values remained below 0.85, the conservative threshold for discriminant validity. The highest HTMT ratio was observed between Information Overload and Psychosomatic Tension (0.78), which remained well below the cutoff. (Henseler et al., 2015) This pattern confirms that all seven constructs measure distinct phenomena rather than representing redundant dimensions. Discriminant validity was conclusively established. HTMT ratio represents heterotrait-monotrait correlation ratio; values \u0026lt;.85 indicate discriminant validity (constructs measure distinct phenomena). All HTMT values in Table 5 range from 0.54 to 0.80, substantially below the threshold, confirming discriminant validity\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStructural Model Results: Direct Effects\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFollowing measurement model validation, the structural model was estimated to test the eight hypothesized relationships. Path coefficients, standard errors, t-statistics, and 95% confidence intervals obtained via bootstrapping (5,000 resamples) are presented in Table 6.\u003c/p\u003e\n\u003cp\u003eTable 6: Structural Model Results Direct Effects (H1-H6)\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003ePath\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026Beta;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003et-value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e95% CI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003eH1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSC, PC \u0026rarr; DE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.038\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e10.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e[0.34, 0.48]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003eH2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSC, PC \u0026rarr; PT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.041\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e9.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e[0.29, 0.45]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003eH3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSO, IO \u0026rarr; PT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.036\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e14.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e[0.45, 0.59]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003eH4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSO, IO \u0026rarr; DE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.039\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e8.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e[0.27, 0.41]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003eH5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eDE \u0026rarr; DD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.032\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e14.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e[0.41, 0.53]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003eH6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003ePT \u0026rarr; DD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.035\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e11.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e[0.33, 0.47]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAll direct hypotheses were supported at p \u0026lt; .001 significance level. The pattern of effect sizes reveals theoretically meaningful relationships, interpreted based on the framework in Table 7. H1-H2 (Intrinsic Stressors): Social comparison and privacy concerns significantly affected disempowerment (\u0026beta; = 0.41, p \u0026lt; .001) and psychosomatic tension (\u0026beta; = 0.37, p \u0026lt; .001), supporting cross-dimensional effects. The stronger disempowerment effect aligns with theory; however, substantial psychosomatic tension effects (\u0026beta; = 0.37) demonstrate affective stress manifestations. H3-H4 (Extrinsic Stressors): Social and information overload most strongly affected psychosomatic tension (\u0026beta; = 0.52, p \u0026lt; .001) versus disempowerment (\u0026beta; = 0.34, p \u0026lt; .001). Significant disempowerment effects (\u0026beta; = 0.34) confirm cross-dimensional prediction: information overload reduces perceived mastery, generating loss-of-control appraisals. H5-H6 (Mediators on Disengagement): Disempowerment (\u0026beta; = 0.47, p \u0026lt; .001) and psychosomatic tension (\u0026beta; = 0.40, p \u0026lt; .001) both significantly predicted disengagement. Disempowerment\u0026apos;s stronger effect suggests loss-of-control appraisals constitute the more potent disengagement driver, though both mechanisms operate meaningfully.\u003c/p\u003e\n\u003cp\u003eTable 7. Effect Size Interpretation Framework\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eEffect Size (\u0026beta; or f\u0026sup2;)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eVerbal Interpretation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eResearch Context\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e0.01-0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eSmall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eSmall but meaningful psychological effects\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e0.05-0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eSmall-to-Medium\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eTypical psychological intervention effects\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e0.15-0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eMedium\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eSubstantial psychological impact\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e0.25+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eLarge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eLarge practical significance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003col\u003e\n \u003cli\u003eH1 (SC, PC\u0026rarr;DE): \u0026beta;=0.41 \u003cem\u003elarge\u003c/em\u003e - intrinsic stressors substantial direct predictor of disempowerment\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eH3 (SO, IO\u0026rarr;PT): \u0026beta;=0.52 \u003cem\u003elarge\u003c/em\u003e - extrinsic stressors strongest predictor of psychosomatic tension\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eH5 (DE\u0026rarr;DD): \u0026beta;=0.47 \u003cem\u003elarge\u003c/em\u003e - disempowerment potent predictor of disengagement\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eIndirect (Disempowerment pathway): 0.706 \u003cem\u003elarge\u003c/em\u003e - substantial mediation magnitude\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cem\u003eStructural Model Results: Mediation Effects\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIndirect effects were calculated to test whether disempowerment (H7) and psychosomatic tension (H8) mediate relationships between stressors and disengagement. Mediation was evaluated via bias-corrected bootstrapping with 5,000 resamples. The results for disempowerment and psychosomatic tension are summarized in Table 8 and Table 9, respectively. Effects were considered significant if 95% confidence intervals excluded zero and p \u0026lt; .05.\u003c/p\u003e\n\u003cp\u003eTable 8: Mediation Effects via Disempowerment (H7) \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eStressor \u0026rarr; Disempowerment \u0026rarr; Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eIndirect Effect\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e95% CI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eSignificance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eSocial Comparison \u0026rarr; Disempowerment \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.193\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.141, 0.250\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eP \u0026lt;. 001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003ePrivacy Concerns \u0026rarr; Disempowerment \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.193\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.141, 0.250\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eP \u0026lt;. 001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eSocial Overload \u0026rarr; Disempowerment \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.160\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.111, 0.214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eP \u0026lt;. 001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eInformation Overload \u0026rarr; Disempowerment \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.160\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.111, 0.214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eP \u0026lt;. 001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eTotal Mediation via Disempowerment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.706\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.598, 0.820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eP \u0026lt;. 001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 9: Mediation Effects via Psychosomatic Tension (H8)\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eStressor \u0026rarr; Psychosomatic Tension \u0026rarr; Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eIndirect Effect\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e95% CI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eSignificance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eSocial Comparison \u0026rarr; Psychosomatic Tension \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.148\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.104, 0.197\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003ep \u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003ePrivacy Concerns \u0026rarr; Psychosomatic Tension \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.148\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.104, 0.197\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003ep \u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eSocial Overload \u0026rarr; Psychosomatic Tension \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.208\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.158, 0.263\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003ep \u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eInformation Overload \u0026rarr; Psychosomatic Tension \u0026rarr; Digital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.208\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.158, 0.263\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003ep \u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 329px;\"\u003e\n \u003cp\u003eTotal Mediation via Psychosomatic Tension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.712\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 93px;\"\u003e\n \u003cp\u003e0.591, 0.840\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003ep \u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eBoth mediation pathways were fully supported. Notably, disempowerment mediation (total indirect effect = 0.706) and psychosomatic tension mediation (total indirect effect = 0.712) demonstrated comparable magnitude, supporting the theoretical prediction of simultaneous dual-pathway operation. Neither pathway overwhelmingly dominated; rather, both contributed substantially to understanding disengagement etiology. The near-equivalence (difference = 0.9%) provides strong evidence for COR Theory\u0026apos;s cascading-effect mechanism.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eModel Predictive Accuracy\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTable 10: Model R\u0026sup2; Values and Effect Sizes\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"621\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003eEndogenous Variable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003eR\u0026sup2;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 434px;\"\u003e\n \u003cp\u003eInterpretation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003eDisempowerment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e0.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 434px;\"\u003e\n \u003cp\u003eModerate-to-strong; stressors explain 46% of DE variance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003ePsychosomatic Tension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 434px;\"\u003e\n \u003cp\u003eStrong; stressors explain 51% of PT variance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003eDigital Disengagement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 434px;\"\u003e\n \u003cp\u003eStrong-to-very-strong; stressors + mediators explain 67% of DD variance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe structural model demonstrates substantial predictive accuracy, as detailed in Table 10. Stressors collectively explain 46% of disempowerment variance and 51% of psychosomatic tension variance, indicating that stress mechanisms are substantially, though not exclusively, determined by identified stressor categories. Critically, when both mediating pathways are included, the model accounts for 67% of the variance in digital disengagement. This suggests that nearly two-thirds of the reasons why young Indonesians disengage from social media platforms can be traced back to these stressors, providing a robust specification of disengagement etiology. This R\u0026sup2; value substantially exceeds typical effect sizes in psychological disengagement research (typically R\u0026sup2; = 0.35-0.50), suggesting theoretical and empirical advancement.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eModel Validation and Robustness\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eHarman\u0026apos;s test yielded 30.8% variance (\u0026lt; 50% threshold), confirming negligible common method variance. The dual-pathway model achieved superior fit (R\u0026sup2; = 0.67 vs 0.52 single-pathway; +29% explained variance), validating psychosomatic tension\u0026apos;s distinct contribution. Multi-group comparisons across age, gender, and platforms showed no significant between-group differences (permutation tests p \u0026gt; .05), confirming model generalizability. These convergent robustness checks establish validity and stability across demographic subgroups.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eAll eight hypothesized relationships received empirical support (N = 1,500 Indonesian users; tenure \u0026le;10 years, age 18-35). Intrinsic and extrinsic stressors demonstrated significant direct effects on both disempowerment and psychosomatic tension. Both mediators independently predicted digital disengagement through simultaneous dual pathways with comparable magnitude (disempowerment indirect = 0.706; psychosomatic tension = 0.712). The integrated model explained 67% of disengagement variance, substantially exceeding literature standards (R\u0026sup2; typically 0.35-0.50)\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCOR Theory Validation and Theoretical Contributions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe central theoretical contribution of this study resides in correlational validation of COR Theory\u0026apos;s cascading-effect mechanism predictions. Cross-sectional evidence demonstrates associations consistent with cascading predictions whereby stressor exposure predicts simultaneously manifested disempowerment and psychosomatic strain. While these patterns align with COR Theory predictions regarding simultaneous multidimensional resource depletion, cross-sectional design precludes definitive causal inference. Longitudinal and experimental designs in future research will strengthen confidence regarding temporal causal mechanisms underlying these correlations. Prior research examining social media stress has operated under compartmentalized frameworks assuming unidirectional pathways. The present findings challenge this assumption through three mechanisms:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eSimultaneous Dual-Pathway Operation\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe nearly equivalent pathway magnitudes (disempowerment = 0.706, psychosomatic tension = 0.712, difference = 0.9%) provide robust evidence for COR Theory\u0026apos;s prediction of simultaneous, parallel resource-depletion mechanisms rather than dominant single pathways. Importantly, the comparable magnitude (50.1% disempowerment vs 49.9% psychosomatic) indicates both mechanisms contribute equally to disengagement ethology neither can be dismissed as secondary or weaker driver.\u003c/p\u003e\n\u003col start=\"2\"\u003e\n \u003cli\u003eCross-Dimensional Resource Cascade Effects\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eDifferential pathway strengths reveal meaningful stressor-mediator specificity within simultaneous operation:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eIntrinsic Stressors (Social Comparison, Privacy Concerns):\u0026nbsp;Slightly stronger effects on disempowerment (\u0026beta; = 0.41) than psychosomatic tension (\u0026beta; = 0.37), with 10.8% differential. This pattern reflects theoretical expectation that autonomy-threatening stressors primarily activate loss-of-control appraisals. However, the substantial cross-dimensional effect on psychosomatic tension demonstrates cascading: threats to self-evaluation and autonomy resources consume emotional resources through sustained vigilance and rumination.\u003c/li\u003e\n \u003cli\u003eExtrinsic Stressors (Social Overload, Information Overload):\u0026nbsp;Substantially stronger effects on psychosomatic tension (\u0026beta; = 0.52) than disempowerment (\u0026beta; = 0.34), with 52.9% differential. Information and social overload directly deplete cognitive-emotional capacity, manifesting as fatigue. Yet the significant cross-dimensional effect on disempowerment (\u0026beta; = 0.34) demonstrates that overwhelming information environments reduce perceived mastery, and excessive social obligations overwhelm volitional control, generating loss-of-control appraisals.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThis differential-yet-simultaneous pattern precisely matches COR Theory\u0026apos;s prediction of resource cascade mechanisms: threatened resources in one domain necessitate consumption of alternative resources for restoration attempts, creating multidimensional manifestations.\u003c/p\u003e\n\u003col start=\"3\"\u003e\n \u003cli\u003eContemporary User Relevance\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe sample characteristics (mean tenure = 6.8 years, age modal = 25-29 years) capture users experiencing contemporary platform characteristics with maximum relevance. Unlike earlier studies including long-tenure early adopters, this specification ensures findings reflect current algorithmic and interface conditions shaping disengagement in the digital landscape that contemporary users actually experience. The R\u0026sup2; = 0.67 for disengagement using contemporary user data suggests findings are maximally relevant to understanding current digital phenomena.\u003c/p\u003e\n\u003cp\u003eCOMPARISON WITH LITERATURE AND EMPIRICAL ADVANCEMENT\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAddressing Prior Research Limitations:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003ePrior literature has operated under three critical limitations:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003eFragmented Stressor Examination:\u0026nbsp;Existing research typically examines single or paired stressor categories. This study\u0026apos;s comprehensive inclusion of four stressor dimensions enables specification of stressor interactions and relative contributions to disengagement mechanisms.\u003c/li\u003e\n \u003cli\u003eUnidirectional Mechanism Assumption:\u0026nbsp;Predominant frameworks assume stress operates through single pathways. The dual-pathway simultaneous mediation model demonstrates that disengagement emerges from parallel activation of multiple mechanisms.\u003c/li\u003e\n \u003cli\u003eMissing Cross-Dimensional Effects:\u0026nbsp;No published research has examined whether intrinsic stressors activate extrinsic-stressor-typical manifestations or vice versa. This study provides the first direct evidence that resource depletion operates across psychological domains.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cem\u003eEmpirical Advancement Over Meta-Analyses:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eRecent meta-analytical reviews synthesizing 150+ studies confirm associations between social media use and psychological distress, yet report wide effect size heterogeneity (r = 0.18 to 0.64). This heterogeneity likely derives from unmeasured mechanism complexity. The present study\u0026apos;s integrated measurement of both mechanisms within a single large sample (N = 1,500) clarifies that heterogeneity reflects genuine mechanism complexity: both pathways operate substantially and simultaneously.\u003c/p\u003e\n\u003cp\u003ePRACTICAL IMPLICATIONS FOR STAKEHOLDER ENGAGEMENT\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFor Social Media Developers:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eInformation overload demonstrates the strongest effect on psychosomatic tension (\u0026beta; = 0.52). Platform redesign should prioritize content filtering mechanisms allowing substantive user control over information volume rather than maximizing information presentation. Social overload similarly demands implementation of absence signals and asynchronous messaging reducing perceived responsiveness obligations. Privacy concerns require genuine transparency enabling perceived control. Social comparison mitigation demands algorithmic reconsideration of curated content amplification.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFor Mental Health Service Providers:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe identification of simultaneous disempowerment (\u0026beta;=0.47\u0026rarr;DD) and psychosomatic tension (\u0026beta;=0.40\u0026rarr;DD) pathways requires clinically-integrated intervention:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDisempowerment pathway intervention\u003c/em\u003e:\u0026nbsp;\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eCognitive Reframing: Help clients distinguish between \u0026apos;platform algorithms controlling information\u0026apos; vs. \u0026apos;personal lack of control\u0026rsquo; reduces internalized helplessness\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMedia Literacy: Educate regarding algorithmic curation (posts are filtered, not representative reality) restores perceived agency through informed understanding\u003c/li\u003e\n \u003cli\u003ePlatform Literacy: Teach use of privacy settings, feed control features, notification management restores perceived mastery/autonomy - Implementation: Group cognitive-behavioral intervention addressing platform-specific helplessness cognitions (6 sessions, 1.5 hr each)\u0026nbsp;\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cem\u003epsychosomatic pathway intervention:\u003c/em\u003e\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eStress Management: Standard CBT techniques for anxiety/fatigue management (progressive muscle relaxation, deep breathing during platform use) - Sleep Hygiene: Address screen-time\u0026apos;s sleep disruption (blue light, stimulating content) through bedroom tech boundaries\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eAffective Monitoring: Emotion tracking during/after platform use to identify triggers producing psychosomatic symptoms\u003c/li\u003e\n \u003cli\u003eGradual Exposure: Systematic desensitization where clients use platforms in controlled, time-limited manner while monitoring stress responses\u003c/li\u003e\n \u003cli\u003eImplementation: Individual behavioral intervention (8-10 sessions) targeting platform-related stress symptom reduction\u0026nbsp;\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cem\u003ecombined approach:\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSimultaneous pathway activation implies single-pathway interventions risk incompleteness. Most effective clinical approach combines cognitive reframing + stress management, each targeting pathway-specific mechanisms. Efficacy metrics should assess both outcomes: (1) disempowerment reduction (sense of control restoration) + (2) psychosomatic tension reduction (stress symptom alleviation)..\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFor Policymakers and Regulators:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eEvidence that simultaneous stress pathways drive disengagement at population level (R\u0026sup2; = 0.67) suggests substantial policy-relevant public health implications. Regulatory frameworks should elevate algorithmic transparency to equivalent priority as data protection. Platforms should disclose known psychological effects comparable to pharmaceutical labeling. Duty-of-care obligations should address high-use populations at elevated risk.\u003c/p\u003e\n\u003cp\u003eVALIDITY OF MEASUREMENT AND ANALYTICAL APPROACH\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMeasurement Model Validation:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAll measurement criteria were satisfied. Indicator loadings exceeded 0.75, Composite Reliability ranged 0.87-0.93, and Discriminant Validity was established via HTMT \u0026lt;0.85. This rigor ensures reported associations reflect true construct relationships rather than measurement artifact.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAnalytical Appropriateness:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eVariance-based PLS-SEM represents optimal approach for this investigation: (1) handles complex multivariate models with multiple mediators; (2) prioritizes predictive accuracy enabling robust mediation testing; and (3) accommodates theoretical focus on specific postulated mechanisms. Two-stage analysis (measurement validation followed by structural testing) ensures observed relationships reflect true construct associations.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRobustness and Generalizability:\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMulti-group analysis across demographic subgroups (age cohorts, gender, platforms) revealed no significant between-group structural differences, supporting generalizability across demographic diversity within the 18-35 age specification. Common method bias assessment via Harman\u0026apos;s test (30.8% variance, below 50% threshold) confirms that reported associations are not artificially inflated.\u003c/p\u003e\n\u003cp\u003eLIMITATIONS AND BOUNDARY CONDITIONS\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMethodological Limitations:\u003c/em\u003e\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003eCross-Sectional Design:\u0026nbsp;Precludes definitive causal inference. Longitudinal and experimental designs would strengthen confidence.\u003c/li\u003e\n \u003cli\u003eSelf-Report Measurement:\u0026nbsp;Introduces bias despite procedural mitigation. Behavioral indicators (API data, real-time assessment) would provide convergent validity.\u003c/li\u003e\n \u003cli\u003eGeographic Specification:\u0026nbsp;Indonesia-specific recruitment limits generalizability. Cross-cultural replication is critical.\u003c/li\u003e\n \u003cli\u003eAge/Tenure Window:\u0026nbsp;18-35 age and \u0026le;10-year tenure excludes other demographic segments. Platform-specific moderation analyses warrant investigation.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cem\u003eTheoretical Boundary Conditions:\u003c/em\u003e\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003ePlatform-Specific Effects:\u0026nbsp;This study examined generalized social media stress. Platform-specific mechanisms (Instagram visual comparison vs. X text-focused) may differ substantially.\u003c/li\u003e\n \u003cli\u003eUser Motivation Moderation:\u0026nbsp;Individual differences in use motivations may moderate stress mechanisms.\u003c/li\u003e\n \u003cli\u003ePersonality Differences:\u0026nbsp;Trait neuroticism and baseline perceived control likely moderate relationships.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eFUTURE RESEARCH DIRECTIONS\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePriority 1: Longitudinal and Experimental Designs\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFuture research should employ longitudinal designs enabling causal inference. Panel studies tracking participants across 6-12 months would examine temporal stress-disengagement sequences and whether disengagement produces stress reduction. Experimental manipulation studies randomly assigning users to algorithmic exposure conditions would provide causal evidence complementing observational data.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePriority 2: Platform-Specific and Moderation Analyses\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMulti-platform studies could isolate which stressor types are most salient on specific platforms. Moderation analyses examining personality, motivations, and usage patterns could identify user profiles most vulnerable to documented mechanisms, enabling targeted intervention design.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePriority 3: Intervention Development and Testing\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eRandomized controlled trials should test interventions targeting identified stress mechanisms. Intervention conditions might include algorithmic transparency enhancement, cognitive reframing, guided disengagement protocols, and combination interventions. Efficacy evaluation would enable evidence-based recommendations regarding highest-impact modalities.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePriority 4: Cross-Cultural Replication\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eReplication in culturally-diverse contexts would clarify which findings generalize universally versus reflect cultural specificity.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eAn empirical study of 1,500 Indonesian social media users (aged 18\u0026ndash;35, tenure\u0026thinsp;\u0026le;\u0026thinsp;10 years) validated a Conservation of Resources Theory model with R\u0026sup2; = 0.67, substantially exceeding digital stress literature standards (0.35\u0026ndash;0.50). Digital disengagement stems from two equally potent, independent mechanisms: loss-of-control appraisals (disempowerment: 0.706) and psychosomatic stress manifestations (0.712) both triggered by intrinsic and extrinsic stressors. This dual-pathway model demonstrates that resource depletion operates simultaneously across multiple psychological domains rather than activating isolated responses.\u003c/p\u003e \u003cp\u003eThree stakeholders require action: platform developers must redesign algorithmic systems to reduce psychological burden; mental health practitioners must implement dual-pathway interventions targeting both disempowerment and psychosomatic manifestations; policymakers must establish regulatory frameworks prioritising algorithmic transparency alongside data protection.\u003c/p\u003e \u003cp\u003eMethodological limitations include cross-sectional design, self-reported measures, Indonesia-specific recruitment, and restricted age/tenure parameters. Future research demands longitudinal designs, cross-cultural replication, and neurobiological investigation. As artificial intelligence exponentially amplifies user demands, empirically mapping stress mechanisms becomes urgently necessary for protecting psychological wellbeing at population scale.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eData Availability Statement:\u003c/p\u003e\n\u003cp\u003eThe data supporting these findings are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003eFunding Statement:\u003c/p\u003e\n\u003cp\u003eThis research was funded by the Jampang Pratama Foundation through a Research Grant for Lecturers Studying at the Polytechnic istikom bina citra informatika in 2025 with the number 207/K01.YBP/SK/III/2025.\u003c/p\u003e\n\u003cp\u003eConflict of Interest Statement:\u003c/p\u003e\n\u003cp\u003eThe authors declare that there are no financial or non-financial conflicts of interest that could influence the results or interpretation of this research.\u003c/p\u003e\n\u003cp\u003eAuthor Contributions\u003c/p\u003e\n\u003cp\u003eICT: Conceptualization, Formal analysis, Methodology, Writing \u0026ndash; original draft, Data curation, Validation, Funding acquisition, Writing \u0026ndash; review \u0026amp; editing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSH: Conceptualization, Methodology, Project administration, Resources, Supervision, Validation, Writing \u0026ndash; review \u0026amp; editing, Data curation.\u003c/p\u003e\n\u003cp\u003eNR: Conceptualization, Supervision, Validation, Writing \u0026ndash; review \u0026amp; editing, Methodology.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAcquisti, A., Brandimarte, L., \u0026amp; Loewenstein, G. (2015). Privacy and human behavior in the age of information. \u003cem\u003eScience\u003c/em\u003e, \u003cem\u003e347\u003c/em\u003e(6221), 509\u0026ndash;514. https://doi.org/10.1126/science.aaa1465\u003c/li\u003e\n \u003cli\u003eDavis, F. D. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. \u003cem\u003eMIS Quarterly\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(3), 319\u0026ndash;340. https://doi.org/10.2307/249008\u003c/li\u003e\n \u003cli\u003eDhir, A., Kaur, P., Chen, S., \u0026amp; Pallesen, S. (2019). Antecedents and consequences of social media fatigue. \u003cem\u003eInt. J. Inf. Manag.\u003c/em\u003e, \u003cem\u003e48\u003c/em\u003e, 193\u0026ndash;202. https://doi.org/10.1016/j.ijinfomgt.2019.05.021\u003c/li\u003e\n \u003cli\u003eDwivedi, Y. K., Hughes, L., Baabdullah, A. M., Ribeiro-Navarrete, S., Giannakis, M., Al-Debei, M. M., Dennehy, D., Metri, B., Buhalis, D., Cheung, C. M. K., Conboy, K., Doyle, R., Dubey, R., Dutot, V., Felix, R., Goyal, D. P., Gustafsson, A., Hinsch, C., Jebabli, I., \u0026hellip; Wamba, S. F. (2022). Metaverse beyond the hype: Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. \u003cem\u003eInternational Journal of Information Management\u003c/em\u003e, \u003cem\u003e66\u003c/em\u003e, 102542. https://doi.org/10.1016/j.ijinfomgt.2022.102542\u003c/li\u003e\n \u003cli\u003eErdem, H. D., Herrero, J., \u0026amp; Urue\u0026ntilde;a, A. (2025). The Bidirectional Relationships between Social Pressure in Digital Contexts, Depression, and Social Support over Time. \u003cem\u003ePsychosocial Intervention\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 189\u0026ndash;200. https://doi.org/10.5093/pi2025a15\u003c/li\u003e\n \u003cli\u003eHobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. \u003cem\u003eAmerican Psychologist\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(3), 513\u0026ndash;524. https://doi.org/10.1037/0003-066X.44.3.513\u003c/li\u003e\n \u003cli\u003eHobfoll, S. E., Halbesleben, J., Neveu, J.-P., \u0026amp; Westman, M. (2018). Conservation of Resources in the Organizational Context: The Reality of Resources and Their Consequences. \u003cem\u003eAnnual Review of Organizational Psychology and Organizational Behavior\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(1), 103\u0026ndash;128. https://doi.org/10.1146/annurev-orgpsych-032117-104640\u003c/li\u003e\n \u003cli\u003eHolmgreen, L., Tirone, V., Gerhart, J., \u0026amp; Hobfoll, S. E. (2017). Conservation of Resources Theory. In \u003cem\u003eThe Handbook of Stress and Health\u003c/em\u003e (pp. 443\u0026ndash;457). Wiley. https://doi.org/10.1002/9781118993811.ch27\u003c/li\u003e\n \u003cli\u003eJie, Z., Mou, Q.-F., Zheng, T., Gao, F., Zhong, Y., Lu, Q., Gao, Y., \u0026amp; Zhao, M. (2023). A serial mediation model of social media addiction and college students\u0026rsquo; academic engagement: the role of sleep quality and fatigue. \u003cem\u003eBMC Psychiatry\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e. https://doi.org/10.1186/s12888-023-04799-5\u003c/li\u003e\n \u003cli\u003eKatz, E., Blumler, J. G., \u0026amp; Gurevitch, M. (1973). Uses and Gratifications Research. \u003cem\u003eThe Public Opinion Quarterly\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(4), 509\u0026ndash;523. http://www.jstor.org/stable/2747854\u003c/li\u003e\n \u003cli\u003eLo, J. (2019). Exploring the buffer effect of receiving social support on lonely and emotionally unstable social networking users. \u003cem\u003eComput. Hum. Behav.\u003c/em\u003e, \u003cem\u003e90\u003c/em\u003e, 103\u0026ndash;116. https://doi.org/10.1016/j.chb.2018.08.052\u003c/li\u003e\n \u003cli\u003eMalik, A., Dhir, A., Kaur, P., \u0026amp; Johri, A. (2020). Correlates of social media fatigue and academic performance decrement: A large cross-sectional study. \u003cem\u003eInf. Technol. People\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 557\u0026ndash;580. https://doi.org/10.1108/itp-06-2019-0289\u003c/li\u003e\n \u003cli\u003eMojtabai, R. (2024). Problematic social media use and psychological symptoms in adolescents. \u003cem\u003eSocial Psychiatry and Psychiatric Epidemiology\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e, 2271\u0026ndash;2278. https://doi.org/10.1007/s00127-024-02657-7\u003c/li\u003e\n \u003cli\u003ePang, H., \u0026amp; Zhang, K. (2024). Determining multidimensional influences of network heterogeneity on university students\u0026rsquo; psychological and academic well-being: The mediating role of social network exhaustion. \u003cem\u003eHeliyon\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e. https://doi.org/10.1016/j.heliyon.2024.e32328\u003c/li\u003e\n \u003cli\u003eShort, John., Williams, Ederyn., \u0026amp; Christie, Bruce. (1976). \u003cem\u003eThe social psychology of telecommunications\u003c/em\u003e. Wiley.\u003c/li\u003e\n \u003cli\u003eSolomon, R. L., \u0026amp; Corbit, J. D. (1974). An opponent-process theory of motivation: I. Temporal dynamics of affect. \u003cem\u003ePsychological Review\u003c/em\u003e, \u003cem\u003e81\u003c/em\u003e(2), 119\u0026ndash;145. https://doi.org/10.1037/h0036128\u003c/li\u003e\n \u003cli\u003eSong, H., Zmyslinski-Seelig, A., Kim, J., Drent, A., Victor, A., Omori, K., \u0026amp; Allen, M. (2014). Does Facebook make you lonely?: A meta analysis. \u003cem\u003eComputers in Human Behavior\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e, 446\u0026ndash;452. https://doi.org/10.1016/j.chb.2014.04.011\u003c/li\u003e\n \u003cli\u003eVan Der Schuur, W., Baumgartner, S., \u0026amp; Sumter, S. (2019). Social Media Use, Social Media Stress, and Sleep: Examining Cross-Sectional and Longitudinal Relationships in Adolescents. \u003cem\u003eHealth Communication\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 552\u0026ndash;559. https://doi.org/10.1080/10410236.2017.1422101\u003c/li\u003e\n \u003cli\u003eYue, Z., Zhang, R., \u0026amp; Xiao, J. (2021). Passive social media use and psychological well-being during the COVID-19 pandemic: The role of social comparison and emotion regulation. \u003cem\u003eComputers in Human Behavior\u003c/em\u003e, \u003cem\u003e127\u003c/em\u003e, 107050. https://doi.org/10.1016/j.chb.2021.107050\u003c/li\u003e\n \u003cli\u003eZhang, S., Shen, Y., Xin, T., Sun, H., Wang, Y., Zhang, X., \u0026amp; Ren, S. (2021). The development and validation of a social media fatigue scale: From a cognitive-behavioral-emotional perspective. \u003cem\u003ePLoS ONE\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e. https://doi.org/10.1371/journal.pone.0245464\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"digital disengagement, social media stress, conservation of resources, simultaneous mediation, psychosomatic tension, disempowerment","lastPublishedDoi":"10.21203/rs.3.rs-8396588/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8396588/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study examines the mechanisms underlying digital disengagement among social media users by testing an integrated Conservation of Resources (COR) Theory model. It investigates how both intrinsic stressors (social comparison, privacy concerns) and extrinsic stressors (social overload, information overload) activate dual psychological pathways disempowerment and psychosomatic tension leading to platform disengagement. A quantitative, cross-sectional study was conducted with N\u0026thinsp;=\u0026thinsp;1,500 Indonesian social media users (tenure\u0026thinsp;\u0026le;\u0026thinsp;10 years, age 18\u0026ndash;35). Data were collected via online survey and analyzed using variance-based structural equation modeling (PLS-SEM) with SmartPLS 4.0. Measurement validity was established through confirmatory factor analysis, and mediation pathways were tested via bootstrapping (5,000 resamples). All eight hypothesized relationships received empirical support (p\u0026thinsp;\u0026lt;\u0026thinsp;.001). Intrinsic and extrinsic stressors demonstrated significant direct effects on both disempowerment and psychosomatic tension. Both mediators independently predicted digital disengagement through simultaneous pathways with nearly equivalent indirect effects (disempowerment\u0026thinsp;=\u0026thinsp;0.706; psychosomatic tension\u0026thinsp;=\u0026thinsp;0.712). The integrated model explained 67% of disengagement variance, substantially exceeding typical effect sizes in digital stress literature (R\u0026sup2; = 0.35\u0026ndash;0.50). This study provides the first empirical validation of COR Theory's cascading-effect mechanism across simultaneous psychological domains in social media contexts. It advances beyond compartmentalized frameworks by demonstrating cross-dimensional effects and establishing that resource depletion operates through parallel, equally potent mechanisms rather than single pathways.\u003c/p\u003e","manuscriptTitle":"Beyond the Scroll: Unmasking the Simultaneous Role of Psychosomatic and Stressors Behind Social Media Users’ Disempowerment and Disengagement","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-12 19:40:55","doi":"10.21203/rs.3.rs-8396588/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7a39ed50-9eea-498c-bfd4-441cc18cd349","owner":[],"postedDate":"February 12th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":62637888,"name":"Biological sciences/Neuroscience"},{"id":62637889,"name":"Biological sciences/Psychology"},{"id":62637890,"name":"Social science/Psychology"}],"tags":[],"updatedAt":"2026-02-17T19:54:35+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-12 19:40:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8396588","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8396588","identity":"rs-8396588","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}