Emotion work in the Hellenic frontline services environment: how it relates to emotional exhaustion and work attitudes

Abstract

The relationship of emotional work with emotional exhaustion and a variety of attitudes towards the job and the organization was investigated in a sample of Hellenes (Greek) employees in the banking industry who were performing frontline service jobs. Emotion work was conceptualized and operationalized in terms of five dimensions: surface acting, deep acting, frequency of emotional display, intensity of displayed emotion, and variety of displayed emotions. The results suggested a weak mediating role of emotional exhaustion in the relationship between emotion work and work attitudes. The results also revealed the presence of a substantial number of interaction effects in the relationship of emotion work with emotional exhaustion and work attitudes. In particular, emotion work was related to emotional exhaustion solely by means of interactions, and the largest part of the relationship of emotion work with work attitudes was also due to interaction effects. In the cases that main effects were present these were mainly caused by an allegedly secondary dimension of emotion work, the variety of displayed emotions, and not by any of the primary dimensions. The national cultural characteristics provided an account for the identified relationship pattern between emotion work, emotional exhaustion and work attitudes. In general, the study revealed the importance of the interactions between the dimensions of emotion work, and was also suggestive of the role of the national cultural context in the way emotion work is associated with outcomes that are of interest to employees and organizations.

Keywords:

• emotional exhaustion
• emotional labour
• Greece
• interaction
• national culture
• work attitudes

Notes

1. The terms ‘emotion work’ and ‘emotional labour’ signify exactly the same construct and should be seen as synonymous (see Zapf 2002, pp. 238–239).

2. It has been legitimately suggested that the construct of emotion work should be expanded to include interactions with coworkers as well (Ashforth and Humphrey 1993; Tschan et al. 2005; and see Totterdell and Holman 2003). However, the present work adhered to the original or traditional meaning of the construct as pertinent exclusively to interactions with customers. Empirical findings indicate that the requirement for displaying prescribed emotions is substantially greater (by a factor of three) in interactions with customers than in interactions with coworkers (Tschan et al. 2005).

3. These two studies, by Brotheridge and Grandey (2002) and by Brotheridge and Lee (2003), utilized the same sample and measured four of the dimensions of emotion work with identical scales (only the assessment of frequency of emotional display was accomplished with different items, see Brotheridge and Grandey 2002, pp. 23–24 and pp. 34–35, and Brotheridge and Lee 2003, pp. 369–370). To avoid misunderstandings, the studies had different aims. However, the parts of these studies that are pertinent to the present work are identical; hence their respective findings are treated as a single report.

4. The Hellenic educational system is structured as following: elementary school (six years duration); gymnasium (three years duration; equivalent to junior high-school); lyceum (three years duration; equivalent to senior high-school); technical/vocational education (variable duration; leads to a technical/vocational qualification); higher education (three to four years duration; equivalent to a polytechnic or two/three-year college degree); highest education (four to five years duration; equivalent to a university or college degree); and postgraduate education. Attendance to gymnasium level is compulsory, hence, this was the minimal educational qualification included in the questionnaire.

5. This increase in the amount of variance accounted for by a set of predictors after the addition of what is in essence a control variable (emotional exhaustion in the present case) must not be seen with surprise or suspicion. It can be explained by means of the relationship between the new variable in the equation and the other predictor variables, which can result in changes in the relationship of the latter with the criterion (e.g. see Howell 1987).