This research investigated job satisfaction as a group-level construct. The group-level job satisfaction construct was labelled "group task satisfaction", and defined as the group's shared attitude towards its task and the associated work environment. It was proposed that group task satisfaction develops from homogeneity in group members' job attitudes. However, once group members become aware that these attitudes are characteristic of the group, group task satisfaction comes to be perceived independently of individual job satisfaction, as a group characteristic. Consequently, group task satisfaction was conceptualised as being independent of, although closely related to, the level of individual job satisfaction within the group.
The validity and utility of the construct of group task satisfaction was tested in three empirical studies. The participants in the first study were 436 undergraduate students (representing 175 groups) who were completing group projects as part of their assessment. Group members' ratings of their group's task satisfaction demonstrated good within-group agreement and significant between-group variability. Aggregated group task satisfaction ratings were correlated with the mean level of individual job satisfaction within the group, the difficulty of the performance goal set by the group and the quality of the group's work. Li addition, the pattern of relationships exhibited by group task satisfaction was different from the pattern of relationships exhibited by other group constructs and aggregated individual job satisfaction.
The second study explored the construct of group task satisfaction in an organisational setting. Data were obtained from 71 groups working in nine different organisations. Group task satisfaction ratings obtained from group members correlated significantly with supervisor ratings of organisational citizenship behaviour, and had a marginally significant correlation with supervisor ratings of performance. They were also strongly correlated with group members' ratings of their own job satisfaction, the quality of the group's processes, the perceived importance of the group's work and the level of autonomy experienced by the team. The strength of the relationship between ratings of group task satisfaction and ratings of individual job satisfaction was found to be moderated by inclusiveness, such that group members who spent more time working on the group's task rated their own job satisfaction and the group's task satisfaction more similarly than group members who spent less time working on the group's task. In Study 2 the discriminant validity of group task satisfaction was tested further by confrasting the pattern of relationships exhibited by group task satisfaction and group affective tone. Group task satisfaction was clearly differentiated from individual-referenced measures of group affective tone, but the differentiation between group task satisfaction and group-referenced measures of group affective tone needs further research.
The third study was conducted in a laboratory setting. In this study, group task satisfaction was manipulated by providing either positive or negative social information to each group. This manipulation did not have an effect on group performance, possibly because the group task satisfaction manipulation was not effective across all of the experimental conditions. However, group members' ratings of group task satisfaction were positively correlated with group performance (measured from the accuracy of the group's work on a bonus calculation task). The level of task interdependence was also manipulated. This manipulation was found to affect the strength of the relationship between group task satisfaction and individual job satisfaction.
In this study, matching items were used to measure group task satisfaction and individual job satisfaction. A consensus measure of group task satisfaction was also tested. It was found that even when group task satisfaction and individual job satisfaction were measured with the same items, there were differences between group attitudes and individual attitudes. However, the consensus measure of group task satisfaction displayed the same pattern of relationships as the individual-rated measure of group task satisfaction.
Together, these studies show that when individuals work in groups they develop shared attitudes, even under fairly minimalist conditions. Group members are aware of the shared attitude within the group towards its task, and their ratings of group task satisfaction are correlated with (but not equivalent to) measures of individual job satisfaction and group performance. It was concluded that there is utility in treating job satisfaction as a group-level construct and continuing to investigate how it affects individuals' experience of working in groups, and group outcomes such as performance and organisational citizenship behaviour.