processes (Keltner and Gross 1999, Morris and Keltner 2000). The strength of a unit’s anger culture also appears to be reinforced over time through the process of attraction-selection-attrition (Schneider, 1987, Schneider et al., 1995), in which people gravitate to the work because they are inspired by its purposes and comfortable with the role of anger expression or suppression in the context.
Finally, our findings also suggest that cultural differences will influence how an employee responds to display rules and what work units’ display rules will be. Other studies have already linked culture and anger expression (Kitayama Mesquita and Karasawa 2006, Kumar 2004). Cultural differences emerged in our study when we examined social work departments located within large urban hospitals in two different countries that had different anger norms. The unit in Singapore was identified as suppressed, while the unit in the US was contingent. Both groups found it unacceptable to express anger towards clients, but the US group found it acceptable to express anger when they were working on behalf of their clients. In Singapore, we did not have reports of social workers expressing anger to others on behalf of their clients the way the US social workers did. We believe the Singapore social workers may have reported more suppressed anger norms in their work unit than the social work department in the U.S. because of the cultural differences. That is, anger expression is typically suppressed in Asian cultures to save the face of both the expresser and the target (Ting-Toomey and Oetzel 2002).
Our findings also extend the emotion labor literature. Rather than focusing primarily on emotion regulation processes and their consequences for organizations (see Grandey 2000), we demonstrate the importance of expanding research to consider a broader spectrum of emotional expressions. While the emotion labor literature has revealed the utility of regulating anger expression for certain classifications of employees (Rafaeli and Sutton 1989, Sutton 1991) and has considered the impact of such regulation on those engaged in it at the individual level (Beal et al. 2006, Cote 2005, Hochschild 1983, Morris and Feldman 1996), until now there has been little attention to or differentiation between the individual- and meso-level outcomes of anger expressions. Our research develops these distinctions by showing that different forms of anger expression can yield both positive and negative outcomes at the individual and