Abstract:

This study investigates the role of drop-out in virtual collaboration. Specifically, the main objective is whether there is a difference between groups with drop-out versus groups without drop-out relating to group processes regarding the task- and social level of collaboration, as well as relating to group performance. Looking at group processes, on the task-level goal orientation and task completion are involved. In respect of the social level group cohesion and taking responsibility are of relevance. While goal orientation focuses on the impact of the goal, task completion forms the main purpose of collaboration. Group cohesion describes the tendency for a group to stick together and remain united; taking responsibility is central for the whole collaboration to prevent dysfunctional phenomena. Drop-out may change virtual collaboration on the task- and on the social level in the following way: The task-level may be involved in drop-out as contributions relevant for the task solution are suddenly missing. On the social level, remaining group members may feel left alone and helpless when their group members leave collaboration.
Group performance is based on the task solutions of all group members. As drop-out reduces the amount of group members and specifically their contributions, also performance may be affected by drop-out. Thus, the main question we investigated is whether there were differences between groups with drop-out versus groups without drop-out in respect of the group processes, namely on a task- and social level, as well as in terms of group performance.
To gain further insights, we evaluated the seminar “Development and Implementation of virtual Learning Scenarios” at two points in time during the winter term 2012/2013. Thirty-eight participants mainly studying pedagogy were asked to evaluate the virtual seminar in a survey rating the above mentioned dimensions at two points of time. They received the first online questionnaire to be filled-in exactly in the middle of the semester, and the second at the end of the semester. This questionnaire included the four dimensions goal orientation, task completion, group cohesion, and taking responsibility for evaluating virtual collaboration using a five-point-Likert scale from 1 (“do not agree”) to 5 (“totally agree”). To measure drop-out, we asked participants whether there was a change of participants in their group. We also investigated group performance using all 42 task solutions of the groups. These task solutions were rated according to a specific coding scheme validated by two persons. Inter-rater reliability was sufficient.
First results based on t-test analysis showed that at the first point of measurement, groups with drop-out significantly differ from groups without drop-out in all four dimensions indicating that groups with drop-out evaluated goal orientation, task completion, group cohesion, and taking responsibility significantly lower than groups without drop-out. At the second point of measurement, task completion, group cohesion, and taking responsibility were again evaluated significantly lower in groups with drop-out in comparison to groups without drop-out. Furthermore, looking at group performance, again groups with drop-out received lower scores than groups without drop-out. These results indicate that drop-out is of great importance for further collaboration in virtual learning environments which must be investigated in further detail.

Keywords:

Drop-out, collaborative learning, task-level and social level, group performance.