Relationship between workplace ostracism and turnover intention among nurses: the sequential mediating effects of emotional labor and nurse-patient relationship

ABSTRACT This study aimed to elaborate on the mechanism by which workplace ostracism influences turnover intention through exploring the sequential mediation effects of emotional labour and nurse-patient relationship. Using a sample of 379 nurses collected from a time-lag survey in a tertiary public hospital in China, we applied structural equation modelling techniques to test our hypothesized model. Our findings revealed that workplace ostracism positively influenced surface acting and deep acting. Workplace ostracism influenced turnover intention through the sequential mediation of surface acting and nurse-patient relationship. The findings of this study imply that nurses should receive education and training in emotional management skills to deal with workplace ostracism. Besides, fostering positive nurse-patient relationships may help reduce nurses’ turnover intention.


Introduction
The turnover rate of Chinese nurses has been high in recent years, further aggravating the shortage of nurses (Yang et al., 2017). It is thus critical for nursing managers to develop an in-depth understanding of the factors affecting nurse turnover. Workplace ostracism increases employees' emotional burden, which leads to turnover intention and turnover (Pierre et al., 2019). However, the inherent mechanism by which workplace ostracism affects nurses' turnover intention is yet to be identified. This study sheds light on the process from workplace ostracism to turnover intention among Chinese nurses by explaining how workplace ostracism influences turnover intention through emotional labour and nursepatient relationship (NPR). According to the theory of emotion regulation, nurses who suffer from workplace ostracism might respond to emotional episodes at work by modifying their expressions (surface acting) and feelings (deep acting). Such responses may consume more of their emotional resources, which undermines NPR. Poor NPRs can lead to disputes and even patient violence, which decrease nurses' affective commitment to the organization. According to the conservation of resources (COR) theory (Hobfoll, 2001), as affective commitment to the organization decreases, individuals tend to develop a tendency to seek new jobs to prevent continued loss of resources. Therefore, a decrease in affective commitment results in turnover intention among nurses (Moreno-Jiménez & Gálvez-Herrer, 2012). In conjunction with the previous arguments, we propose the following hypotheses:

Participants and procedures
The participants were nurses working at a tertiary hospital in Sichuan, China. Two waves of survey were conducted. The first wave was conducted in June 2017. It consisted of a 'demographic profile', a 'workplace ostracism' scale, and an 'emotional labour' scale. Convenience sampling was adopted for this wave, with 420 hospital nurses selected as participants. Specifically, the researchers put each questionnaire copy into an envelope and distributed the envelopes among the participants. 400 valid questionnaires were recovered in Wave 1. Wave 2, which was carried out in November 2017, consisted of a 'nurse-patient relationship' scale and a 'turnover intention' scale. We coded the 400 valid questionnaires recovered in Wave 1, sent them to the participants to fill out, and recovered 379 valid questionnaires. The participants averaged 30.0 years old and had 8.6 years of professional experience on average. The sample consisted of an overwhelming percentage of female nurses (99%), and most of them (70%) were married. The majority of participants had received higher education (67% bachelor's degree holders, 30% college graduates).

Measures
Workplace ostracism was measured by the Workplace Ostracism Scale developed by Ferris et al. (2008). The scale consists of ten items. Emotional labour was measured by the Emotional Labour Scale developed by Brotheridge and Lee (2003) and Grandey (2003) with five items in the dimension of surface acting and three items in deep acting. NPR was measured by the Chinese version of the Nurse-patient Relationship Scale developed by Ma et al. (2020). The scale consists of nine items. Turnover intention was measured by Weng and Xi (2010) Turnover Intention Scale. The scale consists of four items. All the scales were measured on a six-point Likert scale. Table 1 below shows the Cronbach's α values of each scale in this study.

Ethical considerations
This study was approved by the Departmental Ethics Committee in the hospital where our research was conducted.

Results
The correlation analysis is reported in Table 1. A structural equation model was constructed to validate the model. The results of Table 2 and Figure 1 show that workplace ostracism had positive effects on both surface acting and deep acting. Thus, Hypothesis 1a and Hypothesis 1b were validated. The study further showed that surface acting had a significantly positive effect on NPR, whereas deep acting had no significant effect on NPR. NPR had a significantly negative effect on turnover intention.
The Bootstrap method was used in this study to estimate and validate the mediating effect. As shown in Table 3, the indirect effect of workplace ostracism on NPR through surface acting was −0.11 with a bias-corrected confidence interval (CI) of [−0.04, −0.14] at the 95% level, and the CI did not include 0. Therefore, Hypothesis 2a was validated. The indirect effect of workplace ostracism on turnover intention through surface acting and NPRs was 0.06 with a bias-corrected CI of [0.11, 0.02] at the 95% level, and the CI  excluded 0. Therefore, Hypothesis 3a was validated. Since the effect of deep acting on NPR was not significant, Hypothesis 2b and 3b were not validated.

Discussion
First, this study validates the positive influence of workplace ostracism on emotional labour. This is because people tend to use strategies to cope with emotional episodes at work. For example, they may ingratiate themselves with those who ostracize them (Wu et al., 2012). Similarly, when suffering workplace ostracism, nurses tend to adopt ingratiation tactics (such as conformity and flattering) to deal with issues in interpersonal relationships for the sake of avoiding conflicts. In this process, nurses tend to perform more emotional labour, such as surface acting and deep acting, to maximize collective benefits. Besides, our results showed that workplace ostracism influenced NPR through the mediation of surface acting. In comparison, the impact of workplace ostracism on NPR through the mediation of deep acting was not significant. Possible reasons for this are described below. The inconsistency between facial expressions and internal feelings resulting from surface acting leads to a reduction and loss of resources (Grandey, 2000). In the long run, nurses will exhibit stress (Kammeyer-Mueller et al., 2013) and poorer job performance (Goodwin et al., 2011), which are detrimental to good NPRs. However, the alignment of inner and surface emotions in deep acting will not lead to stress and health problems for the individual (Thomas & Abhyankar, 2014). This may lead to less resource consumption and may even facilitate an increase in resources (Goussinsky & Livne, 2016).
Third, the research results validated the sequential mediation of surface acting and NPR on the relationship between workplace ostracism and turnover intention. The results  indicate that good NPRs can serve as a resource to 'compensate' for the loss of resources caused by surface acting, which in turn can help reduce nurses' turnover intention. Finally, the sequential mediating effects of deep acting and NPR on the relationship between workplace ostracism and turnover intention were not validated. This is also related to the insignificant effect of deep acting on NPR.

Implications for practice
Hospital managers should take measures to prevent and intervene in the occurrence of workplace ostracism. Moreover, emotional management courses should be highlighted in nursing curriculum and on-the-job-training programs to develop nurses' emotional management skills to deal with the emotional demands of their work. Lastly, it is important to foster a quality NPR from which all stakeholders may benefit.

Limitations and future research
This study has two main limitations. First, this study surveyed the nurses from only one tertiary hospital in Sichuan Province, China. The size and coverage of the sample are limited. Second, the cross-sectional survey is not the best choice for exploring the causal relationships between variables. Future research may carry out surveys among nurses in a larger scope and conduct a longitudinal study to figure out the relationships between variables.

Conclusion
This study provides new insights into the association of workplace ostracism and turnover intention among nurses. A hypothesized model was established based on the theory of emotion regulation and the COR theory. The time-lag study method was adopted to conduct a questionnaire survey among nurses in a hospital in China. An analysis of 379 questionnaires was performed. The analysis of sequential mediation revealed that the surface acting of nurses caused by workplace ostracism would undermine nurse-patient relationship, leading to increased turnover intention.