Limitations and further research
Our study is subject to a number of limitations, which provide the basis for further research in this area. First, we purposefully chose to keep industry and home country constant, thus being able to focus our investigation on language-induced effects. Future re- search could investigate language-induced emotions and related leadership strategies in other industries and home countries to dis- cover similarities and differences across contexts. In addition, given the focus on one home country, we had a disproportionately large number of interviewees from this country. However, given that our sample still includes MNT leaders from five different nationalities and subordinates from 19 different countries, we are confident that our data provide sufficient spread to arrive at differentiated argu- ments about effective strategies of managing language-induced emotions. In addition, our approach also entailed advantages related to our background as researchers. Speaking the home country language German as our mother tongue and being very familiar with German business culture helped us to make sense of informants' accounts in a way that preserved the authenticity of their perspec- tives.
Second, space restrictions did not allow us to discuss in detail possible cultural preferences for particular emotion management strategies. However, a large majority of our interviewees across different cultural backgrounds embraced the above outlined emotion management strategies. Consequently, we found national culture to be of little relevance for the specific language-based challenges of MNT leaders. We nevertheless encourage future studies to explore in greater depth how the solutions to language-induced emotions play out in different cultural environments.
Third, we focused in particular on emotion management as performed by MNT leaders, since this emerged to be of particular rel- evance to our respondents. This focus implied, however, that we did not account for team members' self-management and, hence, their contribution to emotions at work. We acknowledge the validity of Stewart idea that entire teams actively engage in the regulation of team emotions and refer to Hinds recent study on global virtual team members coping with language-induced emotions.
Fourth, while our interviews did not allow us to capture intra-individual variation in emotions in real time, we argue that our critical incident-based interviewing technique is well suited to capture the dynamic nature of individual emotions. This technique focuses informants' attention on specific events, which triggered particularly intense emotional experiences. It is therefore in line with affective events theory, which posits that particular events are a chief source of emotions. Nevertheless, we encourage future research studying the intra-individual variation in language-induced emotions specifically with event-based design.
Fifth, we did not explicitly examine how language-induced negative emotions and the positive leadership outcomes of language- related emotion management influence team performance, as the relationships between negative emotions and performance and between leadership outcomes and performance have already been investigated to large extents. Moreover, since team performance is a multifaceted construct with context-specific indicators, we argue that general performance measures would not adequately capture the dynamics of MNT leadership. However, our study indicates the beneficial effect of language-related emotion management strategies on MNT performance by providing clear evidence for its positive effects both in terms of mitigating negative emotions and creating directly positive leadership outcomes. We encourage qualitative follow-up inves- tigations to examine these relationships in more depth and quantitative studies to test them statistically.