The aim of this research was to determine how shared leadership behaviors impact team processes and performance under long-duration and isolated contexts. Investigators were also interested in identifying the antecedents (e.g., personality) of shared leadership emergence, how shared leadership enables team autonomy, and the dynamic nature of shared leadership behaviors over the course of a team’s lifespan. Investigators used NASA's Human Exploration Research Analog (HERA) facility and their laboratory to study these behaviors.
The primary focus of analog data collection was to examine team leadership in contexts whereby the team is operating within isolated, confined, and extreme environments (ICE). Analog data collection focused on the nature of team leadership (and what constitutes effective leadership) over time as most of what is known regarding team leadership has occurred outside the context of space flight and similar ICE environments and has been static in nature. With regards to team leadership not only were the types of leader behavior investigated, but the degree to which these behaviors were distributed throughout the team (i.e., enacted collectively) and the relation to key team processes and outcomes were examined.
The goal of the laboratory investigation was to supplement data collection in the analog environment, where the focus was more on temporal components with respect to team leadership. Efforts with respect to this thrust were primarily geared toward data cleaning and analysis, along with the coding of a subset of the team’s communication for team leadership behaviors.
During this study three thrusts have been worked on that serve to examine the following research questions: 1) what leadership forms and behaviors best enable team performance in isolated, confined environments; 2) what is the role of time and contextual features in determining the type of team leadership needed; and 3) what are enablers to horizontal peer-to-peer leadership in these type of mission critical teams.
1) Longitudinal Data Collection in NASA Analog:
Results from the analog indicated that while leadership was a mixture of hierarchical and shared, it was most often distributed through the crew, as seen in both the 14- and 30-day HERA campaigns. Most often, results indicated no single member engaging in leadership functions of a specific type (i.e., action, transition, interpersonal) to the exclusion of other team members engaging in the specified group of leadership functions. Results also point to the functionality of sharing leadership. The sharing of transition phase leadership behaviors was significantly correlated with teamwork, cohesion, and effectiveness. Similar results were obtained with respect to the sharing of action phase leadership behaviors. Finally, the sharing of interpersonal leadership behaviors was significantly related to teamwork and cohesion. Trends for the sharing of interpersonal leadership being positively related to team effectiveness were evidenced, but did not reach significance most likely due to low sample size.
2) Laboratory Investigation:
Results from the laboratory study tended to provide further support for what was being seen in both the analog environment as well as the archival analysis. First, results suggested that team leadership behaviors were most often distributed throughout the team and not solely enacted by a single individual in the team. Second, incorporating the temporal component investigators studied how often different behaviors were performed across missions. Results indicated that over time, the variety of leadership behaviors increased as well as the density of those behaviors (i.e., the degree to which leadership was distributed through the team or shared), with individuals, on average, having a higher density of leadership behaviors in later missions as compared to earlier missions.
Investigators also studied whether a more objective indicator, participant vocal intensity, might be an appropriate metric for recognizing the degree of shared leadership during particular team performance phases of interaction (i.e., transition, action). Results indicated that the team’s vocal intensity was significantly, positively related to shared leadership. However, in the high autonomy condition, this relationship was only true for action oriented shared leadership density and not for transition leadership density. Essentially, these results appear to indicate that talking more during more stressful time periods was likely to be associated with leadership behaviors. However, when autonomy was low (i.e., more structured), vocal intensity was predictive of both transition and action-oriented shared leadership. This has implications in particular in terms of the role of vocal intensity serving as a proxy for measuring leadership during high stress, action heavy performance episodes, but less so for planning phases of similar high stress teamwork contexts.
Finally, analyses were conducted to examine a set of enablers to shared leadership. Results indicated that at the individual level, motivation to lead was significantly predictive of individuals’ leadership behavior density, as perceived by their team members. This also held at the team level, where the average of team members’ motivation to lead was significantly predictive of shared leadership. However, this was only true for the sub-dimension of affective motivation to lead, implying that shared leadership in this situation may be driven by teams whose members truly enjoy taking on leadership roles. Results also suggested that personality may serve as an enabler to shared leadership.
Results demonstrated that at the team level, average extraversion and openness to experience were significant, positive predictors of shared team leadership, although agreeableness, conscientiousness, and emotional stability were not. This pattern did not significantly change across autonomy or distribution conditions. At the individual level, higher openness to experience was associated with higher individual leadership behavior density, implying that those who were more open to unknown or unfamiliar situations were more likely to step up and take on leadership responsibilities more often. Interestingly, common personality predictors of leadership were not significantly associated with individuals’ leadership behavior density ratings from their teammates, including emotional stability, extraversion, and conscientiousness.
3) Analysis of Archival Data: A final thrust of the project was to use a bottoms-up approach to understanding team leadership in ICE. In essence, the investigators employed a mixture of historiometry and the critical incident technique to analyze archived accounts of teams operating in ICE contexts. In doing so, they focused on teams operating in the following ICE contexts – space flight, polar exploration, and long-duration, round the world, ocean sailboat racing. Using the above methodologies, 152 critical incidents pertaining to team leadership were extracted from 124 data sources consisting of blogs, autobiographies, biographies, and journals. Each critical incident was then independently coded by three subject matter experts with respect to (1) the leadership behavior represented, (2) whether the leadership behavior was enacted by a single individual or multiple individuals (i.e., individual versus shared), (3) formality of leadership (i.e., informal, formal), and (4) locus of leadership (i.e., internal, external). Consensus meetings were held to resolve any discrepancies.