Cameron Marlow, a PhD candidate at the MIT media lab, has put together a survey about weblogs to further his dissertation research, and it looks like a lot of folks I know have filled it out. As a fellow dissertator, I was glad to be able to help him out, and I think the project is a great idea. (From his links, he’s also doing all sorts of other cool stuff, as well.) However, I’ve got to say I found myself a little frustrated with the limits set by his questions in my attempts to respond to his survey. The MIT media lab, I think, could use a research methods seminar.
The survey’s minor infelicities — it asks if you met your family online; there’s no selection for “colleague” when it asks how you know someone; it requests occupational data in categories that reveal more about the author than about possible respondents — don’t take away from its value. More problematic, though, are the ways it frames the crawl it does on your blog: upon finding 'a href' tags (er, including in my case a link to an explanation of the 'rel' attribute that may have been helpful to Marlow), it asks you to classify the link, offering these options: weblog, weblog entry/post, personal homepage, part of my weblog, other. Are these really the only sorts of pages to which webloggers link? (I’ve heard a number of qualitative researchers observe that “other” often tends to get a whoooole lot of responses.) Actually, though, this may have been a really smart reason for Marlow’s limiting of the form’s options — but I fear he might have sacrificed research accuracy for research efficiency. As one commenter observed, it’s also troubling to see what the survey’s list of transactional motivations for blogging doesn’t include: learning isn’t on there, for example. Nor is the pleasure of writing.
Like I said, the project seems important, smart, and well-intentioned — but I also worry it’ll likely return some problematically dirty data.
Good points. I found myself frustrated when taking the survey as well.