The effect of perceived stress for work engagement in volunteers during the COVID-19 pandemic: the mediating role of psychological resilience and age differences

Perceived stress and work engagement

It is widely acknowledged that shifts throughout a pandemic might function as major aggravation that persuades an elevated amount of stress. Whenever one experiences the constraints of such stress surpassing their coping mechanism or adaptabilities, it might trigger elevated negative emotions, and physical sensations, resulting in a burdensome disruption in their work engagement (Lazarus & Folkman, 1984; Liu et al., 2019). Perceived stress, rather than the objective stressful event, is more relevant to focus on; objectively stressful events may occur, but how they are perceived and affect individuals may vary. Perceived stress is defined as a personal psychological response and subjective feelings when confronted with negative situations (Zheng et al., 2019). Negative emotions are closely correlated with perceived stress (Schiffrin & Nelson, 2010; Spada et al., 2008). Existing studies illustrate that emotion is the key determinant of work engagement, especially negative emotions (Firdaus, 2019; Liu et al., 2019). Yet, research on the correlation between perceived stress and work engagement comparatively has been under examined.

The mediating role of psychological resilience

Existing literature has found that positive psychological resources (e.g., psychological resilience) can help individuals control and adapt to adjoining environments, improving their work engagement (Corso-de Zúniga et al., 2020; Kašparkova et al., 2018; Mache et al., 2014). Researchers have shown a correlation between poor psychological resilience and an exceeding level of work disengagement (Villavicencio-Ayub, Jurado-Cárdenas & Valencia-Cruz, 2014). Psychological resilience reflects people’s adaptation process and positive attitude in the face of negative events (such as stress, trauma, adversity, frustration), which is characterized by vigor (high level of energy and physical activation), dedication (feeling of pride and enthusiasm with one’s work), and absorption (being happily immersed in the work; Schaufeli et al., 2002).

The integrated psychological resilience model proposed by Kumpfer (1999) emphasizes the effect of selective perception, cognitive reorganization and other processes on psychological resilience. Personal cognitive processes of events, such as the cognition and evaluation of stressful events, can affect psychological resilience. In a similar vein, considering both the internal and external contexts, Mandleco & Peery (2007) have suggested the system model theory, which emphasizes cognitive tendency (such as interpretation and perception of stressful events) is an internal influencing factor, which could have a great impact on psychological resilience. Previous studies have also confirmed that perceived stress negatively affects college students’ psychological resilience—where elevated perceived stress denotes a comparatively lower level of psychological resilience (Liu & Wang, 2016; Khosravi & Nikmanesh, 2014). Therefore, perceived stress is likely to reduce the work engagement of volunteers by reducing their psychological resilience.

Characteristics of variables in volunteers at different ages

Previous studies have divided adulthood into three different stages: early (18–35 years old), middle (35-60 years old) and late (post-60 years old; Baikeli et al., 2021). In China, a large number of volunteers are aged between 18 and 60 years. With economic development and social changes, although the three different stages of adulthood are not separated clearly, it should be noted that different developmental tasks are completed at different stages of adulthood. Specifically, with increased age, people’s cognition of difficult situations and pressure as well as their attitude towards work could have some corresponding changes (Almira et al., 2015; Wille et al., 2014). For early adults, they may lack the ability to solve some problems and have low social experience, resulting in a lack of positive psychological qualities when dealing with problems (Hershey & Farrell, 1999). However, with age, problem-solving skills and abilities improve and more complex and difficult situations can be addressed more actively—often showing more resilience and optimism in the face of setbacks (Lipsitt & Demick, 2012). As a result, resilience may also increase with age.

According to the job demands-resources (JD-R) model, the influencing factors of work engagement are divided into demands and resources factors (Bakker & Demerouti, 2017; Schaufeli, 2003). However, this theory only emphasizes the significance of job characteristics and situational considerations of work engagement, ignoring the key role of individual cognitive factors (Li et al., 2014). With age, cognitive factors play a more substantial role than external situational factors for adults (Ramos, Gregor & Georg, 2016). Therefore, older adults’ work engagement may be easily influenced by internal factors. It has been found that there are differences in resilience at different stages of adulthood (Smith & Hayslip, 2012). Generally, the resilience in early adulthood is not mature, and it is difficult to cope well with stressful events. Moreover, due to the lack of social experience, setbacks tend to be treated with subjective experience. Compared with the early adults, those in middle adulthood tend to view difficulties more positively. Even if they often encounter failures in the process, those in middle adulthood are more likely to persevere rather than give up easily. The resilience of the middle adults is relatively mature and more flexible in dealing with complex problem behaviors (Riehm et al., 2020). Consequently, there might be an age-dependent variation in the mediating influence of resilience on perceived stress and work engagement. Compared to the relatively mature mid-adults, the psychological resilience of the early adults is more likely to mediate between perceived stress and work engagement.

The present study

In sum, this study examined whether resilience during the outbreak of COVID-19 pandemic mediated the relationship between perceived stress and work engagement among volunteers. Based on existing literature, we predicted that perceived stress negatively impacts work engagement only through resilience, and the mediating effect of resilience was significant in volunteers in early adulthood.

Participants and procedure

In early November 2021, 981 volunteers (M_age >18) were recruited online from an area in Gansu Province, China, where the first large-scale COVID-19 outbreak occurred in November 2021. Community volunteers included those working on the frontlines of the pandemic (excluding those in healthcare); students with medical and psychological backgrounds, teachers, civil servants, and freelancers formed the volunteer community. Community volunteers were mainly responsible for order maintenance, food distribution, emotional relief, and other work. Because of the severity of the COVID-19 pandemic at the time of the study, the volunteers recruited were taking on a heavy workload, with at least four hours of volunteer work a day until the pandemic was over. In order to reduce direct contact during the COVID-19 pandemic, we adopted the online questionnaire distribution method. We set up an online group for the subjects and distributed the link of the online self-assessment questionnaire to the recruited subjects at the same time. Before initiating the online self-reporting survey, mandatory informed consent was collected from each participant. With a 92.8% effective response rate, 910 valid completed questionnaires were compiled. The average age of the respondents was 37.45 (SD = 3.16) years, with 402 male and 508 female participants. In this study, the age of volunteers was divided into early adulthood and middle adulthood, corresponding to 18–35 years old and 36-60 years old, respectively (Baikeli et al., 2021). There were 503 in early adulthood, 260 male and 243 female corresponding average age of 26.03 (SD = 4.81) years, and 407 in middle adulthood, 195 male and 212 female with an average age of 40.36 (SD = 3.01) years. All the volunteers who participated in this study worked on the front lines of epidemic prevention work for nearly a month, and none reported a positive COVID-19 test result. We have received the permission to use the Chinese version of the Connor-Davidson Resilience Scale (C-CDRISC) and the Chinese version of the Utrecht Work Engagement Scale (C-UWES) from the copyright holders duly for this present research purpose. This present study completely complied with the Code of Ethics of the World Medical Association (Declaration of Helsinki) and its later amendments. Furthermore, the Ethical Board of Northwest Normal University has approved this study with ERB Number 20210256, dated: 01/11/2021. Participants’ demographic characteristics are presented in Table 1.

Measures

Statistical analysis

For this present study, IBM SPSS v26.0 software was used for descriptive statistical analysis, including percentages, means, standard deviations, T-tests, and correlational analyses. For the purpose of hypothesis testing, IBM AMOS v25.0 was used to construct the models and compare structural equation models in multiple groups (Arbuckle, 2009). Power analysis was run using a priori sample size calculator for structural equation models. The required sample size was 323, with 80% power (Soper, 2020). Selected CFI, TLI, RMSEA, SRMR, Chi-square (χ²), and degrees of freedom (df) as the fitting index of the model, which CFI > 0.09, TLI > 0.09, RMSEA<0.08, SRMR<0.05 referred to a reasonable fit (Browne & Cudeck, 1993; Kline, 2005).

Characteristics and relationship of study variables in volunteers at different ages

Mediating effect of resilience

Structural equation model (SEM) was used to investigate the mediating effect of volunteer resilience on perceived stress and work engagement. Perceived stress was an exogenous latent variable, and work engagement and resilience were endogenous latent variables (Fig. 1). As latent variables, perceived stress included crisis perception (a1) and perceived coping (a2). The latent variables of resilience included tenacity (b1), self-reliance (b2) and optimism (b3). The latent variables of work engagement include vitality(c1), dedication (c2), and concentration (c3). The results showed that the fitting indexes of the mediation model were χ² = 57.97, df = 17, CFI = 0.98, TLI = 0.98, RMSEA = 0.05, SRMR = 0.03 and the fitting indexes were all satisfactory.

The path coefficients and relationships among the three variables can be seen from Fig. 1. The perceived stress of volunteers shows a significant direct negative predictive effect on resilience and work engagement, while resilience has a comparatively significant positive predictive effect on work engagement. This suggests that resilience mediates the relationship between perceived stress and work engagement.

Differences in the mediating effect of resilience for volunteers at different ages

Volunteers aged 18 to 60 years were divided into two groups, early and mid-adulthood. According to previous study findings, there are significant distinctions in perceived stress, resilience and work engagement in different age groups. In order to determine whether the mediating effect of resilience is structurally consistent across ages, this study examined the mediating models of early and middle adulthood. The results showed that the indexes of early adult model were χ² = 69.71, df = 17, CFI = 0.98, TLI = 0.96, RMSEA = 0.06, SRMR = 0.04. The indexes of the model were χ² = 60.02, df = 17, CFI = 0.98, TLI = 0.97, RMSEA = 0.04, SRMR = 0.03. In general, all the fitting indexes are in an acceptable range and can be compared across groups (Wen et al., 2004). On this basis, this study adopts the method of multi-group comparison in SEM to set the equivalent model, and the fitting results of each model are demonstrated in Table 4. The result shows that the model between M1 and M2 was Δχ²/ df = 2.74, p = .000; the model between M2 and M3 was Δχ²/ df = 2.50, p = .000; the model between was M3 and M4 Δχ²/ df = 3.68, p = .000; and the model between M4 and M5 was Δχ²/ df = 2.17, p = .000. The difference of fitting index ΔTLI and ΔCFI between the two models is less than 0.01. This indicates that each equivalent model is valid (Cheung & Rensvold, 2002), suggesting the mediating model of resilience has the same significance and underlying structure in early and mid-adult volunteers.

Comparing the results of structural equation models for the different age groups, it was found that there were significant age differences in the impact of resilience on work engagement. The standardized path coefficients for the mediating models for the different age groups are shown in Fig. 2. For middle adult volunteers, the direct path coefficient of resilience to work engagement was not significant (β = 0.06, p = .132), and the direct path coefficients of perceived stress to work engagement were not significant (β = −0.02, p = .095). However, the path coefficient of perceived stress to resilience was significant (β = −0.59, p = .000). This suggests that resilience does not mediate between perceived stress and work engagement in mid-adult volunteers.

Study strengths, limitations, and future research scope

The findings of this present study may have practical as well as theoretical implications. Apparently, it is inescapable for volunteers to perceive elevated levels of stress during the COVID-19 outbreak. During this unprecedented moment, the findings of this present study provide insight into the particular pathways through which perceived stress results in decreased work engagement of volunteers. Stress at work is inevitable, but our research suggests that increased resilience might help keep workers engaged regardless of the stress. Hence, it is necessary to look into a practical ancillary means for volunteers to build resilience and, if at all conceivable, shift focus to volunteers to implement a successful intervention.

Previous studies without COVID-19 have found that perceived stress does affect the work engagement of different workers (Abbas et al., 2021; Adanaqué-Bravo et al., 2023; Cheng & Kao, 2022; Kumar et al., 2021; Romero-Martín et al., 2022; Saleem, Malik & Qureshi, 2021; Zhang et al., 2021). For this study, we selected workers who were most responsible for controlling COVID-19 during the pandemic. When stay at home orders were in place, volunteers made sacrifices to their personal safety to ensure their communities did not spread COVID-19. Although the sample pertained to COVID-19 volunteers, the volunteers included in the study came from a variety of backgrounds; therefore, it is plausible that these findings could be extended beyond a COVID-19 context. The results of this study support those of previous studies in terms of how perceived stress and resilience impacts work engagement. This study adds a new contribution to the literature by identifying a mechanism of perceived stress that affects work engagement of volunteers across different age groups during COVID-19. Future work should continue to explore how volunteers’ work engagement is impacted as well as identifying other mechanisms or moderating factors that could impact this relationship. Overall, the results of this study not only work to fill a gap in the literature but can also be useful for informing interventions to increase work engagement for volunteers so that the prevention and control of COVID-19 can be better maintained and monitored. The results of this study could also be informative for future public health emergencies, where the government may similarly establish a volunteer team.

This present study is attributed with some inevitable limitations. Self-reported survey questionnaires measured the research variables. In contrast to the thorough analysis involved in a qualitative method, certain important information was likely left out of a quantitative assessment. Furthermore, we recruited Chinese samples exclusively, which undoubtedly thwarted generalizability. Hence, it is necessary for future research to investigate and study volunteers in epidemic areas in other countries. Eventually, the substantial direct impact between perceived stress and work engagement recommends that other mediators or moderators might not have comprised through this study. Considering that the COVID-19 epidemic may seem unrestrained for some people, the potential to regulate negative feelings in response to stressful circumstances and adopting emotional regulation tools should be considered as a moderator. Future research might benefit from taking a broader view of the identified components in order to fill in the gaps in our existing findings of the connections between them.