Date of Award


Degree Type


Degree Name

Doctor of Philosophy


School of Business


Faculty of Business and Law

First Advisor

Dr Peter Standen

Second Advisor

Associate Professor Landis Barratt-Pugh


Research on leadership typically assumes a single leader who has managerial authority over the group, unit or organisation. Shared leadership is an emerging concept of leadership as a group-level phenomenon. It builds on antecedents such as democratic leadership, semi-autonomous and self-managed work groups, participative decision-making and co-leadership that are typically studied as variations of leadership by a single leader. Shared leadership is seen as more distributed, informal and emergent than these. Recent empirical research shows shared leadership can have beneficial effects on a variety of group process and outcome variables. However, so far its effects on creativity have not been empirically examined. This is surprising, since creativity is an important response to increased competition and rapid change in the business environment. Much creativity research identifies important pre-requisites that are more likely to be found in shared than hierarchical leadership. Improved creativity may be one of the most valuable benefits of shared leadership.

This study provides empirical evidence on this relationship from a naturalistic experiment in which student groups were allowed to self-manage over a three-month creative project. In assessing shared leadership, two methodological innovations were introduced. First, previous studies have either used aggregated measures of group performance, or more recently the measures of group ‘degree centrality’ (degree of hierarchy) or ‘density’ (degree of sharing) developed in Social Network Analysis research. However, none of these measures by themselves adequately captures the distinction between hierarchical and shared leadership, although the SNA measures are potentially more precise. Following recommendations of previous authors, this study explored the combined use of centrality and density to better reflect the underlying construct. A second refinement was to use a general construct of leadership based on Bass and Bass’s (2008) extensive literature review, rather than constructs such as transformational leadership that have a narrower theoretical base and tend to assume a hierarchical context. Results from items measuring sharing of Bass and Bass’s five ‘leadership functions’ were compared with a ‘global’ measure of leadership sharing. Creativity was assessed by a panel of judges who rated the groups’ creative outputs (movies), rather than the more common method of rating creativity in the work process.

The results provide evidence for the hypothesised link between shared leadership and creativity that, although qualified by aspects of the study design, suggests further research is worthwhile. Implications for future research on both leadership and creativity are explored, along with consequences for the practice of management. The issues of how to best measure shared vs. hierarchical leadership, and how much a construct can reflect both forms of leadership, are of particular relevance to the future development of this field.

In summary, this study offers the first evidence directly linking shared leadership to work group creativity, and suggests improvements to current methods for measuring the extent of leadership sharing in a group.