When is it Ethical to Research Online Communities?
I’ve been a member of a medical Webmaster’s mailing list now for years. Most months, the conversations and discussion drift by me because they don’t interest me or they’ve been discussed to death in the past. (You can only have so many of the same discussions over and over again before tiring of them.)
Recently the topic of ethics and morality came up in regards to data mining mailing list archives (or really, any discussion forum that’s easily accessible on the Internet). This discussion was spurred by the recent publication of an article where the researchers had surreptitiously done research on a cancer support group (apparently without the group’s knowledge or consent). This is not the first time I’ve come across this behavior from professional researchers and it’s becoming increasingly disturbing.
Why do these researchers’ behaviors present a potential ethical dilemma? Two main reasons — the words are being used for an unintended purpose without the original authors’ permission and may be archived for all time. Because they take support group’s members’ words (often, without context) and publish them in a paper format, archived for all time. Sometimes pseudonyms accompany those words as the attributed author. Unfortunately, pseudonyms online can be the equivalent of an online identity and are just as easily traceable as a proper name.
The resulting discussion on the medical Webmasters mailing list focused on whether permission is granted by default, since mailing list’s archives can often be found through a search engine or elsewhere on the Internet. In other words, because it’s so easy to find these discussions online, the conversation is public. And because it is public, I can use that conversation (or “data”) in my research without anyone’s permission.
A number of side issues were brought up that I don’t think have much relevance to this discussion. Copyright really doesn’t apply, despite each contributor’s words being copyrighted once posted to a mailing list (unless they’ve given specific, written permission to waive their copyright rights). It’s not an issue here because the Fair Use copyright case law establishes that research is a legitimate use of copyrighted materials where permission need not be obtained. Other issues regarding the technology and ease of accessing archived discussions online were also presented. These too are basically irrelevant to the central ethical dilemma.
But something about hearing how easily justified this behavior was in some researchers’ minds made me rethink this last point. The justifications were largely based upon the technology — “this is what the Internet is all about.” Making information public. If you don’t want your words public, then don’t post to a support group.
Posting to a support group online is intended to help themselves or someone else get through a difficult period in their lives. There is a difference to posting to a mailing list or other online forum that may be public and publicizing one’s words on such a list. This is an important distinction lost upon the researchers. Just because something is public doesn’t mean ethics’ rules don’t abide. Nor does it give the researcher carte blanche for publicizing those words to further their own professional or research goals.
Technology doesn’t change the ethics of the situation, however. Just because something is more public than something else doesn’t make it more ethical to study without permission. Imagine a cash-strapped, small town allows a small cancer support group to use the town’s auditorium space for its weekly meeting. Do those group members now have no expectation of the same privacy and respect for that privacy than if they were to hold the same meeting in a room with a closed door? Just because a researcher could hide in the audience seats doesn’t mean he is now justified to do so. Just because a researcher comes across a support group’s conversation online doesn’t give him the automatic right to conduct research on that conversation. Technology alone cannot be used as a legitimate justification for such behavior.