Blog research: fine line between public/private

Researchers, IRBs should calibrate protections based on bloggers' privacy intent

As researchers and IRBs navigate the brave new world of Internet research, it's tempting to lump various types of research venues together — applying the same rules to social media networks such as Facebook that they use for Internet chat rooms and blogs.

But that would be a mistake, says Linda Eastham, MSN, RN, FNP, a nursing instructor at the University of Virginia in Charlottesville. Eastham recently published a paper in the journal Research in Nursing and Health about ethical decisions in conducting research on the content of blogs. She says the accessibility of information and privacy expectations are very different depending upon the venue.

When Eastham, who is a doctoral candidate at Virginia Commonwealth University, decided to study the content of blogs that chronicle people's health care experiences, she encountered difficulties in writing her proposal to her IRB.

"I assumed that I would simply turn to the literature and that the literature would answer my questions," she says. "But I couldn't actually find an answer to my question, which was specifically about blogs. The literature addressed social network sites or Internet research in general — it was very global, but not specific."

So Eastham delved further into the topic, and found it to be far more nuanced than she had thought. Despite the public nature of many blogs — a named or anonymous writer posting continual updates for others to read — there are actually degrees of difference that can signal a blogger's intent to be more or less private.

Those degrees of difference, Eastham argues, should help a researcher form a similarly nuanced approach to the question of whether it's ethical to quote from blogs as a sole source of data.

"Initially, I think many people jump to the conclusion that if you put it on the Internet, of course it's public," she says. "But being on the Internet doesn't necessarily make it public. People's intent, I think, is what's missing from that perspective."

Identification risk

If a researcher is simply pulling information and quotes from blogs, but not surveying or interacting with the bloggers personally, the major ethical consideration for researchers and for IRBs is whether the information used contains identifiable private information. Even if the blogger is anonymous, Eastham points out, quoting a blog can indirectly identify someone. If the blogger uses his or her pseudonym used elsewhere on the Internet, it can lead someone to information that can help identify the blogger.

When determining whether to use something from a blog, Eastham considers a number of the blog's characteristics:

— Is the blog private? When setting up a blog, an author can choose to have only certain people see it, and may require a password to view it. A researcher may hear about this blog from someone who has access to it or may find a link to it elsewhere.

— Has the blog been indexed for search? While a blog may not be strictly private or password-protected, the author may choose not to make it searchable by Google or other search engines. Again, while a researcher may not find this blog on a routine search, it may be linked to from other similar blogs.

"The researcher might be looking at Topic A and find six blogs," Eastham says. "And one blog author may follow Blogs X, Y and Z. By a snowball effect, you might find other blogs that theoretically wouldn't come up on a search engine."

— Does the blog turn up only in cache form? This would indicate that someone started a blog, decided to discontinue it and deleted it.

— Does the author allow readers to subscribe to a RSS feed, which delivers new posts to them?

— Does the author allow reader comments? Does he or she answer them and interact with readers?

Honoring privacy intent

Eastham says answers to these questions can form a picture of the degree of privacy that the blog's author intends. The more private the blog appears to be, the more care should be used in quoting material from that blog, she says.

"I think it's an issue of considering the intent of the blog author and doing our utmost as a researcher to honor that intent," she says.

Researchers could respond to higher levels of privacy on a blog by excluding content from it, by obtaining informed consent from the blogger or by masking certain details to minimize the risk of identification.

Eastham says that a conversation between the IRB and the researcher can help clarify how to calibrate privacy protections for blog material.

In her own case, when she first approached an IRB to talk about researching blogs, "The response was, 'Oh, you're going to have to get everybody's permission,'" Eastham says.

She eventually went back to the IRB with a plan that excluded blogs that existed only in cache and any password-protected blogs. She specifically sought out blogs that allowed reader responses. That new plan, she says, met with the IRB's approval.

"I think it would be healthy for there to be dialogue, to have the researcher and the IRB be clear about what the researcher wants to do," Eastham says. "Are these blogs going to be indexed for search? How are you going to locate these blogs? And let the researcher explain."

Reference

Eastham LA. Research using blogs for data: Public documents or private musing? Res Nurs Health 2011 Aug;34(4):353-61.