I’ve posted about this study here, but just never really got around to posting about it again until now. (This semester was a whirlwind!) At any rate, here is a chopped-up (for length) and still mildly rough version (I write better than this, I promise!) of the write-up for my research methods microstudy–this one a replication of a 2004 survey on collaborative writing practices.
Noël, Sylvie, and Jean-Marc Robert. “Empirical Study on Collaborative Writing: What Do Co-authors Do, Use, and Like?.” Computer Supported Cooperative Work: The Journal of Collaborative Computing 13.1 (2004): 63-89.
Noël and Robert’s 2004 study (based on a survey conducted in 2001) set out to describe how collaborative writers work. Though many of their results dealt with group organization and leadership, intragroup communication, and satisfaction or productivity levels with collaborative projects, the main goal of their study was to “verify whether CSCW [Computer Supported Collaborative Work] tools were being used by collaborative writers and, if not, to discover just what technologies people prefer for collaborative writing” (69). The researchers were surprised to find that the most common tools for participants’ collaborative writing were not collaborative work tools, but instead were simply e-mail and word processors (usually Microsoft Word).
Since their 2001 survey, improvements in available bandwidth have allowed for more widespread use of Web-based tools—Noël and Robert mention this as part of their study, a point of comparison to earlier research on writing groupware. Also, the number of widely-available, and easy-to use collaboration tools, both free and proprietary, has increased—Google Documents (officially launched in 2007) is only one tiny example. An exponential spread in the use of wikis, blogs, and social networking websites as platforms for collaboration has also taken place in the last five years. This replication of their original study examines to what extent collaborative authoring practices have changed in the last five years, specifically regarding the software collaborative writers use to author their projects and the ways that software affects their work processes.
Methods and Participants
Because Noël & Robert’s original survey apparatus was included with their article, I was able to directly replicate many of their original questions, focusing and adapting their survey into a form appropriate for my focus on tools and processes. Of their original 44 questions, 20 were retained, adding answer selections to multiple-response questions as appropriate (i.e. adding text messaging, VOIP and social networking sites to questions about communication technologies, or adjusting the list of publication types in question 18), or adjusting their 7-point Likert scale to a 5-point scale. The major additions to the original apparatus were questions 12-15, which asked about synchronous online meetings; these were modeled on Noël and Robert’s question 10, which asked about face-to-face meetings. Questions about document audience, group roles, and project timelines were cut to simplify and focus this survey to a manageable microstudy scope.
In the spirit of replication, data was collected using the Internet; Noël and Robert discuss the strengths and weaknesses of such an approach at length in their article (69). The survey was created with www.surveymonkey.com and was distributed through hyperlinked e-mail and Twitter. Recruiting emails were sent to two mailing lists (WPA-L and TECHRHET) and the faculty e-mail list at my university; a recruitment tweet was sent on the same day, with the #twitrhet hashtag added. Reminder e-mails and a reminder tweet were sent out one week after the survey’s start date. Due to this delivery method, participants were largely from composition, writing, and communication disciplines. Noël & Robert’s original survey had only 42 responses over the course of a month (May 2001); this survey had a total of 105 over the course of two weeks (October 12-28, 2009).
Because they’re looong, I’m not posting them here at this time. You can get the basic idea in the next section. If you’re really interested, you can download my synthesis of the results in .pdf.
Noël and Robert found that despite improved broadband and networking capabilities, collaborative writers were not making much use at all of computer-supported collaborative writing applications (79). Five years later, it seems that collaborative writing practices have changed (at least in terms of this group of participants). Few participants in this study reported using specialized groupware; however, if free collaborative word processing software like Google Documents could be counted among that group, then many of the ways that collaborative writers work have been changed. The benefits of synchronous technology for composing, editing, and discussing collaboratively-written texts can be seen in many of the responses to this survey.
Noël and Robert point out a number of possible reasons for lack of success of CSCW systems: users must go to the trouble of not only installing the particular software on a variety of (potentially incompatible) systems but also learning a new tool and adopting it into already-developed work processes. Other problems include the complexity of many such systems, which tend to work against user acceptance. What can be seen from this limited study is that simple, integrated systems like Google Documents that allow for synchronous work are popular with collaborative writers. One of the possible approaches for designers of collaborative writing systems that Noël and Robert point out is to “integrate more collaborative features into popular word processors” (83). Google Documents is certainly one example of this approach. Despite this popularity, however, the predominantly used method of writing was still parallel rather than synchronous. Again, software limitations were more often cited than writer preference.
I was surprised—considering their widespread use on the Web—to see so few respondents reporting use of wikis for their collaborative writing. Reasons for its lack of adoption were most often related to lack of synchronous writing capability and a history feature perceived to be incompatible with the kinds of documents most respondents reported working on.
It is worth noting that this study has been rendered outdated in the few weeks since its completion. Google Wave, still in preview mode, is the kind of synchronous, integrated composing environment that will undoubtedly change the way collaborative writers work. Such a study as this one will have to be replicated after this software becomes more widely used.