Sociolinguistics and CMC

I've found it very frustrating in my department when colleagues decide to use CMC data without really understanding the structures of it. A lot of times it comes from laziness and wanting an easy data set, but even when their intentions are in the right place, they still don't know enough about the community and discourse structures to fully analyze things. (And then they come to me for advice, of course.)

I've been working on some paper ideas lately about how you can't be an ethnographer of an online community without being an insider. That you lose so much if you come from outside the structure and study it, and you're locked out of so much information in social networking sites if you don't actually make an effort to be part of the community. It's sort of related, though, to be an academic and come to online social network sites just to grab data and not actually understand the culture that is going on there or the conventions of communication.

Okay, off my soapbox now.
 

It seems to me that it is quite likely to be problematic if researchers are studying language datasets taken from Internet discourse without being cognizant of the structures of such discourse. I suppose its possible that over time, people doing this work might notice some of the differences, but then again they might not. Joshua, what is the rationale (as far as you can tell) people use for using online discourse in this manner?
 

rae actually raises a good point about doing ethnography online, which is an obvious issue here - in ethnography as i've been exposed to it, the researcher may or may not actually immerse themselves in all of the practices of the community, but in either case their status as researcher is typically known by the group being studied. while i've seen some really good internet ethnography done while engaging with the group and in the medium (lynn cherny's work comes to mind), there's also a tendency to lurk around a group for awhile and call it ethnographic research. assumedly, if ethnographic work is going to be done in an online group where the research isn't aware of or interested in theories behind online communities, this kind of lurking may be the norm for research. i don't necessarily agree that you can't do internet ethnography without becoming part of the group, though - i just think you'll get different persepectives depending on how you approach it.

i think part of the motivation is easy access to interesting data, to be honest. rather than looking for little communities of practice that focus on rationalizing eating disorders through community discourse about them, for example, you can find a perfect online community doing just that. Rather than trekking down to this group with a lot of video equipment to record interactions, you can just copy and paste their conversations into a transcript. Rather than needing a contact to really establish yourself within this group, you might be able to just log in with a nice degree of anonymity - and this possible anonymity among group members will probably also make it easier to get IRB approval for your research. And if you've already found a local group who does this and choose to study them, looking at this online group will help make claims that this isn't just a local phenomena, but has even spread out among the interwebs.

there are probably lots of good reasons, though just as many potential bad ones.
 

Hmmm...so two issues really [aside-just now as I was typing this I became quite aware of my use of elipses and I thought of you, Joshua!] the first being the recurring issue of insiderr/outsider status in ethnography; if you read the AoIR ethics guidelines (and from discussions I've seen at their conferences) it looks like more people who do online ethnography tend to favor disclosing researcher status, but I've heard plenty of good arguments for the other side. As far as needing to be a member already, there is plenty of traditional ethnography that involves the researcher not being a member prior to undertaking research, I'm not sure why, from the community perspective, that would be any different online. Anytime you undertake research on a new community, online or offline, there will be time at the beginning devoted to getting up to speed.

Now, online discourse/language, I think that's another matter, b/c the structure of the specific channel does make a difference. For example, a few years ago I did a discourse analysis project on a fan listserve (where yes, I had previously been a member). One of the things that interested me about this list was that it was distributed as digest-only, and therefore allowed for posts that were constructed with a high degree of intertextuality that made for very interesting ways of expressing ideas. If a researcher interested in language just went there to grab a chunk of talk, I think it's quite possible that not understanding the structure and affordances of the medium would potentially lead to incorrect/misleading conclusions about how language was used.

Plus it's just lazy.