AN EMPIRICAL STUDY ON THE IMPACT OF ANONYMITY AND PROXIMITY ON STATUS INFLUENCE IN A GSS ENVIRONMENT

Abstract

An empirical study was conducted to help understand how GSS under certain settings affect the status influence within decision making groups. A laboratory experiment was employed to test the hypotheses presented in the study. The communication modality in group interaction process was manipulated by focusing on cue and anonymity while holding other factors constant. The cues including verbal, visual, and textual were controlled by varying support level between manual and GSS support and proximity between face-to-face and dispersed. The anonymity was controlled by either revealing or concealing the identity of the contributors. Preference task which occurred frequently in organizations was chosen in the study. A total of 240 first year undergraduate students from the DISCS, National University of Singapore, participated in the experiment. They were randomly assigned to 60 groups together with 4 teaching assistants from the same department who acted as confederates. The GSS employed was SAGE (Software Aided Group Environment) developed at National University of Singapore. The dependant variables in the study measured status influence, sustained influence, residual disagreement, perceived status influence, perceived self influence, and perceived decision confidence. The experimental data were analyzed using JMP by SAS Institute Inc. It was found that: 1. The manual group experienced higher status influence and sustained influence than the face-to-face and the dispersed GSS groups. 2. Residual disagreement was lower in the manual group than in the face-to-face and the dispersed GSS groups. 3. Perceived self influence in the manual group was lower than in the dispersed GSS group. 4. There were no significant differences between the manual group and the face-to-face or dispersed GSS groups in mean responses of perceived status influence and perceived decision confidence. 5. No significant differences were found in all the six dependent variables either between the face-to-face and the dispersed GSS groups or between the identified and the anonymous groups. By employing GSS technology with the removal of verbal communication, status influence can be significantly reduced at the expense of higher residual disagreement. A context with strong status influence is more efficient in resolving disagreements. This study has also shown that anonymity, dispersion, or removal of visual cue help enhance perceived self influence after removing the verbal cue. These findings are helpful for clarifying the uncertainties and for anticipating the consequences in the adoption of GSS technologies, especially when effects of status influence are concerned. However, as pointed out in this study, several related issues require further investigation. This line of endeavour should continue before researchers in the field of GSS can claim complete understanding of patterns of social influence under GSS settings.