The respondents in this research were modern service industry workers from two units in Shanghai. The average score for the six items on the work satisfaction scale was 3.25, indicating significant room for improvement.
The single-factor analysis of work satisfaction indicated that it may be associated with age, education level, work schedule, length of employment, and gender [21–25]. The relationships between age and work satisfaction showed a U-shaped trend. This is perhaps because younger employees may receive more care and face fewer difficulties associated with living independently, while older employees may have rich life experiences and a stable network of relationships. By contrast, middle-aged workers may face more challenges and setbacks in life and thus have lower work satisfaction. The same seemed to hold true for the length of employment. Those hired within the last 5 years showed higher life satisfaction and social support. Regarding education, those with a junior high school education showed the highest work satisfaction. This could be partly related to lower psychological expectations as well as the distribution of the survey population: most with a junior high education had a shorter length of employment. Previous studies have also noted differences stemming from work schedules. This could be because shift work can require long hours and is characterized by less peer support since work hours are usually staggered with colleagues.
However, according to the multiple linear regression analysis, general demographic characteristics explained little in terms of work satisfaction. The adjusted R2 was only 0.082, suggesting that focusing on work schedule, marital status, and other demographic variables might not be effective for improving work satisfaction. Indeed, there is a need for more focus on factors such as social capital and perceived stress.
Many studies have investigated the relationship between social capital and work satisfaction. Imam et al. [26] suggested that social capital can enhance work satisfaction. Huang [27], moreover, reported that social capital could improve job satisfaction in hospitals in Taiwan. In the present study, Pearson’s correlation for social capital and work satisfaction was 0.711 and was significant at the 0.05 level. Further linear regression analysis showed a strong protective effect of social capital on work satisfaction. Forms of social capital such as corporate identity, mutual assistance, and trust may help employees enhance their sense of belonging, cope with challenges at work, and experience improved work satisfaction. Strömgren et al. [28], Shin et al. [29], and others have verified as much, especially in terms of the significant effect of trust (including interpersonal and organizational) on work satisfaction.
Regarding perceived stress, the simple correlation analysis showed that it was negatively associated with work satisfaction. Stratified regression analysis confirmed that perceived stress had a negative predictive effect on work satisfaction. The current fast-paced, competitive work environment has placed higher requirements on workers’ self-regulation ability and learning ability, among others. When stress levels become high and are perceived by staff, they may adversely affect mental health, work positivity [30], and job satisfaction to the point of resigning from their jobs. Enterprises could counter such outcomes by providing services such as psychological counseling, mutual aid platforms, vocational training, and other means of social support [31]. In this way, social capital may affect work satisfaction by intervening in stress perception.
Since previous research has mainly focused on the relationship between the two factors of social capital and work satisfaction, we explored three factors—social capital, perceived stress, and work satisfaction—using Amos 24.0 to construct the structural equation model. The test for mediating effect was based on Baron and Kenny’s [32] in-turn-test theory: given the existence of significant relationships among the independent variables, dependent variable, and intermediary variable, the correlation coefficient between the independent and dependent variables should be tested to determine whether there is a complete or partial mediation effect.
Since the results of the previous correlation tests and multiple linear regression satisfied the prerequisites for mediating effect, further verification of direct and indirect effects showed that perceived stress played a partial mediating role between social capital and work satisfaction. In other words, social capital may directly affect work satisfaction, or it might also indirectly influence work satisfaction by improving stress perception ability. Therefore, work satisfaction may be improved by strengthening employees’ social capital, thus aiding both enterprises and society.
This study has some limitations. First, it was designed as a cross-sectional study, and the respondents were employees at just two modern service institutions in Shanghai. As such, the findings have poor generalizability. Second, the mediating effect of social capital and work satisfaction only incorporated perceived stress, and the model contained limited variables. Future research can further expand the relevant factors and other outcome variables. Despite these limitations, this study can serve as a supplement for work on the relationship between social capital and work satisfaction and provide a reference for dealing with the problem of low work satisfaction among modern service industry workers.