Contingency theory predicts that using computer-mediated communication to accomplish complex collaborative work will be difficult, especially for tasks that require interactive, expressive communication. This proposition was examined in an experiment in which 67 three-person groups of MBA students completed two collaborative writing projects under either Computer Only, Computer + Phone or Face-to-Face communication conditions. The effects of these manipulations on group processes and performance were examined using data obtained from questionnaires and scores on the projects themselves. Although communication modality did not affect project performance, being restricted to computer-mediated communication made completing the work more difficult and diminished the participants' satisfaction with their work and with the other members of their work groups. The results also provide partial support for the idea that tasks that require more intensive communication, such as project planning, were more difficult than those that can be completed more independently, but this premise was not consistently confirmed. Taken together, these findings tend to confirm the contingency hypothesis regarding the difficulty of accomplishing work that involves ambiguous goals, multiple perspectives, and information that is susceptible to multiple interpretations without an interactive multiperson communication medium, such as face-to-face meetings. However, the results also suggest that modifications of contingency theory are required to incorporate the evidence that individuals can, to some extent, adapt to restricted communication channels. Further research designed to examine patterns of adaptation under various task/technology combinations is recommended.
This study examines the influence of anonymity on group process in groups using a group decision support system (GDSS) with an idea-generating task. Group members whose contributions were anonymous generated more comments, were more critical and probing, and were more likely to embellish ideas proposed by others than were those whose contributions were identified by name. Implications for group support research are discussed.