[Air-L] a question about privacy protection and copyright in Internet research

Tue May 10 09:37:58 PDT 2011

I'm glad for the clarification, because now I know I disagree :).

The cases where the published documents ("texts") actually are
relevant to human subjects seem to be the area that is most
interesting to me. And I suspect that there is not a single published
document that does not require the researcher to be ethically
reflexive in that regard.

I suspect more people would agree with me that such ethical
reflexivity was necessary if not for the practical implication of
declaring that humans are involved (even if they may not be "subjects"
in a casual sense or in the sense OHRP claims). In other words, I
doubt Jeremy or anyone on this list is suggesting that with certain
types of study, ethics no longer apply.

The problem is much more practical, and that is that by suggesting
that most or many published texts also involve human subjects, it is
then necessary to take your study before a human subjects board.
Although human subjects boards exist for very good reasons, they (in
my experience) are often less familiar with the ethical issues
surrounding an individual's research agenda than the individual is
herself. Moreover, they tend to impede research that does not fit an
existing structure. That acts as a significant brake on innovation in
research, and means that a great deal of research does not get done.

Let's take Twitter as an example. Although there are exceptions (since
human subjects boards don't agree on the interpretation of particular
studies, nor--often--on their charge), Twitter is general considered
to be exempt from human subjects review. It is *so* exempt that most
people who study Twitter do not go to IRBs to get an exemption. If a
general understanding emerged that researchers examining Twitter
required IRB approval, far less good research would be done on
Twitter, and I think that would be a tragedy. (Or, more pointedly, I
think that would do harm to the subjects the process was trying to
protect, by robbing them of a deeper understanding of their social
interactions.)

Will one of these Twitter studies lead to harm? Yes, possibly. There
are parallel examples of people studying open discussions that led to
misinterpretation of the community (at least from the perspective of
the community) and that is equally possible for Twitter. I've talked
about my own ethical dilemma with a recent study of Twitter, and I
expect that this isn't so rare an occurrence. Nonetheless, I
emphatically would not have been able to do that study had it required
IRB approval, both because of the time required to get it approved,
and my own invested time to get it done.

So, I think it's important to split the discussion between "are there
ethical issues?" to which I can't imagine the answer ever being "no,"
and "should this be reviewed by an IRB?" to which I can imagine quite
a number of "nos."

Perhaps we need to think of alternatives. I would love to see journals
requiring a section that discussed (a) whether the research had been
reviewed by an IRB/human subjects board, (b) if so, a public
disclosure of the proposal documents (along with the data), (c) if
not, a discussion of the researchers' thoughts about risks and how
they were mitigated. This to me would result in a much better outcome
than turning things over wholesale to an IRB system that is frequently
clunky at best.

- Alex

On Tue, May 10, 2011 at 12:18 PM, jeremy hunsinger <jhuns at vt.edu> wrote:
> so i guess we should be clear here:
>
> we have nations, states and laws, that is one level of argument
>
> nation states and laws also have ethical review systems
>
> ethical review systems for research is another level of argument
>
> research ethics fit within those ethical review systems and are generally recognized by them
>
> though you may or your discipline may hold standards above and beyond those standards that you need to deal with
>
> my main issue here is to  promote the construction that published documents are not a matter of human subjects, or if they are it is in very rare cases, so rare that they need not really even be considered by human subjects review, which currently they are exempt in the u.s..  I want to resist the academic temptation to turn every textual object into a representational ethical subject and say.. no.. this is just like a book, or this is jut like a newspaper, and in the rare cases when it is not, i want to be very explicit about why it is not.  In terms of databases, I want to compare them to existing database practices in the social sciences and say this is just like x, and we have been treating x like this for z number of years years, if it is not like x, then say why.  However, any time you say why it is not like x, you cannot in my mind be using an argument that they could have used in the past.  In short, I'm trying to encourage and promote a rigorous approach to different
>  iation between subjects and non-subjects in research ethics on the internet, because in my mind, it is needed and we strongly need to resist the temptation to turn everything into a research subject.
>
> I am not trying to say that we do not need judgment, nor am I trying to argue that the traditions we have need updated, what I am saying is that we need rigorous clarity much more than not.  We need to say this is a document, this is not a subject, and this is a subject and this is not a document, here are the categories where there is confusion.
>
> the two categories we seem to be confused about in this discussion are:
>
> privacy
>
> property/copyright
>
> my position on privacy is that a document is public and not private, and if you have private information in a document you should really investigate the nature of that private information, because it has been published and may not even be private anymore.
>
> and the latter category is not really my expertise, we have lawyers on the list that can tell you much more closely about it.  But my suspicion is that most of the issues of privacy are confusions about who actually owns data or owns what rights to the data.  and what that means.
>
>
> so to reiterate, I'm taking a strong line that tries to make clear the differences between document and subject, and I welcome arguments, but i also see it as a process, i admit that i am arguing a standpoint and occasionally i am also playing devil's advocate to force an issue or to force debate on an issue, and I'm happy to be wrong, so long as I am wrong and we have much more clarity.
>
>
>
>
> _______________________________________________
> The Air-L at listserv.aoir.org mailing list
> is provided by the Association of Internet Researchers http://aoir.org
> Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>
> Join the Association of Internet Researchers:
> http://www.aoir.org/
>

-- 
//
// This email is
// [x] assumed public and may be blogged / forwarded.
// [ ] assumed to be private, please ask before redistributing.
//
// Alexander C. Halavais, ciberflâneur
// http://alex.halavais.net
//