Paradoxes of leadership: examining the effects of ambidextrous leader behaviours on follower innovation and work attitudes

This thesis tests the Ambidexterity Theory of Leadership for Innovation (Rosing, Frese, & Bausch, 2011), which suggests that leaders who use two sets of contradictory behaviours at appropriate times (temporal flexibility), can facilitate their followers’ innovation. Although it is argued to be the most effective leadership style for innovation, studies so far have shown mixed results. This thesis tries to answer important questions about its theoretical assumptions and components, some of which have been neglected so far. By following a positivist philosophical approach, and a quantitative methodology, I conducted two studies to examine a theoretical model that I developed, which is based on the ambidextrous leadership model, but takes into consideration further intrapersonal and situational factors. The first study (experiment) examined whether the interplay between leaders’ opening and closing behaviours at the right time, predicts follower innovative behaviours, and whether follower ambidexterity (exploration and exploitation) and motivation mediate the effect of ambidextrous leadership on innovation. This study followed an experimental design, which focused on the temporal flexibility aspect of the theory. A sample of 122 individuals took part in the experiment. Data were analysed on SPSS, using regressions, ANOVA, and mediation analysis, as well as through the Consensual Assessment Technique (CAT). Results were mixed and indicated that temporal flexibility, as well as the main assumptions of the theory were not significant. The second study (daily diary) aimed to replicate the findings from the first study in a natural setting and extend them by investigating the moderating effect of leader-member exchange and trust. This study followed a longitudinal design and focused on the portrayal of leaders’ opening and closing behaviours in a natural context. It employed an experience sampling method approach (daily diary study) to examine daily fluctuations of leaders’ behaviours over a week. A sample of 124 individuals took part in this study. Using linear mixed modelling and SPSS macros for longitudinal models to analyse the data, results were also mixed, with some supporting parts of the theory, while others did not. This study also found no support of the main theoretical assumption that leaders who engaged in contradictory behaviours would facilitate follower innovation. Implications, limitations, and future research suggestions are discussed for both studies.