Introduction

The rise of digital gig platforms has brought unprecedented flexibility to the labor market while simultaneously exposing gig workers to unique challenges within their work environment. Unlike employees in traditional employment relationships, platform-based gig workers typically operate in a “de-organized” environment characterized by a lack of stable organizational affiliation, formal managerial structures, and continuous professional support (Kadolkar et al., 2024). Although these platforms offer considerable autonomy in task selection, scheduling, and workspace, such flexibility often comes at the cost of increased work isolation, heightened career uncertainty, and weakened social support networks (Zhao and Liu, 2024a). Specifically, in traditional organizational settings, social support networks are embedded within hierarchical structures and continuous interpersonal interactions, encompassing diverse sources such as organizations, supervisors, colleagues, and families. These networks provide emotional, instrumental, informational, and evaluative resources that collectively shape individuals’ psychological states and behavioral responses (Chen et al., 2022). By contrast, the de-organized nature of digital gig platforms fundamentally reshapes this landscape: algorithmic management largely supplants traditional supervisory functions, customer review systems become the primary mechanism for evaluating and feeding back on work performance, and the high degree of task fragmentation undermines collegial support, leaving gig workers’ social support networks heavily dependent on platform algorithms, customer reviews, and family emotional interactions (Hajiheydari and Delgosha, 2024). This structural shift indicates that social support within traditional organizations is increasingly insufficient to fully explain the mechanisms shaping gig workers’ psychological states and work performance in platform contexts. Consequently, a systematic investigation of how multidimensional social support on digital gig platforms impacts gig workers’ work performance has become both a critical and timely research endeavor. This study not only extends the explanatory scope of Social Support Theory to emerging work forms but also holds significant practical implications for optimizing platform governance and safeguarding gig workers’ rights.

Existing research indicates that gig workers on digital platforms generally face multidimensional support deficits (Mukherjee and Singh, 2024; Mukherjee et al., 2025). First, platform organizational support is often limited to information provision, technical tools, and compensation mechanisms, while insufficiently addressing workers’ emotional needs and professional development (Caza et al., 2022). Second, the uncertainty and outcome-oriented nature of customer review support create persistent psychological pressures, with negative ratings directly affecting income levels and subsequent work opportunities (Liang et al., 2025). Moreover, the blurred boundaries between work and family life, coupled with the inherent independence of platform work, render family emotional support increasingly critical in alleviating work stress (James, 2023). These characteristics highlight the inadequacy of traditional forms of social support, such as resource provision, leadership care, and coworker collaboration, in explaining work performance in the context of digital gig platforms (Ángeles López-Cabarcos et al., 2022). Although prior studies have examined factors influencing gig workers’ work performance on digital platforms, such as platform design, task complexity, algorithmic fairness, worker experience, job demands–resources alignment, and work autonomy (Mohsin et al., 2022; Watson et al., 2021; Zhang et al., 2024; Zhu et al., 2024a), there remains a lack of systematic theoretical development and empirical validation regarding the specific composition of social support networks in de-organized contexts and their mechanisms of impact on gig workers’ work performance (Mukherjee et al., 2025).

To address this gap, the present study integrates Social Support Theory and Affective Events Theory to develop a chain mediation model that systematically examines how platform organizational support, customer review support, and family emotional support influence gig workers’ work performance through work engagement and burnout. Social Support Theory emphasizes the interactive mechanisms between individuals and their social networks, explaining how emotional, instrumental, informational, and appraisal support shape psychological states and behavioral responses (Chen et al., 2022). Drawing on this framework, we propose a tripartite social support structure tailored to the digital platform ecosystem: platform organizational support primarily entails instrumental and informational support, such as algorithmic scheduling, task matching, and performance feedback to enhance efficiency (Lang et al., 2023); customer review support reflects appraisal support through user ratings and reviews that offer social feedback and behavioral cues (Wu and Huang, 2024); family emotional support provides emotional comfort, helping gig workers navigate the instability and flexibility of their work environment (Nahum-Shani et al., 2011; King et al., 1995).

Second, Affective Events Theory provides a dynamic lens for understanding the psychological processes through which affective events influence individual behavior, explaining how specific events trigger affective reactions and changes in psychological states that ultimately shape behavior (Junça-Silva et al., 2021). Within the integrative framework of this study, Social Support Theory focuses on the sources and structure of supportive events, illuminating how gig workers obtain support from multiple relational domains, including the platform, customers, and family (Chen et al., 2022), whereas Affective Events Theory emphasizes the mechanisms of emotional state formation, clarifying how these social support events influence gig workers’ work performance through pathways of work engagement and burnout (Zhu et al., 2022). The integration of these frameworks not only illuminates how gig workers on digital platforms acquire resources through diverse social relationships but also explicates how multi-dimensional social support (affective events) triggers their work engagement and burnout (affective and psychological reactions), subsequently influencing their work performance (behavioral response), thereby extending the explanatory power and applicability of both theories within this emerging form of employment.

Furthermore, this study incorporates decision-making autonomy as a critical moderating variable. Decision-making autonomy refers to the extent to which individuals can independently determine how tasks are executed, schedules are arranged, and work pace is managed (Clinton and Conway, 2024). In organizational behavior and occupational psychology, it is regarded as a vital contextual resource that enhances employees’ sense of behavioral control and fosters intrinsic motivation and work effectiveness (Liu et al., 2024a). Within the context of digital gig platforms, decision-making autonomy exhibits both distinctiveness and complexity. Although platforms formally grant gig workers the right to select tasks and schedule their time, this autonomy is inherently embedded within algorithm-driven frameworks. Algorithmic management mechanisms such as dynamic pricing, task-matching rules, performance evaluation systems, and feedback ratings, while providing an appearance of freedom, simultaneously impose structural constraints and subtle controls on workers’ decision-making scope (Zhu et al., 2024b). Consequently, decision-making autonomy on these platforms can be understood as a bounded discretion within prescribed rules, serving both as a critical contextual resource and as a challenge to gig workers’ self-management capabilities in the absence of formal organizational support (Wu and Huang, 2024). This characteristic renders decision-making autonomy a complex yet pivotal factor in shaping gig workers’ work states and work performance. Accordingly, this study aims to investigate how platform-based decision-making autonomy moderates the relationships between work engagement and work performance, as well as between burnout and work performance, offering insights for optimizing gig work arrangements and enhancing performance outcomes.

In summary, this study develops a novel chain mediation model grounded in Social Support Theory and Affective Events Theory to systematically explore how platform organizational support, customer review support, and family emotional support (as affective events) affect gig workers’ work engagement and burnout (as affective and psychological reactions), and how these factors further affect work performance (as behavioral responses). It further investigates the moderating role of decision-making autonomy in the relationships between work engagement and work performance and between burnout and work performance. This research seeks to address the following key questions: (1) How do platform organizational, customer review, and family emotional support influence gig workers’ work engagement and burnout? (2) Can work engagement mitigate burnout among gig workers? (3) How do work engagement and burnout affect work performance? (4) How do the three forms of support indirectly affect work performance via work engagement and burnout? (5) How does decision-making autonomy moderate the relationships between work engagement and work performance, and between burnout and work performance?

This study offers several theoretical and practical innovations. First, while prior research has largely focused on single-dimensional social support within traditional organizational contexts, systematic investigations of social support networks in “de-organized” work environments, particularly digital gig platforms, remain scarce (Ángeles López-Cabarcos et al., 2022; Nahum-Shani et al., 2011). Drawing on Social Support Theory, this study is the first to construct a three-dimensional interactive model encompassing platform organizational support, customer review support, and family emotional support, comprehensively elucidating the differentiated mechanisms through which multi-dimensional social support influences gig workers’ work performance, thereby providing a systematic theoretical framework for understanding these dynamics (Lang et al., 2023; Nahum-Shani et al., 2011; Wu and Huang, 2024). Second, by integrating Affective Events Theory, this study is the first to reveal a sequential mechanism in a de-organized context: affective events (social support as) → affective and psychological reactions (work engagement/burnout) → behavioral responses (work performance). Social Support Theory focuses on the sources and structures of supportive events, while Affective Events Theory highlights the mechanisms through which these events elicit emotional responses and shape psychological states. Together, they illuminate how gig workers access resources through diverse social ties and how social support, as a form of affective event, elicits affective and psychological reactions, namely work engagement and burnout, that subsequently shape work performance. This integration extends the theoretical boundaries of both frameworks within the emerging employment context of digital gig platforms (Hajiheydari and Delgosha, 2024). Third, this study is the first to uncover the double-edged effect of decision-making autonomy. While high autonomy can undermine the positive impact of work engagement due to cognitive overload and resource scarcity, it can also alleviate the negative impact of burnout through space for self-regulation (Clinton and Conway, 2024). This paradox highlights the need for platforms to dynamically calibrate autonomy based on task complexity and workers’ skill levels, rather than adhering to a simplistic “more autonomy is better” approach, offering theoretical guidance for balancing flexibility in algorithm design. Finally, the heterogeneity test revealed that the work performance of high-skilled gig workers is more susceptible to the adverse effects of burnout, whereas the performance of low-skilled gig workers remains largely unaffected under burnout conditions, likely due to their pressing economic needs (Behl et al., 2022; Jing et al., 2023). This finding challenges the universality of the traditional Job Demands–Resources model and underscores the necessity for platforms to adopt differentiated support strategies, prioritizing psychological relief for high-skilled workers and strengthening income security mechanisms for low-skilled workers, thereby achieving more precise and human-centered resource allocation.

This paper is organized into five main sections. Chapter 2 defines work performance within the context of digital gig platforms, reviews relevant literature, and develops research hypotheses based on the theoretical framework. Chapter 3 outlines the questionnaire design and data collection procedures. Chapter 4 evaluates the validity of the measurement and structural models, examines mediation and moderation effects, and analyzes the empirical results. Chapter 5 discusses the research conclusions, theoretical contributions, practical implications, limitations, and directions for future research. The Appendix provides the survey instruments used in the study.

Literature review

Work performance on digital gig platforms

Digital gig platforms refer to online labor marketplaces that facilitate short-term, task-oriented matches between service providers (such as freelancers or independent contractors) and service buyers (clients) (Kadolkar et al., 2024). Within this highly flexible yet de-organized work environment, gig workers face challenges such as task instability, the absence of formal employment relationships, and algorithm-driven control exerted by the platform (Wood et al., 2019; Wu and Huang, 2024). These structural features render gig workers’ performance largely dependent on external support systems rather than traditional organizational resources or individual capabilities (Jolly et al., 2021; Nahum-Shani et al., 2011). While a few scholars have employed frameworks such as self-determination theory, social exchange theory, conservation of resources theory, and resource conservation theory to explore the effects of psychological capital, individual characteristics, job satisfaction, and contextual resources on gig workers’ performance (Mohsin et al., 2022; Watson et al., 2021; Zhang et al., 2024), there remains a lack of systematic investigation into how emotional, instrumental, informational, and evaluative forms of social support perceived by gig workers influence their work performance within digital gig platforms.

Unlike traditional organizations, where formal support is typically provided by the organization, supervisors, or colleagues, gig workers rely on more informal and fragmented support networks composed primarily of three sources: platform organizational support (instrumental and informational support), such as task recommendation mechanisms, technological tools, and monetary rewards; customer review support (evaluative support), including ratings and positive feedback; and family emotional support (emotional support), such as emotional understanding and assistance with time coordination (Belanche et al., 2021; Lang et al., 2023; Nahum-Shani et al., 2011). Research within traditional employment contexts has shown that the effective allocation of organizational and individual resources, along with organizational and family support, significantly enhances employee performance, improves interpersonal relationships, and reduces job stress (Ángeles López-Cabarcos et al., 2022; Jolly et al., 2021). However, digital gig platforms differ significantly from traditional employment models. Platform workers are often subject to algorithmic control, job insecurity, and a lack of social protection, exposing them to elevated social and psychological stress (Wood et al., 2019). Simultaneously, the high degree of flexibility, temporariness, and autonomy in gig work demands strong self-management capabilities. In the absence of formal organizational support or a sense of belonging, gig workers’ attitudes and behaviors are more susceptible to the influences of platform resource allocation, customer feedback, and family support (Wu and Huang, 2024). Building on this context, the present study integrates Social Support Theory and Affective Events Theory to examine how platform organizational support, customer review support, and family emotional support influence work performance through their effects on work engagement and burnout. This integrated perspective offers novel theoretical insights and practical implications for understanding the mechanisms underlying performance in platform-based work environments.

Social support theory and affective events theory

Social Support Theory offers a theoretical framework that focuses on the interactive mechanisms between individuals and their social networks, emphasizing the beneficial role of supportive resources in regulating psychological states and behavioral responses (Chen et al., 2022; Gillman et al., 2023). According to this theory, social support provided by relatives, friends, colleagues, partners, and other social ties offers emotional value or practical utility, helping individuals better cope with external stressors and improve their self-regulation and adaptability (Ibrahim et al., 2024). Social support is generally divided into four core dimensions: First, emotional support, which refers to understanding, care, and trust gained through social interactions, contributing to a sense of belonging and emotional security; second, instrumental support, which refers to direct assistance provided by others in terms of material, resources, or actions; third, informational support, which entails access to problem-solving or decision-making information obtained through social exchanges; and fourth, appraisal support, which provides standards for self-evaluation and behavioral adjustment based on feedback from others (Chen et al., 2022). This study innovatively integrates Social Support Theory into the context of digital gig platforms, establishing a three-dimensional social support analysis framework tailored to the platform’s ecosystem characteristics: platform organizational support, which manifests as instrumental and informational support, such as task scheduling optimization, task matching mechanisms, and performance feedback systems to improve work efficiency (Lang et al., 2023); customer review support, primarily reflecting appraisal support, which refers to the social feedback and behavioral reference provided by user ratings and reviews (Wu and Huang, 2024); and family emotional support, which offers crucial emotional support, providing individuals with emotional comfort and psychological reliance in a highly flexible and unstable work environment (Nahum-Shani et al., 2011; King et al., 1995). This theoretical extension not only enriches the application of Social Support Theory in the context of the digital economy but also provides robust theoretical support for exploring the heterogeneous impact of multiple social support pathways on gig work performance.

Affective Events Theory posits that emotionally charged events in the workplace trigger specific affective and psychological reactions, which in turn influence individual behavior (Junça-Silva et al., 2021). The theory comprises three core elements: first, “events,” referring to specific situations or stimuli encountered at work; second, “affective and psychological reactions,” denoting the immediate emotional and cognitive states elicited by these events; and third, “behavioral responses,” which encompass the actions driven by these preceding reactions (Russell et al., 2012; Zhu et al., 2022). The theory underscores the mediating role of affective and psychological reactions in linking external events to behavioral outcomes (Russell et al., 2012). In this study, the theory serves as a process-oriented framework for understanding how social support transforms into affective and psychological reactions among gig workers and, subsequently, affects their behavioral responses. We conceptualize platform organizational, customer review, and family emotional support as affective events in the digital gig work context. When gig workers perceive positive social support from these sources, they are more likely to experience favorable affective and psychological reactions, manifested as work engagement characterized by vigor, dedication, and absorption, which in turn promote positive behaviors and improved work performance (Chompukum and Vanichbuncha, 2025). Conversely, the absence of support or exposure to negative feedback may trigger adverse affective and psychological reactions, leading to burnout, marked by emotional exhaustion, depersonalization, and a diminished sense of accomplishment, ultimately resulting in performance decline (Tong and Spitzmueller, 2024). Within the affective events framework, work engagement and burnout function as pivotal emotional and psychological mediators that indirectly shape gig workers’ performance outcomes (Nimon et al., 2023; Tong and Spitzmueller, 2024).

In summary, this study integrates Social Support Theory and Affective Events Theory to establish a comprehensive and complementary theoretical foundation and analytical framework. Social Support Theory focuses on the structural and interactional sources of support, explaining how gig workers obtain assistance and resources from diverse social connections (Chen et al., 2022). In contrast, Affective Events Theory emphasizes the mechanisms of affective and psychological state generation, elucidating how events influence individual behavior through these pathways (Chompukum and Vanichbuncha, 2025). Integrating these two perspectives enables a full-chain analysis from “external social support events” to “affective and psychological reactions” to “behavioral responses.” This approach not only illustrates how multi-source social support on digital gig platforms impacts work performance via work engagement and burnout but also extends the applicability of both theories within platform-mediated labor environments.

Research hypotheses

Grounded in Social Support Theory and Affective Events Theory, this study systematically explores the relationships and underlying mechanisms between multidimensional social support and work performance among gig workers on digital platforms. Based on this framework, eleven hypotheses are proposed, as illustrated in Fig. 1.

Fig. 1
Fig. 1
Full size image

Theoretical model of work performance.

Social support and work engagement

Social support refers to the resources individuals obtain through social connections with others, groups, or communities, and is generally categorized into workplace support and private life support (Nasurdin et al., 2018). On digital gig platforms, workplace support is primarily manifested in the form of platform organizational support and customer review support. Private life support mainly stems from close familial relationships and emotional care, which promote physical and mental well-being and help alleviate work-related stress (Gillman et al., 2023). Research shows that digital gig platforms provide work opportunities, technological tools, task guidance, and compensation through algorithmic management, granting workers a high degree of autonomy. This structure not only boosts workers’ confidence in handling job challenges but also aligns their behavior with platform goals, thereby enhancing work engagement (Lang et al., 2023).

Furthermore, positive customer interactions during service delivery, such as respectful communication and favorable reviews, can elicit pleasant emotional experiences, bolster psychological resources, reduce stress from negative customer encounters, and mitigate emotional exhaustion (Xiongtao et al., 2021). Family emotional support is equally vital in daily life. Stable and supportive family relationships can alleviate negative emotions and behavioral responses caused by workplace interpersonal conflicts, reduce absenteeism and burnout, and help workers better balance professional and personal demands, ultimately fostering greater work engagement and improving overall performance (Afsari et al., 2023). Based on these insights, the following hypotheses are proposed:

H1: Platform organizational support has a significant positive impact on work engagement.

H2: Customer review support has a significant positive impact on work engagement.

H3: Family emotional support has a significant positive impact on work engagement.

Social support and burnout

As an external resource, social support can stimulate internal motivation, strengthen workers’ sense of identity and belong, and effectively mitigate burnout caused by prolonged exposure to stress (Nahum-Shani et al., 2011). On digital gig platforms, algorithm driven support mechanisms such as task matching, operational guidance, and performance feedback enhance gig workers’ sense of purpose and control, thereby promoting greater work engagement. Moreover, since income and future job opportunities are closely tied to performance, gig workers are often incentivized to remain engaged despite experiencing burnout (Lang et al., 2023). Customer rating systems, a key component of platform evaluation, directly link positive reviews to task allocation and rewards. High ratings not only increase the likelihood of receiving new assignments but also provide additional incentives, which further motivate service quality improvement, increase engagement, and ultimately alleviate burnout (Xiongtao et al., 2021). Additionally, family emotional support plays a crucial role in helping gig workers cope with life stressors, improve emotional well-being, and enhance work-life balance, thereby reducing burnout (Afsari et al., 2023). Based on the above analysis, the following hypotheses are proposed:

H4: Platform organizational support has a significant negative impact on burnout.

H5: Customer review support has a significant negative impact on burnout.

H6: Family emotional support has a significant negative impact on burnout.

Work engagement and burnout

Work engagement and burnout represent two contrasting psychological states, each characterized by distinct dimensions and experiences (Tong and Spitzmueller, 2024). Work engagement comprises three key dimensions: vigor, dedication, and absorption (Schaufeli et al., 2002), whereas burnout is reflected in emotional exhaustion, depersonalization, and reduced personal accomplishment (Lei et al., 2024). Empirical research consistently finds that high levels of work engagement significantly reduce burnout (Gillet et al., 2024; Maslach and Jackson, 1981). Specifically, vigor helps maintain energy reserves and psychological resilience, effectively buffering against stress-induced exhaustion (Galanakis and Tsitouri, 2022); dedication fosters a strong sense of meaning and identification with work, enhancing self-efficacy and preventing disengagement (Chen et al., 2023); and absorption facilitates flow experiences, minimizes distractions, increases productivity, and fosters a sense of achievement, forming a positive feedback loop (Galanakis and Tsitouri, 2022). Drawing from this evidence, the following hypothesis is proposed:

H7: Work engagement has a significant negative impact on burnout.

Work engagement and work performance

Work performance is commonly defined as an employee’s overall effectiveness, productivity, and quality in task completion, typically assessed through observable behaviors and outcomes (Liu et al., 2024a). As a key indicator of work performance, work engagement plays a vital role in enhancing organizational performance (Mohsin et al., 2022). Within the context of digital gig platforms, work performance is particularly critical, as rating systems and customer feedback directly influence workers’ income and future task opportunities. Studies show that higher levels of work engagement significantly improve performance (Moreira et al., 2022). Engagement contributes to performance through three interrelated dimensions: vigor supports sustained energy and motivation; absorption enhances concentration during task execution; and dedication strengthens identification with job value and personal responsibility. Together, these dimensions optimize work efficiency, improve problem-solving capabilities, and ultimately enhance overall work performance (Moreira et al., 2022). Based on theoretical and empirical findings, the following hypothesis is proposed:

H8: Work engagement has a significant positive impact on work performance.

Burnout and work performance

Burnout is a chronic stress syndrome characterized by emotional exhaustion, depersonalization, and a diminished sense of personal accomplishment (Lei et al., 2024). Emotional exhaustion, the core symptom, results in depleted energy and diminished enthusiasm, leading to reduced efficiency (Zhou et al., 2014); depersonalization manifests as emotional detachment and indifference toward work and colleagues, potentially triggering interpersonal conflicts and undermining team collaboration (Nureni et al., 2023); a decline in personal accomplishment causes self-doubt, weakening the motivation to remain engaged (Lei et al., 2024). Empirical evidence highlights a significant negative relationship between burnout and work performance. Emotional exhaustion undermines task focus, depersonalization encourages passive coping strategies, and reduced accomplishment diminishes intrinsic motivation to perform well (Zhou et al., 2014; Nureni et al., 2023). This adverse effect is particularly pronounced on digital gig platforms, where performance declines not only affect task quality but also customer ratings, platform task distribution, and ultimately, economic returns and career prospects (Nureni et al., 2023). Thus, mitigating burnout is essential for improving work performance. Accordingly, the following hypothesis is proposed:

H9: Burnout has a significant negative impact on work performance.

The moderating role of decision-making autonomy

Decision-making autonomy refers to the degree of choice and independence an individual possesses in their work processes (Wu and Huang, 2024). Prior research suggests that moderate levels of decision-making autonomy can enhance employees’ sense of control, self-recognition of competence, and perceived respect. It also offers individuals opportunities for self-actualization and personal development, thereby strengthening intrinsic motivation and improving both job satisfaction and work performance (Liu et al., 2024a). However, within the context of digital gig platforms, decision-making autonomy often manifests as limited autonomy. While platforms ostensibly grant gig workers some discretion in task selection and scheduling, algorithm-driven mechanisms such as dynamic pricing, task matching, and performance evaluation impose implicit constraints on actual decision-making space (Zhu et al., 2024b). Consequently, the moderating effect of decision-making autonomy may vary under different conditions. On the one hand, in high-autonomy scenarios, the opacity of platform algorithms and the complexity of task structures often require workers to engage in frequent task selection and strategic adjustments, imposing additional cognitive demands (Hajiheydari and Delgosha, 2024). These demands may diminish the positive impact of high work engagement on performance. In contrast, moderately structured work arrangements provide clear guidance and allow highly engaged workers to focus on task execution rather than repeated decision-making, thereby enhancing performance outcomes (Wood et al., 2019). On the other hand, under conditions of low autonomy, the platform’s tight control over task flow and scheduling limits gig workers’ ability to self-regulate, increasing the risk of emotional exhaustion for those experiencing burnout (Clinton and Conway, 2024). When autonomy is increased, however, workers can selectively accept tasks based on their personal state, adopt flexible and individualized work strategies, and schedule restorative breaks independently. These strategies can effectively mitigate the negative impact of burnout on work performance (Zhu et al., 2024b). Accordingly, this study proposes the following hypotheses:

H10: Decision-making autonomy attenuates the positive effect of work engagement on work performance.

H11: Decision-making autonomy attenuates the negative effect of burnout on work performance.

Research methodology

This study employed a cross-sectional survey to measure multi-dimensional social support, work engagement, burnout, and work performance among gig workers on digital platforms in Shanghai. The questionnaire was developed based on established scales and refined through a translation–back-translation process, expert review, and a pilot test to ensure reliability and validity. Following data cleaning, confirmatory factor analysis was conducted to assess the structural validity of the scales, and structural equation modeling was used to empirically test the research hypotheses and chain-mediated effects. Although a cross-sectional design cannot capture dynamic changes over time, it is suitable for revealing stable individual differences, particularly in measuring affective responses and psychological states, thereby providing robust empirical evidence for understanding the mechanisms through which multi-dimensional social support influences work engagement, burnout, and work performance (Schaufeli et al., 2002).

Questionnaire design

The questionnaire used in this study consisted of two sections: demographic information and main content. The demographic section gathered background data including gender, age, occupational field, and work experience. The main content comprised four components: social support on digital gig platforms (platform organizational support, customer review support, and family emotional support) as independent variables; gig work states (work engagement and burnout) as mediators; work performance as the dependent variable; and decision-making autonomy as a moderating variable. All variables were measured using a 7-point Likert scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”), allowing quantitative assessment of each construct.

To ensure the cultural relevance of the scales in the Chinese context, all measurement instruments were adapted from validated scales (Biswas and Kapil, 2017; Edwards et al., 2008; King et al., 1995; Kusi et al., 2021; Kuvaas, 2006; Maslach and Jackson, 1981; Schaufeli et al., 2002; Weber et al., 2017), and followed the translation/back-translation protocol recommended by the World Health Organization (2009). Specifically, two bilingual psychology researchers independently translated the English scales into Chinese. Discrepancies were resolved through discussion within the research team to produce an initial Chinese version. Subsequently, two independent translators who were unfamiliar with the original scales conducted back-translations, which were then compared to the original versions. Any inconsistencies were reviewed and revised by the research team in collaboration with subject matter experts to ensure semantic equivalence. Additionally, a pilot test was conducted with 10 randomly selected Chinese gig workers. In-depth interviews were carried out to verify item clarity and comprehension, further refining the wording (Wang et al., 2025).

Measurement scales

The variables were measured using the following instruments: platform organizational support was adapted from Biswas & Kapil (2017) and Kusi et al. (2021), comprising 7 items (α = 0.845), tailored to the characteristics of digital gig platforms, such as “This platform really cares about my well-being.”. Customer review support was based on Weber et al. (2017), consisting of 5 items (α = 0.769), such as “I have received a large number of positive online reviews from customers.”. Family emotional support was measured using the scale by King et al. (1995), with 5 items (α = 0.779), including “Members of my family want to listen to my work-related problems.”. Work engagement was assessed using Schaufeli et al. (2002), divided into three dimensions: vigor, dedication, and absorption, with 13 items (α = 0.918). Example items include “When I get up in the morning, I feel like going to work.” “I find the work that I do full of meaning and purpose.” and “I am immersed in my work.”. Burnout was measured with the Maslach and Jackson (1981) scale, including 10 items across emotional exhaustion, depersonalization, and personal accomplishment (α = 0.894), such as “I feel used up at the end of the workday.” “I’ve become more callous toward people since I took this job.” and “In my opinion, I am good at my job.”. Work performance was assessed using Kuvaas (2006), with 4 items (α = 0.724), including “I almost always perform better than an acceptable level.”. Decision-making autonomy was measured using Edwards et al. (2008), comprising 6 items (α = 0.833), such as “My working time can be flexible.”. See Appendix for details.

Data collection

A cross-sectional survey design was adopted, and data were collected through the professional online platform Sojump, targeting digital gig workers in Shanghai. Two criteria were used for sample screening: First, Job type was identified through the item “What type of gig work do you do? (Online, offline, or both),” selecting only respondents engaged in online or hybrid gig work; and second, IP address verification ensured geographic accuracy. The final sample included only respondents who selected “online” or “both” and whose IP addresses were in Shanghai. Data collection occurred in two phases: a pilot test from March 1–2, 2024, which yielded 100 valid responses meeting quality standards, followed by the formal survey from March 4–15, 2024, yielding 778 responses. After rigorous data cleaning to remove invalid responses, such as those engaged solely in offline gig work, with non-Shanghai IP addresses, incomplete answers, or patterned responses, a total of 500 valid responses were retained, resulting in an effective response rate of 64.27%.

Shanghai was chosen as the research site due to its theoretical significance. As a global pioneer in the digital economy, the city hosts a robust and diversified industrial structure, with approximately 3 million gig jobs (Zhao and Liu, 2024a). The digital gig platform ecosystem in Shanghai offers three key advantages: First, policy support. The Shanghai municipal government has implemented a series of supportive policies to strengthen the recruitment and service systems of digital platforms, thereby facilitating the digital transformation of traditional industries and resulting in higher platform penetration rates compared to regions with less policy support (Han et al., 2024). Second, labor characteristics. The city attracts highly educated and international talent, forming a skill-intensive labor pool. In addition to service-oriented gig roles like food delivery and ride hailing, high end gigs such as legal consulting, IT development, and technical advising are also prevalent, contrasting sharply with talent shortages in many other areas (Jing et al., 2023). Third, the Market Ecosystem. High urbanization and active labor mobility have fostered a multicultural and dynamic market environment, making Shanghai a representative case for exploring the gig economy’s development in urbanized settings (Liu et al., 2024b). This integrated “policy–talent–market” development model provides a representative case for examining digital gig work performance. In a global environment where digital gig platforms frequently encounter employment instability, insufficient social protection, and algorithmic oversight, this study focuses on the role of multidimensional social support, aiming to help gig workers better cope with structural risks, enhance occupational autonomy, and safeguard labor rights (Behl et al., 2022; Gillman et al., 2023). The experience of Shanghai not only provides a practical model for improving the development environment of digital gig work in other regions of China but also offers valuable insights for building a more sustainable global digital gig employment ecosystem.

Data quality assessment

A total of 500 valid questionnaires were collected. To ensure the robustness of the analysis, two methods were employed to determine adequate sample size. First, based on the 10-times rule, the minimum sample size should be ten times the maximum number of paths in the model, which is 11; hence, the required minimum is 110. With 500 valid responses, this requirement was comfortably exceeded (Kock and Hadaya, 2018). Second, a power analysis using G*Power with an effect size of f² = 0.15, α = 0.05, statistical power = 0.8, and seven predictors yielded a minimum required sample size of 103 (Kang, 2021). The actual sample size of 500 also surpassed this threshold, confirming its adequacy for PLS-SEM analysis.

Several tests were conducted to verify data validity. First, Harman’s single-factor test indicated that the first factor accounted for 35.53% of the total variance, below the critical 50% threshold, suggesting no serious common method bias (Aguirre-Urreta and Hu, 2019). To further assess this, “Occupational Fields” was introduced as a marker variable, and the model was re-estimated. The results showed no significant correlations between the marker and core variables, and the structural relationships remained unchanged, confirming the minimal influence of common method bias (Simmering et al., 2015). Additionally, the Kolmogorov–Smirnov test indicated significant deviation from normality for all variables (p < 0.001), justifying the use of PLS-SEM (Massey Jr, 1951). Finally, variance inflation factors ranged from 1.276 to 2.841, well below the critical value of 3, indicating no multicollinearity issues (Thompson et al., 2017). In summary, these checks confirm the validity and reliability of the data, with no notable issues related to common method variance, multicollinearity, or overall data quality.

Demographic characteristics

The demographic data reveals a balanced gender distribution, with males slightly outnumbering females at 51.8%, while females account for 48.2%. In terms of age, the majority of respondents are young to middle-aged workers between 21 and 40 years old, comprising 89.2%, with the largest group being those aged 31 to 40, representing 52.0%. This reflects the central role of this age group in the gig economy. The occupational distribution is predominantly within platform economy-related sectors, including food delivery, courier services, and ride-hailing, which together account for 26.6%, followed by the computer services and software industry at 25.0%, and professional services such as translation and consulting at 19.4%. This highlights the concentration of digital gig platforms within technology-driven and flexible service industries. Additionally, the respondents generally possess substantial work experience, with nearly half (46.6%) having 4 to 10 years of relevant experience, while only 7.2% are newcomers to the workforce. These findings suggest that the study sample embodies the typical characteristics of the digital gig economy, with a predominant focus on young to middle-aged professionals in technology and service sectors, offering a representative snapshot of the digital gig workforce in Shanghai (Zhao and Liu, 2024a). See Table 1 for further details.

Table 1 Respondent’s profile.
Full size table

Results

Measurement model assessment

The measurement model was assessed using SPSS 27 and Smart PLS 4.0. Results indicated that all key constructs demonstrated sound psychometric properties. First, the means and standard deviations of the key variables are presented in Table 2. The social support dimensions, including platform organizational support (M = 5.377, SD = 0.856), customer review support (M = 5.771, SD = 0.724), and family emotional support (M = 5.518, SD = 0.857), all showed relatively high scores, suggesting that gig workers perceived substantial support from platforms, customers, and family members. Similarly, elevated scores for work engagement (M = 5.394, SD = 0.869) and work performance (M = 5.550, SD = 0.816) reflect an overall positive work state and performance level among gig workers. The high mean value for job autonomy (M = 5.243, SD = 0.989) highlights the autonomous nature of gig work. In contrast, the significantly lower and more dispersed score for burnout (M = 2.589, SD = 0.984) reveals notable variation in the extent of burnout perceived across individuals (Zhao et al., 2024).

Table 2 Result of measurement model.
Full size table

Second, the reliability and convergent validity results are shown in Table 2. All constructs demonstrated Cronbach’s α coefficients ranging from 0.724 to 0.918 and composite reliability values (rho_a: 0.727–0.923; rho_c: 0.828–0.930), all exceeding the acceptable threshold of 0.70, indicating strong internal consistency. Convergent validity was also supported, with average variance extracted values ranging from 0.508 to 0.546, surpassing the minimum standard of 0.50. These results suggest that the latent constructs effectively capture the variance of their respective indicators (Zhao and Liu, 2023).

Finally, discriminant validity was assessed using the HTMT ratio and Fornell-Larcker criterion, as shown in Tables 3 and 4. All HTMT values were below the threshold of 0.85, and the square roots of AVE for each construct exceeded their inter-construct correlations. These findings collectively confirm the adequacy of discriminant validity for the measurement model (Zhao and Liu, 2024b). In sum, the measurement model demonstrates satisfactory reliability and validity, providing a solid foundation for subsequent structural model analysis.

Table 3 Discriminate validity (HTMT ratio).
Full size table
Table 4 Descriptive statistics and discriminant validity (Fornell-Larcker criterion).
Full size table

Structural model analysis

This study employed Smart PLS 4.0 to conduct a structural equation modeling analysis to explore the mechanisms influencing work performance on digital gig platforms. The model demonstrated a good overall fit, with a standardized root mean square residual (SRMR) of 0.065, which is below the recommended threshold of 0.08, indicating a satisfactory model-data fit. Regarding explanatory power, the variance explained for work engagement (R² = 0.590), burnout (R² = 0.615), and work performance (R² = 0.542) all reached acceptable levels, suggesting the model possesses strong predictive capability. The results of hypothesis testing are presented in Table 5. Specifically, platform organizational support (0.456***), customer review support (0.207***), and family emotional support (0.227***) all exerted significant positive effects on work engagement, supporting H1–H3. In terms of burnout, customer review support (−0.240***) and family emotional support (−0.200***) showed significant negative effects, supporting H5 and H6, while the effect of platform organizational support was non-significant (0.005, p = 0.918), thus H4 was not supported. Additionally, work engagement had a significant negative effect on burnout (−0.467***), supporting H7. Finally, work engagement positively influenced work performance (0.547***), while burnout negatively impacted work performance (−0.194***), confirming H8 and H9.

Table 5 Structural model and hypotheses results.
Full size table

The findings reveal that platform organizational support, customer review support, and family emotional support effectively enhance gig workers’ work engagement. Among these, customer review support and family emotional support also play a role in mitigating burnout. Within the context of digital gig platforms, platform organizational support primarily provides essential work resources and operational guidance through job opportunities, algorithmic tools, performance feedback, and reward mechanisms, thereby fostering work engagement (Gao, 2023; Lang et al., 2023). However, this support is largely instrumental and informational in nature focusing on task efficiency and output performance, while lacking concern for gig workers’ emotional needs and job security (Zhu et al., 2024b). Thus, while platform support enhances engagement, it does not significantly alleviate burnout. In contrast, customer review support, by offering timely emotional feedback and positive recognition, enhances workers’ self-efficacy and sense of value in their work, leading to greater focus and a positive attitude toward tasks (Xiongtao et al., 2021). Customer reviews, integrated into the platform’s algorithmic mechanisms, influence service ratings, performance assessments, task assignments, and incentive systems. Positive reviews can yield more orders, higher earnings, and reputation growth, thereby substantially boosting motivation (Liang et al., 2025). This form of external social support partly compensates for the limitations of institutional support and contributes to both increased engagement and reduced burnout. Moreover, understanding, encouragement, and care from family members provide emotional attachment and psychological comfort, reinforcing a sense of security and belonging. This enables gig workers to better cope with algorithmic control and customer uncertainty. Such emotional resources help alleviate psychological burdens caused by job insecurity and social isolation, thereby reducing emotional exhaustion and burnout (Bose and Pal, 2020). Furthermore, family emotional support fosters a sense of responsibility and purpose, motivating workers to engage more deeply in their work and strive for better performance to improve family well-being or fulfill familial obligations (De Janasz et al., 2023). This value-driven and emotionally charged support not only strengthens work motivation but also enhances the perceived meaning and purpose of work, promoting sustained work engagement (Bose and Pal, 2020).

Second, the findings indicate that work engagement plays a pivotal role in mitigating burnout. Despite facing multiple challenges, including algorithmic oversight, task pressure, and income volatility, work engagement can be conceptualized, from a dynamic resource’s perspective, as an activating psychological state that facilitates resource gain, whereas burnout reflects a state of sustained resource depletion (Bakker and Demerouti, 2007). On digital gig platforms, the positive outcomes associated with engagement, such as increased earnings and favorable customer evaluations, enhance gig workers’ self-efficacy and buffer the accumulation of negative emotions arising from work demands (Hajiheydari and Delgosha, 2024). Accordingly, the “work engagement → burnout” pathway can be interpreted as a sequential process of resource conversion: highly engaged workers gradually build personal resources through continuous skill development, experience accumulation, and positive feedback, effectively mitigating emotional exhaustion, improving adaptation to platform rules and task requirements, and ultimately enhancing work performance (Liang et al., 2025). This conceptualization not only aligns with the intrinsic logic of resources gain driving psychological improvement as proposed in Conservation of Resources theory but also provides a theoretical foundation for understanding how gig workers sustain motivation and psychological resilience in highly demanding environments (Hobfoll, 1989). Conversely, burnout has a significantly negative impact on performance. Burned-out workers are prone to distraction, emotional detachment, and exhaustion, leading to poor service quality, delays, or negative reviews (Zhao and Liu, 2024a). Given the algorithmic tracking of task completion, response speed, and customer feedback on digital platforms, performance declines are rapidly reflected in platform ratings and job opportunities, potentially affecting income and worker retention (Gillet et al., 2024). Therefore, under highly automated and transparent evaluation systems, the positive effects of engagement and the detrimental impacts of burnout on performance are amplified, making the promotion of engagement and prevention of burnout central to sustaining livelihoods and advancing careers in the gig economy.

Chain mediation effect analysis

The results of the chain mediation effect analysis in this study reveal that the three social support pathways, platform organizational support, customer review support, and family emotional support, each significantly influence work performance through independent and chain mediation effects involving work engagement and burnout, as shown in Fig. 2 and Table 6. First, platform organizational support demonstrates a significant positive effect on work performance through the independent mediation of work engagement (0.250***), but no significant effect through burnout (0.001, p = 0.921). However, the chain mediation effect, where platform organizational support first enhances work engagement and then mitigates burnout, ultimately leading to improved work performance, is significant (0.041***), supporting hypotheses H1a and H1c, but not H1b. Second, customer review support significantly impacts work performance not only through independent mediation of work engagement (0.124***) and burnout (0.039***), but also through a significant chain mediation effect (0.021***), supporting hypotheses H2a, H2b, and H2c. Lastly, family emotional support demonstrates three significant pathways: independent mediation through work engagement (0.113***), independent mediation through burnout (0.047***), and chain mediation (0.019***), supporting hypotheses H3a, H3b, and H3c. These results indicate that social support from different sources influences gig worker work performance through differentiated mediation mechanisms.

Fig. 2
Fig. 2
Full size image

Structural modeling result.

Table 6 Results of mediation analysis.
Full size table

The chain mediation effect analysis further reveals that platform organizational support primarily enhances work performance by boosting work engagement, reflecting the characteristics of algorithmic management in digital gig platforms. Platforms directly stimulate gig workers’ task engagement through real-time data feedback and reward mechanisms (e.g., order rankings, order completion bonuses) (Caza et al., 2022; Wu and Huang, 2024). At the same time, functional support provided by platforms, such as algorithmic tools and navigation optimization, is more focused on improving task efficiency and engagement rather than offering emotional support, which explains why Platform support does not significantly alleviate burnout (Lang et al., 2023). Nonetheless, highly engaged gig workers often accumulate skills and experience resources in repetitive tasks, such as memorizing efficient routes or mastering platform rules, thus enhancing their sense of control over algorithmic management and reducing feelings of helplessness caused by system domination, alleviating anxiety from task uncertainty. This process can partially reduce burnout and ultimately enhance work performance (Hajiheydari and Delgosha, 2024).

Second, the dual effect pathway of customer review support highlights the “double-edged sword” effect of the evaluation system in platform economies. On one hand, visible positive reviews and rewards are directly converted into financial incentives, effectively stimulating gig workers’ work engagement (Xiongtao et al., 2021); on the other hand, negative reviews, leading to financial penalties and ranking declines, create psychological and economic stress, directly resulting in emotional exhaustion (Liang et al., 2025). The platform’s practice of publicly ranking customer reviews creates a “digital panopticon” environment, compelling gig workers to maintain both high engagement and low burnout, which facilitates the formation of the chain mediation effect (Kadolkar et al., 2024; Wood et al., 2019).

Finally, the study reveals the dual effect pathway of family emotional support on gig worker work performance. First, Family emotional support indirectly improves performance through increased work engagement, as emotional encouragement from family members strengthens gig workers’ career identity and sense of achievement, motivating them to engage more deeply in work tasks (De Janasz et al., 2023). Second, Family emotional support enhances performance by mitigating burnout. Emotional comfort and stress relief from family members help reduce gig workers’ negative emotions, enabling them to exhibit stronger psychological resilience in the face of work setbacks. This emotional regulation mechanism plays a crucial role in stabilizing work performance (Bose and Pal, 2020). Furthermore, the psychological energy generated through positive family interactions (e.g., happiness) transfers to the work domain, leading gig workers to actively accept challenging tasks and set higher work goals driven by a sense of responsibility to create a better life for their families, thus ensuring goal persistence (De Janasz et al., 2023).

Moderation effect analysis

To examine the moderating role of decision-making autonomy in the relationship between work engagement and work performance, as well as between burnout and work performance, this study employed SPSS 27 for moderation effect analysis. Additionally, simple slope plots were generated using the ±1 standard deviation method to further explore the direction and intensity of the interaction effects. The results indicate that decision-making autonomy plays a dual role in both relationships. Specifically, in the relationship between work engagement and work performance, when decision-making autonomy is low, the relationship is more pronounced, with a slope of 0.527***. However, in situations with higher autonomy, the positive predictive effect of work engagement on work performance diminishes, with a slope of 0.396***. A similar trend is observed in the relationship between burnout and work performance: under low autonomy conditions, burnout has a stronger negative impact on performance, with a slope of −0.382***, while under high autonomy conditions, this negative effect weakens, with a slope of −0.206***. The interaction terms in the moderation effect regression model were significant, further validating Hypotheses H10 and H11, as detailed in Table 7, Figs. 3 and 4.

Table 7 Moderation effect analysis results.
Full size table
Fig. 3
Fig. 3
Full size image

Moderating effect of decision-making autonomy on the relationship between work engagement and work performance.

Fig. 4
Fig. 4
Full size image

Moderating effect of decision-making autonomy on the relationship between burnout and work performance.

In the relationship between work engagement and work performance, first, gig workers on digital platforms face cognitive overload (Hajiheydari and Delgosha, 2024). A high degree of autonomy creates a complex decision-making environment that requires workers to engage in continuous, multidimensional cognitive processing (Hajiheydari and Delgosha, 2024). Gig workers must balance multiple factors such as task compensation, time constraints, and customer review support; independently manage work pace and prioritization; respond to client needs and complaints; and constantly interpret and adapt to dynamic algorithmic mechanisms (Zhu et al., 2024b). This continuous cognitive processing brings a significant cognitive load, making even highly engaged individuals susceptible to “decision fatigue,” which ultimately undermines the efficiency of core task execution (Kadolkar et al., 2024). Secondly, workers also face a support resource gap. Digital gig platforms commonly exhibit a structural contradiction of “high autonomy, low support” (Behl et al., 2022). On one hand, platforms grant workers high autonomy; on the other, they lack the support resources traditionally offered by organizations, including system training, on-the-job coaching for skill development, social support such as team collaboration and managerial guidance, instrumental support like standardized processes and quality monitoring tools, as well as protective support like minimum income guarantees and grievance mechanisms (Caza et al., 2022; Wu and Huang, 2024). In resource-scarce contexts, highly engaged workers are forced to rely on personal resources to compensate for the lack of organizational support. Over time, this can lead to a “self-exploitation” dilemma, depleting resources and weakening the positive impact of engagement on performance (Lang et al., 2023; Wood et al., 2019).

In the relationship between burnout and work performance, under low autonomy conditions, the platform’s strict control over task processes and time points limits workers’ ability to self-regulate, resulting in a greater risk of emotional exhaustion for those experiencing burnout (Lang et al., 2023). However, a high-autonomy environment first provides workers with crucial recovery opportunities. Workers can adjust their work pace according to their physical and mental state, taking breaks before fatigue accumulates, or choose relatively fewer demanding tasks to regulate stress levels (Wood et al., 2019). Furthermore, workers can adopt work methods that best suit their personal characteristics, reducing unnecessary energy expenditure and effectively interrupting the “stress–exhaustion” vicious cycle (Wan et al., 2024).

Heterogeneity test

To further examine the impact of different occupational types on the study’s results, a heterogeneity test was conducted. Based on 10-times rule and G*Power analysis, considering the minimum sample size requirement of 110 for this study, the sample was divided into two groups to ensure each group met the analysis requirements: high-skilled gig workers (Group 1) and low-skilled gig workers (Group 2) (Van Slageren and Herrmann, 2024). Specifically, Group 1 includes the following occupational types: translation, consultation, professional services, artistic creation activities, information transmission, scientific research, and the computer services and software industry, totaling 321 samples; Group 2 includes takeout, delivery, ridesharing, home care, cleaning, and others, totaling 179 samples. Both groups meet the minimum sample size requirement.

The results for Group 1, high-skilled gig workers, showed that the relationships among the paths remained consistent with the original model, confirming the validity of the original hypotheses. In particular, the negative impact of burnout on work performance was significant, indicating that high-skilled gig workers experience a notable decline in work performance when facing burnout (−0.265**). However, for Group 2, low-skilled gig workers, Hypothesis 9 regarding the effect of burnout on work performance was not supported. Specifically, although low-skilled gig workers experienced burnouts, their work performance did not show a significant decline (−0.059, p = 0.495). Other path relationships remained consistent with the original model. These results suggest that low-skilled gig workers, such as those involved in food delivery, ridesharing, or cleaning, may view platform work as their primary source of income. Despite experiencing burnout, they may maintain relatively high work performance due to income pressures (Jing et al., 2023). This heterogeneity test highlights significant differences in how work performance changes in response to burnout across different occupational types. Specifically, high-skilled gig workers are more sensitive to burnout, while low-skilled gig workers can partially overcome the negative impact of burnout on their performance. This finding emphasizes the importance of considering occupational heterogeneity when studying digital gig platforms.

Robustness test

To further assess the robustness of the model structure, we conducted a reanalysis using the Path-Specific Effects (PSE) method (Liu et al., 2025; Zhou and Yamamoto, 2023). The results indicate a high degree of consistency between the PSE and PLS-SEM findings (see Table 8): the direct effect of social support on work performance (0.413***), the indirect effect (0.292***), and the total effect (0.705***) were all significant. Further analysis showed that social support has a significant indirect effect through work engagement (0.255***), suggesting that work engagement plays a crucial mediating role between social support and work performance. However, the path from social support to burnout to work performance was not significant (0.037, p = 0.227), likely due to the relatively weak effect of platform organizational support on burnout. Overall, the consistency between the PSE and PLS-SEM results in terms of effect direction and significance confirms the robustness of the model. For a detailed summary of all path-specific effects, please refer to Table 8.

Table 8 Total Path-Specific Effects (PSE).
Full size table

Research conclusions and contributions

Research conclusions

This study, integrating Social Support Theory and Affective Events Theory, systematically explores how platform organizational support, customer review support, and family emotional support influence gig workers’ work performance through work engagement and burnout in digital gig platforms. The main findings are as follows:

Firstly, the study reveals that these three types of social support affect work performance through distinct pathways. Platform organizational support primarily enhances work performance by fostering work engagement but has limited effects on alleviating burnout. In contrast, customer review support and family emotional support operate through dual pathways, promoting both work engagement and significantly mitigating burnout, thereby improving work performance. This difference reflects the uniqueness of the platform work environment: organizational support focuses more on productivity enhancement, while external social support addresses the psychological needs of gig workers (Zhu et al., 2024b).

Secondly, the study uncovers a chain mediation pathway through which multidimensional social support alleviates burnout by increasing work engagement, thereby enhancing work performance. This finding highlights the psychological resource transformation process in gig work, wherein the positive psychological capital and experience accumulated through high work engagement effectively counteract the negative effects of burnout, particularly in algorithm-driven work environments (Hajiheydari and Delgosha, 2024).

Finally, the study identifies a double-edged sword effect of work decision-making autonomy. On one hand, high autonomy may exacerbate cognitive burdens and resource scarcity, weakening the positive effect of work engagement on work performance (Hajiheydari and Delgosha, 2024). On the other hand, it provides gig workers with necessary recovery and adjustment space, alleviating the negative impact of burnout on performance (Wood et al., 2019). This finding indicates that decision-making autonomy at work is not inherently a beneficial resource but rather a contextual one, whose effects depend on the interplay of specific task characteristics, available support, and individual psychological states (Mukherjee and Singh, 2024; Mukherjee et al., 2025).

Furthermore, through heterogeneity testing, the study finds key differences in work performance between high-skilled and low-skilled gig workers. High-skilled gig workers are more sensitive to burnout, with psychological stress directly translating into performance decline, while low-skilled gig workers can maintain relatively stable work performance even under burnout, possibly due to stronger economic needs and differences in job nature (Jing et al., 2023). This finding challenges the generalizability of the traditional job demands-resources model in the platform environment.

Theoretical contributions

The theoretical contributions of this study are mainly reflected in three aspects. Firstly, this study is the first to systematically and innovatively examine the mechanisms through which multi-dimensional social support influences work performance on digital gig platforms, providing a critical basis for reconstructing social support networks in “de-organized” work environments. In traditional organizational contexts, social support systems are typically embedded within stable hierarchical structures and sustained interpersonal interactions, with diverse sources, including organizations, supervisors, colleagues, and family, providing emotional, instrumental, informational, and appraisal support, collectively shaping individuals’ psychological states and behavioral responses (Chen et al., 2022). In contrast, the “de-organized” nature of digital gig platforms fundamentally transforms this landscape: algorithmic management replaces most supervisory functions, fragmented work diminishes the basis for colleague support, and gig workers’ support networks become highly dependent on platform algorithms, customer feedback, and family emotional interactions. Reflecting this reality, the study proposes a platform-specific three-dimensional social support framework: First, platform organizational support, essentially algorithm-mediated, delivers instrumental and informational support through algorithmic scheduling, task allocation, and performance feedback, with its effectiveness contingent on algorithmic transparency, fairness, and interpretability (Lang et al., 2023). Second, customer review support serves as a central evaluative mechanism, operating through rating and review systems, with its effectiveness shaped by system design and algorithmic processing, and its influence on gig workers’ behavior and work performance immediately reinforced through the platform’s algorithmic mechanisms (Wu and Huang, 2024). Third, in the context of highly permeable work–family boundaries, family emotional support serves as a critical stabilizing factor, with its value directly influenced by platform-driven work intensity and irregular work schedules (Nahum-Shani et al., 2011). The findings of this study indicate that social support on digital gig platforms has been redefined by platform characteristics such as algorithmic management, evaluation transparency, and work fragmentation, highlighting the streamlined and distinctive nature of support mechanisms in “de-organized” contexts and advancing the adaptive development and theoretical innovation of Social Support Theory in algorithm-driven labor environments.

Secondly, by integrating Social Support Theory with Affective Events Theory, this study innovatively constructs a chained mediation model that systematically explains how affective events, platform organizational support, customer review support, and family emotional support, influence gig workers’ affective and psychological reactions, work engagement and burnout, which in turn affect behavioral outcomes such as work performance (Ibrahim et al., 2024). The results of the chained mediation analysis further confirm the effectiveness of enhancing work engagement in mitigating burnout, deepening our understanding of the relationship between work engagement and burnout (Chen et al., 2023). In contexts of high vulnerability and high autonomy, social support not only directly improves psychological states but also indirectly shapes performance through the “engagement–burnout” pathway, reinforcing the influence of supportive events on behavioral outcomes through both emotional and resource-based channels. Consequently, this study not only confirms the pivotal role of social support in enhancing work performance but also establishes an integrated theoretical framework, revealing the deeper mechanisms through which social support affects behavior via psychological state chains, thereby extending the applicability of Affective Events Theory to non-traditional employment contexts and offering novel theoretical perspectives and analytical foundations for future research on gig workers’ psychological mechanisms and performance formation.

Finally, decision-making autonomy on digital gig platforms exhibits a pronounced “double-edged sword” effect, which also constitutes a central theoretical finding of this study. The mechanism is explicated through dual cognitive and resource pathways: on one hand, grounded in cognitive load theory, excessive autonomy increases continuous decision-making demands and self-regulatory pressure, triggering “decision fatigue,” depleting cognitive resources, diverting attention, and reducing executive functioning, thereby weakening the positive impact of work engagement on work performance (Clinton & Conway, 2024). On the other hand, based on conservation of resources theory, decision-making autonomy functions as a critical contextual resource, enabling gig workers to flexibly adjust task sequences, work pace, and effort allocation, thereby more effectively conserving and optimizing psychological resources in the face of burnout and buffering its detrimental effects on performance (Crayne et al., 2024). Thus, in digital gig platforms, decision-making autonomy can simultaneously diminish engagement benefits while mitigating burnout-related losses, demonstrating a mechanistic duality. This finding overcomes the binary limitations of traditional autonomy research, provides a systematic theoretical explanation for the “autonomy paradox” in digital platforms, and offers a theoretical basis for algorithmically optimized autonomy design.

Practical implications

This study systematically elucidates the mechanisms through which multi-dimensional social support on digital gig platforms influences work performance via work engagement and burnout, as well as the moderating role of decision-making autonomy. Building on these insights, the following practical strategies are proposed across three levels: platform design, policy formulation, and organizational practice.

Firstly, regarding platform design, digital gig platforms should enhance gig workers’ psychological well-being and work performance through algorithmic optimization and functional features. Specific strategies include: improving algorithmic management by demystifying the “black box” of task allocation, performance evaluation, and pricing rules to increase transparency and fairness (Kadolkar et al., 2024); implementing differentiated decision-making autonomy based on task type and skill level, granting high-skilled gig workers greater autonomy in task selection and price negotiation, while ensuring low-skilled workers receive baseline income guarantees, adverse-weather allowances, and clear, transparent rules to balance perceived control and performance outcomes (Han et al., 2024). In addition, platforms should develop digital community features grounded in Social Support Theory, such as official forums, instant messaging groups, and online learning spaces, to encourage informal knowledge sharing, experience exchange, and emotional support, fostering collective strategies for interpreting rules and stress relief, thereby mitigating isolation, enhancing psychological resources, resilience, and performance stability (Hajiheydari and Delgosha, 2024). Platforms can also leverage data-driven monitoring of psychological states, analyzing metrics such as work engagement duration, customer feedback, and communication frequency to provide early warnings and targeted interventions, offering personalized support to gig workers (Lang et al., 2023).

Secondly, at the policy level, gig workers should be provided with fundamental protections and long-term developmental support. Measures include: integrating mental health into social security systems by exploring mental health insurance or assistance funds covering gig workers, offering accessible and affordable counseling, crisis intervention, and employee assistance programs as effective supplements to traditional occupational injury and unemployment insurance (Liang et al., 2025); incorporating well-being protection clauses in platform governance, mandating regulatory or industry standards for algorithm accountability, such as pre-announcement and impact assessment of major rule changes; enforcing rigid safeguards against overwork, including mandatory rest periods and online time limits; and empowering third-party dispute resolution and arbitration mechanisms to prevent algorithmic exploitation or excessive workloads. Simultaneously, policies should encourage platforms to provide vocational training and skill development, offering gig workers sustainable career pathways and promoting employment equity (Shaholli et al., 2024).

Finally, at the organizational practice level, a multi-layered social support network encompassing platform organizational support, customer review support, and family emotional support should be established, with attention to the dual effects of decision-making autonomy. Specifically, organizations should provide skill development, resource allocation, psychological support, and career advancement opportunities to reinforce gig workers’ task execution and professional growth (Zhao & Liu, 2024a). Customer review support should be operationalized through timely and transparent feedback mechanisms, linking positive evaluations to incentives and complemented by periodic recognition, such as “Top Gig Worker” awards, to encourage experience sharing and exemplary practices, thereby enhancing engagement and team morale (Xiongtao et al., 2021). Family support should be facilitated through family-friendly policies, including “interruption-free periods,” reasonable work hours, and family activity rewards, alongside guidance on family relationship counseling and emotional resources to help workers balance work and life (Tóth et al., 2025; Putra et al., 2023). Moreover, organizations should manage the dual impact of decision-making autonomy by providing clear work guidelines, decision-making tools, and dynamic feedback mechanisms to help workers balance autonomy with psychological load (Han et al., 2024). Through coordinated efforts in platform design, policy formulation, and organizational practice, this multi-tiered and finely tuned strategy can systematically enhance work engagement, reduce burnout, improve work performance, safeguard psychological well-being, and support the sustainable development of the digital gig economy.

Research limitations and future directions

This study systematically examined the mechanisms through which multi-dimensional social support affects work performance on digital gig platforms, providing important insights into the reconstruction of social support networks in “de-organized” work environments. Nevertheless, several limitations remain, offering directions for further exploration and refinement in future research.

Firstly, the study focused primarily on Shanghai. Although the gig economy in this region is relatively mature, it may not fully capture variations across China in terms of economic development, platform maturity, and policy support. For instance, second- and third-tier cities may face more pronounced gaps in social protection or skill-matching challenges (He and Leurs, 2024). Moreover, cross-cultural differences in social support patterns may affect the generalizability of the findings (Chen et al., 2023). Future research could expand to multiple regions nationally and even cross-culturally, investigating the heterogeneity of social support mechanisms across different institutional and cultural contexts, particularly in rapidly growing gig economies with underdeveloped social protection systems, such as in South and Southeast Asia.

Secondly, this study employed a cross-sectional survey, which effectively examines relationships among social support, work engagement, burnout, and work performance, providing robust empirical evidence (Schaufeli et al., 2002). However, cross-sectional data cannot capture the dynamic fluctuations of these variables over time, nor can it definitively establish the causal direction between work engagement and burnout (Agrawal et al., 2022). Future research could adopt longitudinal or event-based designs to investigate gig workers’ real-time emotional and psychological responses to specific occurrences, such as platform policy changes, negative customer feedback, or family conflicts, as well as the evolving interplay between engagement and burnout. Additionally, incorporating individual difference factors, such as personality traits, affective tendencies, burnout resilience, and autonomy preferences, could further enhance explanatory power and illuminate complementary mechanisms both across and within individuals.

Thirdly, this study did not systematically compare social support mechanisms across different platform types and industry contexts. Delivery, ride-hailing, and knowledge-service platforms differ significantly in task structure, employment models, and labor relations, which may affect both the demand for and effectiveness of social support (Zvavahera et al., 2024). Future research could conduct multi-platform and cross-industry comparisons, with particular attention to emerging forms of social support within the “creator economy,” thereby developing a more generalizable and forward-looking theoretical framework.

Fourth, while this study examined the moderating role of decision-making autonomy on the relationships between work engagement and work performance, as well as between burnout and work performance, it did not fully capture the potential influence of other contextual factors. Future research could adopt a contingency perspective, investigating how task complexity, platform algorithm transparency, and worker skill levels modulate the impact of decision-making autonomy. This could support the development of a dynamic balance model between “autonomy benefits” and “autonomy costs,” providing a more nuanced theoretical understanding.

Finally, this study has yet to explore how artificial intelligence reshapes social support mechanisms. With the increasing adoption of generative AI, affective computing, and automated management tools on platforms, the form, content, and effectiveness of social support may be undergoing significant changes (Kim and Lee, 2024). For example, can AI-driven real-time feedback partially replace platform organizational support? Does it help mitigate uncertainty and isolation in gig work? These emerging questions represent promising directions for future research.