Why do workaholics experience depression? A study with Chinese University teachers

Introduction

The impact of job stress on the physical and psychological health of worker has been extensively studied with regard to job characteristics. However, workers’ personal characteristics may be other important factors. One personal characteristic associated with workers’ health is workaholism. Campbell and colleagues suggested that workaholism had been an emerging topic in academic research, and several researchers have conducted studies on workaholism (Campbell et al., 1976). Oates compared workaholism with alcohol addiction and created the concept of workaholism (Oates, 1968; Van Wijhe et al., 2013) which was defined as individuals whose health, happiness, relationship, and social functions are affected by too much work (Campbell et al., 1976; Porter, 2001). Robinson considered workaholism as an obsessive–impulsive disorder of individuals, which is reflected by the demand of self-obsession. Workaholics could not control their working habits and are excessively addicted to work; thus, they ignore other aspects of life (Robinson, 1999). The models of workaholism could be grouped into medical addiction and psychological addiction (Malinowska and Tokarz, 2014). The hypothesis of medical addiction model revealed that long-time work could lead to excessive adrenaline of an individual, which could make him or her more dependent on his or her positive emotion (such as happiness). Such dependence inspires him or her to work for a longer time, to produce more adrenaline, and to have more positive feelings. Thus, a persistent addiction circulation was formed (Finkle and Shin, 2014; Peng et al., 2014b). According to the psychological addiction model, the behavior of workaholism was not driven by exciting factors from a long-time work, but by specific interests (including respect and money) (Shimazu et al., 2010). Combining the behavior and the interests of an individual, his or her work would be his or her resources and functions. If an individual ceases to work, his or her functions would not be fully developed and even cause “withdrawal” reaction including panic. Thus, workaholism could be referred to as the psychological dependence of an individual or addiction to his own work behavior (Zhao et al., 2014). Several researchers believe that workaholism is an obsessive and poor work state with the cost of self-sacrifice (Peng et al., 2013b; Zhang et al., 2014b). However, excessive work could not only harm families and social relations, but could also lead to other problems including anxiety, depression, ulcer, chronic headache, high blood pressure, apoplexy, and even suicide (Agostini et al., 2014; Claude et al., 2014; He et al., 2014), which could seriously affect physical and mental health.

Job burnout is a major ill-health outcome of workaholism. Job burnout refers to physical, emotional, and mental exhaustion state caused by heavy work, which is indicated by the feeling of hating work (Bianchi et al., 2013; Peng et al., 2014a). Job burnout is characterized by body tiredness, depressive mood, exhausted creativity, and low sense of value (Peng et al., 2013a). In general, a workaholism may keep working hard to meet work requirements and maintain performance, yet sacrificing physical, emotional, and mental health. According to research, long-term workaholism is not related to work performance, and continuous investment of time and energy may not lead to an expected achievement (Ouyang et al., 2015; Porter, 2004). The above researches show that job burnout as a mediator can affect workaholism and depression. Thus, once the balance of payment and achievement has been broken, workaholism can further harm the physical and psychological health of an individual. For example, an employee can experience job burnout including emotional exhaustion, depersonalization, which reduces personal accomplishment.

Previous studies also revealed that workaholism is associated with depression. Depression (also called depressive disorder) is the major type of mood disorder that displays clinical features of significant and lasting depressive mood (Yu and Fan, 2014). According to clinical experience, depressive mood is inconsistent with its environment (Tully and Cosh, 2013) and is characterized by low spirit, desperation, inferiority, and even pessimism (Diener et al., 2013). Researchers have discussed the relationship of workaholism and depression, that is, excessive work can lead to depression (Abdolmanafi et al., 2011; Robinson and Kelley, 1998). For instance, Morris and Charney (1983) suggest that workaholics experience anxiety and even depression when they are deprived of work. Furthermore, studies have also revealed the positive relationships between job burnout and various mental health problems; this is particularly true for distress and depression. It seems that those three variables are highly related with each other.

Although previous studies provided adequate evidence which demonstrated the relationships between workaholism and job burnout, and job burnout and depression, further studies were still needed to elucidate the trilateral relations. First, it was necessary to test the concurrent effect of workaholism and job burnout on depression so that we could have an overall prospect about the relationships among them. To our knowledge, no study has been conducted to explore the mediator effect of job burnout on the relationship between workaholism and depression. This study was thus aimed to fill this gap. We hypothesized that job burnout mediated the impact of workaholism on depression. Second, most related researches were conducted in Western culture of individualism. It is also important to confirm these findings in culture of collectivism, take Chinese culture as an example. In summary, this study selected Chinese college teachers as the sample and discussed the mediating effects of job burnout on workaholism and depression.

Method

Results

Descriptive statistics and correlation analysis

First, the inter-correlations of all the latent variables were calculated to test the significance of correlation coefficient. Means, SDs, and inter-correlations for all the variables were presented in Table 1. The results showed that workaholism was positively correlated with depression and job burnout. In addition, job burnout and depression were also positively correlated.

OPEN IN VIEWER

Table 1.

Means, standard deviations (SDs), and correlations of workaholism, job burnout, and depression.

	Mean	SD	1	2	3
1. Workaholism	21.73	4.48	1
2. Job burnout	33.08	13.16	0.431**	1
3. Depression	66.67	7.56	0.439**	0.417**	1

**

p < 0.01.

Measurement model

Then, a two-step procedure was adopted to analyze the mediation effect in order to confirm the structural relations of the latent structured model (Anderson and Gerbing, 1988). First, the measurement model of the four latent variables was tested to assess the extent of goodness of fit represented by its indicators, respectively. If index of confirmatory measurement model meet the requirements, then the maximum likelihood estimation would be used to test the structural equation modeling (SEM) (Peng et al., 2013a). The following four indices were utilized to evaluate the goodness of fit of the model: (a) χ²/df, (b) the standardized root mean square residual (SRMR), (c) the root mean square error of approximation (RMSEA), and (d) the comparative fit index (CFI) (Hu and Bentler, 1999). In this study, a model was considered to have a good fit if all the path coefficients were significant at the level of 0.05, χ²/df was below 3, SRMR was below 0.08, RMSEA was below 0.08, and CFI was 0.95 or more (Lian et al., 2014; Yu et al., 2014).

To control the magnification of errors and improve the reliability and normality of the resulting measurements (Nasser-Abu Alhija and Wisenbaker, 2006; Xiao et al., 2014), three item parcels were created for depression with the factorial algorithm proposed by Rogers and Schmitt (Zhang et al., 2014a). Therefore, fewer indicators were used and data had better fit values. The mean of items in one parcel was used as an observed variable (Rogers and Schmitt, 2004).

Evaluating the adequate suitability of the measurement model to the sample data required the confirmatory factor analysis to be conducted. As shown in Figure 1, three latent constructs (workaholism, job burnout, and depression) and eight observed variables were included in the measurement model. All the indices of the measurement model showed to be suitable to the data: χ²/df = 1.08, RMSEA = 0.014, SRMR = 0.028, and CFI = 0.99. All the factor loadings for the indicators on the latent variables were significant (p < 0.001), which indicates a high representation of the latent constructs by their indicators.

OPEN IN VIEWER

Figure 1.

The measurement model.

Structural model

For the first step, the direct effect of the predictor variable (workaholism) on the dependent variable (depression) without mediators was tested. The directly standardized path coefficient was significant, β = 0.50, p < 0.01. Then, the partially mediated model (see Figure 2) which contained mediator (job burnout) and a direct path from workaholism to depression was tested. The mediational model showed a very good fit to the data, χ²/df = 1.21, RMSEA = 0.025, SRMR = 0.031, and CFI = 0.98. The effect of workaholism on depression through job burnout was 36.09 percent.

OPEN IN VIEWER

Figure 2.

The final structural model.

The confidence interval of direct and indirect effects

The mediating effect of job burnout between workaholism and depression was tested for a significance by adopted the bootstrap estimation procedure (a bootstrap sample of 1000 was specified). Table 2 shows the indirect effects and their associated 95 percent confidence intervals (CIs). As shown in Table 2, workaholism had significant direct effect on depression, β = 0.37, p < 0.01. In addition, the indirect effect of workaholism on depression through job burnout was also significant, β = 0.13, p < 0.01.

OPEN IN VIEWER

Table 2.

Direct and indirect effects and 95 percent confidence intervals (CIs) for the final model.

Model pathways	Estimated effect	95 percent CI Lower bonds	95 percent CI Upper bonds
Direct effect
Workaholism → depression	0.372	0.205	0.533
Workaholism → job burnout	0.509	0.376	0.639
Job burnout → depression	0.251	0.106	0.379
Indirect effect
Workaholism → job burnout → depression	0.128	0.054	0.198

Discussion

This study investigated the concurrent effect of workaholism and job burnout on depression and further examined the mediator effect of job burnout for the relationship between workaholism and depression in Chinese college teachers. SEM was used to determine the direct role of workaholism on depression, as well as the mediating roles of job burnout. Results show good level fit indices. In the partially mediated model, a positive relationship between workaholism and depression was found in this study, which suggests that teachers with higher scores on workaholism were more likely to have higher levels of depression. The intermediary role of job burnout has been analyzed by discussing the relationship of workaholism and psychological health among college teachers and based on the work requirement–supply theory (Kristof, 1996).

First, this research proves that workaholism is harmful to the psychological health of college teachers. Self-identified workaholism may be found in a third of working population and widely exists in senior management, technical personnel, and other groups with higher education, especially among college teachers (Andreassen and Griffiths, 2012; Villella et al., 2011). Workaholism indicates that an individual cannot control his or her work, which compels him or her to put all energy to work (Robinson, 1999). Specifically, a workaholic is extremely addicted to work that he or she put all his time and energy into work and neglects the other aspects of life. A workaholic might deviate from his or her emotion and cognition when working for a long time and under work pressure. When work becomes the priority in life, psychological health can be affected. Long-time work can also be exhausting and may lead to depression. Such results are consistent with the research conclusions of predecessors (Shimazu et al., 2010).

Another finding of the study was that the structural model supported the mediation effect of job burnout between workaholism and depression in Chinese college teachers, that is to say, the path from workaholism to depression through job burnout was significant. The requirement–supply theory (Kristof, 1996) discusses the mechanism of workaholism. According to this theory, job burnout plays a significant mediating role in the relationship of workaholism and depression, which coincided with this study. Work gradually consumes the physical and psychological resources of employees, which leads to job burnout and harm on physical and psychological health (Peng et al., 2013a). Job burnout may worsen the bad reaction caused by work requirements and may also harm their mental health. College teachers are faced with the pressure of teaching and scientific research, but they have relatively little income and welfare. The more they work hard, the more they feel depressed. Thus, college teachers with workaholism gradually lose their enthusiasm to work, passively treat their work and colleagues, and lose their sense of achievement, which can further harm their psychological health (Isenbarger and Zembylas, 2006). Moreover, the result successfully proves the availability of requirement–supply model that has been applied in workaholism research, which can inspire future researchers to explore further the differences of workaholism and work involvement.

The findings of this study have applications in strengthening the importance of counseling and interventions in college teachers with workaholism as there is a high prevalence of workaholism in them. Workaholism among college teachers has a negative effect on their psychological health because workaholism broke the balance of work requirements and work resources. High work requirements and low work resources lead to job burnout, which further affects their mental health. Investing more on educational resources could reduce job burnout among college teachers. Therefore, a proper social support should be provided to necessarily control the catalytic effect of depression among college teachers brought by workaholism, which would ultimately be transformed to job burnout. Future research should examine more examples to produce more extensive results. In addition, local research with multi-peculiarity, multi-methods, and long-time following type should be considered.