In lieu of an abstract, here is a brief excerpt of the content:

herding cats online Real Studies of Virtual Communities dmitri williams and li xiong The data-collection methods were real, even if the places weren’t. So, when did we realize our research was going to be a little different? Was it the time when one potential research subject said to another, “Cry more, noob!” or the time we were accused of running a pyramid scam? Maybe it was when lawyers shut one study down. Or perhaps it was the time when colleagues passing by our of‹ce gave us the Uh huh, sure you’re doing research look for the fourth time. We knew going into the work that this was “brave new world” territory, but we still didn’t fully appreciate exactly how brave or how new. In this chapter, we will share our experiences and lessons learned in collecting data from online populations. In particular, we will walk through the process, describing the pitfalls of conducting systematic online research with that most dif‹cult of quarry: the anonymous online gamer. And although our methods had to evolve with a changing world, we sought at every step to retain and translate all of the social science requirements that make a study valid and generalizable—to let rigor duke it out with messy reality. The studies from which we draw these notes took place from 2002 to 2007, and all involved a type of online game known as a massively multiplayer online game, or MMO for short. If you have not seen one, imagine 122 a table top Dungeons and Dragons session taking place online—with several thousand people all over the world. Millions of players now populate these games, spending about 24 hours a week online with each other rather than in front of a television (Yee 2006). That bizarre milieu of orcs, elves, accountants , housewives, and broadband connections is our workplace. The source material for this chapter is based on three major studies that we conducted, each of which generated a large data set that led (or is in the process of leading) to publications in peer-reviewed journals. The ‹rst study came from Williams’s dissertation and involved the MMO Asheron’s Call 2. Despite the several ups and downs to be recounted here, that effort was productive (Williams 2002, 2003, 2006a, 2006b, 2006c, 2006d; Williams and Skoric 2005). The second study was a two-part test of World of Warcraft (WoW) users (Williams, Caplan, and Xiong 2007; Williams et al. 2006), and the third study is an ongoing effort with Sony Online Entertainment that we cannot talk about completely because of nondisclosure agreements (NDAs). No worries. We did not know what an NDA was in the beginning either. (It is a legal document that limits what you can say to people who are not associated with the project or company.) So let us start at the very beginning. Let us assume that you are an aspiring graduate student or junior faculty member ready to jump into the world of online data collection. You have essentially three distinct phases to plan for. The ‹rst is the groundwork phase, which includes ‹guring out where the data will come from as well as dealing with various bureaucracies and culture clashes within the ‹eld. Having fought these initial battles, next comes the actual execution of the plan, including the sampling, recruitment , surveying, and general wrangling of the subjects. Last comes the analysis itself, which we will not cover in much detail since it is available in the papers cited earlier. planning: send lawyers, guns, and money . . . Some social scientists are lucky enough to work with existing data sets. In those cases (e.g., see Freese in this volume), the challenges mostly involve the data not being exactly what is the most relevant or appropriate to their research questions. In our case, the data have almost always been what we wanted. The problems arose from having to justify the data collection in the ‹rst place. And here we faced three consistent major hurdles: complying with rules regulating the protection of human subjects, corporate relations , and political considerations. Herding Cats Online • 123 You Want to Do What? Institutional Review Boards are a cautious lot by nature. After all, it is their job to safeguard the public from our clipboard-carrying, white-lab-coatwearing , electroshock-inducing, prison-experimental ways. So who can blame them for starting out with suspicion? Still, our problems were initially less about risk to subjects than about explaining what...


Additional Information

Related ISBN
MARC Record
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.