Abstract:
The IS environment is riddled with difficulties rendering it complex in nature. It involves
multiple interactions among elements, multiple actors, multiple actor roles, various
degrees of freedom and multiple settings for distribution. Unfortunately, research has
paid limited attention to finding the silver bullet that could manage complexity in systems,
including IS systems. The complexity of IS, as well as the frequent changes to the
agreed-upon requirements common in this environment, causes uncertainty. To improve
our understanding of IT employeesÕ perceived work complexity, this research drew on
the constructs of shared leadership and team performance.
More specifically, the objective of this research study was to develop and test a model
of how work complexity, as perceived by IT employees, influences their team
performance, and the role that shared leadership plays. A new construct, perceived work
complexity, was created. Two existing dimensions of team performance, namely
effectiveness and efficiency, and three types of shared leadership, transformational,
transactional and directive shared leadership, were studied. Shared leadership was
hypothesised to positively predict team performance and negatively predict perceived
work complexity, based on prior research. Perceived work complexity was also
hypothesised to negatively predict team performance. Other factors, such as age and
working at different locations, were also considered.
To test the research model a survey methodology was adopted. Data was collected from
IT professionals in South Africa using an online questionnaire which was developed
using existing literature and multi-item scales to operationalise the studyÕs variables.
Using a non-probability snowball sampling approach, the questionnaire was
administered to IT employees who were invited to participate in the study. Data was
collected over three months and a total of 204 useable responses from IT professionals
in South Africa were collected.
After removing missing data, and checking for outliers, the data was subjected to
confirmatory and exploratory factor analyses to ensure validity and reliably. First, an
exploratory factor analysis to ensure unidimensionality was carried out; this was followed
by a confirmatory factor analysis to prove the convergent and discriminant validity of the
constructs. After proving the validity and reliability of the constructs, composite variables
of the latent factors were created. Finally, hypothesised relationships were tested using
structural equation modelling techniques to test the proposed model.
ii
The final results showed that perceived work complexity negatively predicts team
performance, and that shared leadership negatively predicts perceived work complexity
and positively predicts team performance. Additional analyses suggested that perceived
work complexity can be seen as either a mediator or moderator. Perceived work
complexity partially mediated the effect, reducing it, that shared leadership has on team
performance. The moderating effect of perceived work complexity showed that there is
a significant positive relationship between shared leadership and team performance at
high and low levels of perceived wok complexity. The study also showed that working at
different locations increases perceived work complexity and that age reduces the
perception of shared leadership.
The perceived work complexity phenomenon is important to both academia and IT
management practice. By determining the extent to which perceived work complexity
and shared leadership are important to team performance, this study adds to the growing
body of knowledge on perceived work complexity and ways to resolve key IT
management issues. The results have useful implications for practice.