Previous qualitative research has demonstrated that employees in a wide variety of
service jobs engage in ‘acting’ in order to conform to the service role (Ashforth & Tomiuk, 2000).
Two aspects of job seemed to predict more acting. The direction of relationships with job
satisfaction supported that more acting is performed when employees are displeased with the job
(Adelmann, 1995; Parkinson, 1991), especially surface acting. A mediation test with a nested model showed that job satisfaction’s relationship with affective delivery, demonstrated in the service
literature, was fully mediated by acting. Liking one’s job impacted affective delivery with customers to the extent that it related to surface or deep acting. This gives new meaning to the importance of the ‘happy worker’; that worker may be less likely to put on an act with customers.
Perceiving display rules predicted more deep acting, but not surface acting, supporting the idea that this is a good faith response to work demands (Rafaeli & Sutton, 1987).
Deep acting, or working on inner feelings to appear authentic to customers, had a positive influence on observed interactions with customers, as predicted, when controlling for the level of job satisfaction, surface acting, and emotional exhaustion. This supports that deep acting has the power to convince an audience, as suggested by the Stanislavsky method of acting and Hochschild (1983, see introductory quote), but is this effortful process stressful for the employee? The deep acting and emotional exhaustion relationship was non-significant in the current study beyond surface acting. This is interesting given the inherent effort in deep acting, as seen in the items referring to ‘working hard’ to feel certain emotions. The payoffs of deep

acting, reduced emotional dissonance and positive reactions from customers, may restore
resources of the employee in a way that surface acting cannot. This suggests future implications
for training employees in deep acting, by raising awareness of emotional cues and teaching
techniques of emotion regulation as demonstrated elsewhere (Totterdell & Parkinson, 1999).
Surface acting is faking in bad faith – would an observer recognize the inauthenticity and view the service worker as less capable at affective delivery? With this front-line sample, the more surface acting reported, the lower the perceived affective delivery: the perceived warmth, sincerity and courtesy with customers. Those who report more surface acting were also more likely to be caught breaking character with customers, which seemed to be due to the mediating role of emotional exhaustion. Emotional exhaustion was more likely to be experienced when employees engaged frequently in surface acting, as suggested by previous research on emotional dissonance (e.g., Abraham, 1998, Morris & Feldman, 1997), and this state of depletion leaves one with few resources to keep from breaking character. Another possible explanation is that the employee may try to cope with emotional exhaustion by revealing their negative feelings to customers in order to reduce the emotional dissonance experienced. This provides further support for the damage from burnout in service work (e.g., Wright & Cropanzano, 1998).