Challenges and Recommendations
Overall, our experiences using the wiki were both mixed and surprising. While any new technology obviously brings with it limitations as well as benefits, what was most interesting to us was how--despite the common assumption that wikis enhance collaboration--our students did not find the wiki helped to achieve deep collaboration given their existing situation. Rather, many saw it as an impediment to the deep collaboration they were (already) engaging in prior to the introduction of the wiki. There were several reasons for this tendency. While it became clear that the choice of Clearspace software inhibited student collaboration in specific ways, making the completion of the assignment more difficult than we or the students would have liked, some of the challenges we faced apply to wikiware more broadly. Many students, for example, complained that site hierarchy was challenging to understand and to navigate. Although the particulars of Clearspace's organization of spaces and subspaces may not have been self-evident for our students, wikiware in general typically suffers from a lack of clear navigation paths and a confusing overall hierarchical structure. Ironically, this lack of rigid hierarchy and the fact that users can create or manipulate their own hierarchies seems to be one of the characteristics that earn wikis their reputation for being "democratic." Regardless, and more fundamentally, our research team determined the application of wikiware in our courses to be disappointing in terms of deep collaboration because of three factors.
The first factor that worked against deepening collaboration through the wiki was an attribute of our courses: Our students were co-located and most had already formed more-or-less close-knit communities. In this context, students noted, meeting in person made it easier both to compile documents and to critique the work of peers. Although we had hoped that the transparency of the wiki would allow students to preserve and negotiate differences in the iterated revisions, they seemed to prefer not to make such critiques without the added communication tools of body language and facial expressions when face-to-face meetings were still an option. Admittedly some students--those who did not know each other well before forming a group or those in groups with conflicting schedules who found it difficult to set up face-to-face meetings--found the wiki to be an added benefit for engaging in group work. For some students, the particulars of the software we had chosen made writing on the wiki less appealing than writing in Word and exchanging documents or using Google Documents, and other students remarked that they enjoyed having the wiki available as a common place to share all their documents. In fact, the most frequent users employed the wiki as a clearinghouse, a place to store documents or trade drafts which had been modified using other tools, such as Microsoft Word's track changes feature. These observations were first made in our 2007 pilot study, but were reinforced by remarks and usage patterns of students in the Spring 2008 PDI course. But this use of the wiki--as a clearinghouse--did not translate to more nuanced, more frequent interactions. Although we hoped having the wiki space available would encourage further contributions and interactions among group members, many students felt they already were collaborating "deeply" enough using other tools.
The second factor that worked against wiki-enabled deep collaboration in our courses was an attribute of wikiware: the potential of one contributor to overwrite the work of another. Many of our students complained that if multiple students were editing their group's documents simultaneously, someone's work would be overwritten due to the way versions are saved, regardless of whether users were separated by a distance or co-located. To avoid the potential for overwriting, only one student would work on a wiki page at a time, thus making simultaneous engagement impossible. Hence, the wiki encouraged asynchronous contributions--i.e., turn taking--which was one of the primary practices the research team had hoped to improve upon. Such "slowing down" of the writing process had the potential of fostering deep collaboration, because it made each version's changes more transparent and open for feedback, which was exactly what we had hoped the wikispace would provide. In reality, however, the fear of "overwriting others' work" became a serious implementation challenge and one that may have undermined the very goal of deep collaboration as we envisioned it, with intimate negotiation over content at each stage of its generation. The risk of overwriting another's work by not knowing whether a team member was simultaneously working on the same section at a given time provided a strong incentive for students to break up the task into discrete sections, assigning each to a single person for a specified period of time (e.g., until the next class meeting). In other words, students iterated each section individually and then handed-off their work for modification by others. While we had hoped wikiware's ability to update documents in real-time would encourage more back-and-forth exchanges among more individuals during the writing process, such exchanges were achieved only to a limited extent. Furthermore, regardless of the extent to which the quantity of interactions increased due to the fact that multiple contributions were required as part of the assignment, the quality of exchanges indicated that the wiki did not deepen the collaboration among student team members nor did it lead to improvements in their writing process or outcomes.
Perhaps related to the two factors listed, above the third factor that seemed to work against students' use of the wiki for deep collaboration had to do with the overall length of time allotted to complete a relatively narrow and specifically delineated type of work. The wikis were introduced mid-semester and the time-to-task we specified in the assignments themselves was only two weeks. Although groups were encouraged to continue using the wiki to complete their final projects, and indeed some did, for most groups use trailed off after the requirements were fulfilled. Granted the instructors deliberately chose to force "wiki traffic" within a limited time-frame. However, given the complications of the interface, however, perhaps the learning curve involved with using Clearspace might have been more readily absorbed during a longer window of involvement or ironically, if the assignment had been more complex. In fact students' responses to the second paper survey support this analysis. One student wrote, "[t]here was little time to begin using the wiki for the assignment," while another student complained that the wiki was, "[o]bnoxious. Frustrating. It is useful to use a wiki when the project is the only thing you are doing. But, when the project is only a few hours a week, going online to chat with the group when you could just wait until class is relatively easier. Furthermore document sending is just as easy to do over email. If this project took 20 hours a week, and if it was difficult to get the team together to discuss, it would be very useful. But it is not really in this instance,"and a third student suggested that they would "[n]eed more time in order to fully experience it. The collaborative process was working at the same time on the same work."Perhaps larger projects that unfold over a longer time-period might benefit more from such wiki-aided use. In the future, instructors may want to introduce the wiki earlier in the semester, even before the assignment is distributed to give students a chance to play around in the interface and thus become more familiar with it.
Our research suggests that the trust and respect necessary for deep collaboration is something that develops along with interaction, whether it is facilitated by a wiki or conducted face to face. Our goals of measuring and then correlating the degree of social networking and cohesiveness with (deep) collaboration on the wiki were difficult to achieve for several reasons. First, it was impossible to isolate the specific factors leading to deep collaboration. This difficulty was magnified by the fact that our wiki teams were small and that many of the team members already knew each other well. Second, the prerequisites for achieving deep collaboration--such as trust and respect among group members--could be developed independently of, and in parallel with, the wiki's usage, making it impossible to isolate the wiki's contributions to these factors. Third, while we hoped the wiki would increase the frequency and quality of interactions among group members who did not already know one another, the incentive to interact through the wiki was clearly much higher among group members who did not already know one another or who were not co-located.
Despite these limitations, however, both instructors were pleased with the way that the wiki assignment encouraged students to engage in the writing process earlier than they otherwise might have. Moreover, the wiki provided a public forum for student work to be displayed, so that students and instructors alike could see who was writing what and when. The wikis also made the evaluation of individual contributions easier to accomplish, especially in courses where students' writing grades were based on individual contributions. Although students complained mostly about specific technical limitations of the software, we and they are aware that all new technologies come with interface challenges and wikiware platforms will be improved in time and adjusted to the specific needs and goals of particular user communities. Setting these concerns aside, however, our students' analysis of their use of wikis suggests they could be potentially powerful tools for harnessing collaboration in teams that are not geographically co-located. Additionally, for teams that are co-located, wikis might function best as a clearinghouse for document storage--a function that emerged from how our groups actually used the wiki. Wikis also allow distributed, process-based, iterative writing to be much more transparent for those who wish to understand (and evaluate) the evolution of a document and the contributions made by specific individuals. Thus, while we certainly do not recommend against using wikis on collaborative writing assignments--they provide valuable administrative and assessment opportunities--we do recommend against employing them as a substitute for other trust-building activities that develop team cohesion or collaborative-writing skills-building activities that facilitate peer-review and editing capacities.
Continue reading the Conclusions section.