Cohesiveness and performance in an organisational setting : an empirical setting.
Although cohesiveness has been studied for more than half a century, there is no common
definition and associated measure. The current focus of research has moved from general
explanations to context specific approaches. This particular thesis focuses on organisationally
based, limited life, project teams. Despite the increasing use of these types of teams over the
last 15 years, there is no published literature dealing directly with them, and no instrument
designed specifically to measure their cohesiveness. This study has defined the construct of
cohesiveness for these teams, using team members themselves, and a psychometrically sound
measure of cohesiveness has been devised. It is made up of two sub-scales: task and people,
and 20 component items. This tool is designed not only as an indicator of cohesiveness but
also as a diagnostic tool to allow teams to increase their cohesiveness.
Meta-analytical research has found that cohesiveness and performance are related, but that a
number of factors moderate this relationship, e.g. type ofteam (Mullen and Copper, 1994).
Using the task/people cohesiveness measure, three studies relating cohesiveness to
performance and other variables were undertaken. Since project teams are of limited life and
each team's output is unique, a correlational design was utilised. Altogether, data from over
500 teams was collected. Studies were carried out with 'real' teams and also a number of
teams undertaking a business simulation game. Cohesiveness was found to be related to
performance across a basket of performance measures in both circumstances. Task
cohesiveness was more strongly related to performance than was people cohesiveness. By
comparing levels of performance feedback, evidence was found to support Mullen and
Coppers' (1994) suggestion that performance and cohesiveness have a reciprocal effect on
one another and that the performance to cohesiveness effect is the larger of the two. The
relationship of a number of variables to both cohesiveness and also the cohesivenessperformance
relationship were investigated.
This thesis contributes to the literature by using Social Identity Theory as an approach to
defining and producing a measure of cohesiveness for a specific type of group - project
teams. Its application has provided data on the cohesiveness-performance relationship that
supports earlier meta-analytical findings where these are comparable.