Abstract

In spite of the large extensive literature reporting the effects of motivational variables on performance, very few studies have focused on the dynamics at the group level. None has empirically tested the transference of perceptions of group efficacy to different teams, nor have these studies taken advantage of computer mediated communications (CMC) as a research tool, where members of a group can interact in a virtual space, thereby suppressing many personal biases. Most of the discussions concerning the relationship of these variables have been held at the theoretical level, thus awaiting empirical testing. These issues seem to be particularly relevant to the new trends in education and human resources practices, which emphasize team collaboration and encourage or require the participation of individuals in multiple and diverse groups. The purpose of this research is twofold: (i) to investigate the effects of efficacy-performance spirals at the individual and group levels, as produced by the manipulation of feedback, and (ii) to study the transference of efficacy beliefs (about the self and about the group) to new working groups. Computer-mediated communication will be used as a tool to investigate these issues. The theoretical framework for this research is based on Bandura' self-efficacy theory, and Lindsley's et al . (1995) model of multilevel efficacy-performance spirals. According to this model, feedback, task uncertainty and task experience are factors affecting the generation of efficacy-performance spirals. Since, among several variables, the effect of feedback on self-efficacy-performance spirals at the individual level is well established in the literature, this variable was manipulated. Three research questions were proposed: (1) Do the efficacy-performance cycles generated at the group level affect individual perceptions of self-efficacy? (2) Do individual members transfer their perceptions of group-efficacy to other groups when faced with similar tasks? (3) Do individual members transfer their perceptions of self-efficacy to other groups when faced with similar tasks? Data was collected from a sample of 86 undergraduate students in a Mexican University. The virtual groups consisted of four to five students. These teams were divided into two conditions: with feedback and without feedback. Participants worked on two activities of collaborative writing, each to be performed with a different group. Various variables which, according to the literature, may affect self-efficacy were measured before, during, and after completion of the tasks. Appropriate multivariate techniques were used for data analyses. Results indicated a positive relationship between the perceptions at the group and individual levels while testing for transference of efficacy beliefs between both activities (r = 0.58, p < .01 for the individual level, and r = 0.52, p < .01 at the group level), thus suggesting some kind of transference had occurred. However, t-tests yielded significant differences (t = -3.07, p < .05 at the individual level, and t = 4.2, p < .0001), thus suggesting that these perceptions change positively from one group to the next. The MANCOVA analysis of perceptions while working in the same group seemed to confirm this conclusion. The effect of group efficacy-performance spirals to the individual level indicated no significant results. However, descriptive data suggest differences worthy of further research, and the need to consider other strong variables in the model. Implications for theoretical interpretations, for further research, and for pedagogical practice are discussed.