Service organizations encourage employees to express positive emotions in service encounters, in the hope that customers “catch” these emotions and react positively. Yet customer and employee emotions could be mutually influential. To understand emotional exchanges in service encounters and their influences on customer outcomes, the current study models the interplay of emotional contagion and emotional labor, as well as their influence on customer satisfaction. Employees might catch customers’ emotions and transmit those emotions back to customers through emotional contagion, and employee emotional labor likely influences this cycle by modifying the extent to which emotional contagion occurs. Data from 268 customer-employee dyads, gathered from a large chain of foot massage parlors, confirm the existence of an emotion cycle. Deep acting, as one type of emotional labor used by employees, hinders the transmission of negative emotions to customers, whereas surface acting facilitates it. Both customer emotions and employee emotional labor thus have critical influences on service encounters. The findings highlight the importance of understanding the potential influence of customer preservice emotions and the presence of an emotion cycle during service delivery.

This study attempts to extend the theoretical and empirical work of previous studies on abusive supervision in light of new insights into supervisor–subordinate guanxi (s-s guanxi) and two distinct dimensions of organizational citizenship behaviours. Data were collected from a survey of 280 employees and their 280 direct supervisors in a Chinese organization. Results indicate that abusive supervision is positively related to subordinates' organizational citizenship behaviours towards individuals (OCBI), whereas there is no relationship between abusive supervision and organizational citizenship behaviours towards the organization (OCBO). Moreover, our analyses establish a mediating role of s-s guanxi in the relationship between abusive supervision and OCBI and OCBO. Our results also demonstrate that s-s guanxi more strongly affects OCBI than OCBO. Implications and specific suggestions for practitioners are discussed.

Despite researchers’ increasing attention on customer sexual harassment, few studies have investigated its effects on the service performance of frontline employees. This study examined the link between customer sexual harassment, as perceived by frontline employees, and their service performance by focusing on the mediating role of difficulty in maintaining display rules and the moderating role of traditionality. The results from a field survey of 359 supervisor–subordinate dyads in a chain of restaurants in China provided evidence that difficulty in maintaining display rules mediates the negative relationship between customer sexual harassment and service performance. In addition, Chinese traditional values attenuate the relationship between customer sexual harassment and difficulty in maintaining display rules and the mediating effect of difficulty in maintaining display rules. Implications for theory, research and management practice are discussed.

This article reports two studies aimed at developing a theory-based multidimensional measure of workgroup emotional climate (WEC) and exploring its relationship with workgroup effectiveness. In Study 1, a four-factor theory of WEC is derived from a review of the climate and emotions literature, followed by the use of both qualitative and quantitative methods to operationalize the model and develop a survey measure. An initial sample of 396 workgroup members provided the data for the exploratory factor analysis and item response theory analysis. Confirmatory factor analysis (CFA) of data from another sample of 334 workgroup members further confirmed the proposed structure. In Study 2, multilevel CFA of data collected from 840 workgroup members from 148 workgroups provided construct, consensual, and discriminant validity. We also examined the relationship between WEC and workgroup effectiveness. The resulting four-factor, 16-item measure demonstrated robust psychometric properties, with acceptable levels of reliability and validity.