[Air-L] Air-L Digest, Vol 51, Issue 31

Bates, Benjamin J bjbates at utk.edu
Thu Oct 30 07:29:30 PDT 2008


As an IRB member at my university, the key initial question is whether
the email archive has been published (i.e., whether it is publicly
available).  As a general rule, content analysis of pre-existing,
publicly available, materials do not require IRB review as it does not
involve interaction with research subjects.
If there is a question about whether the material is publicly available,
you might want to go ahead and file an IRB - and the key concern will be
in making sure you don't use materials in a way where "private" comments
are divulged in a way that can be linked to a specific individual.  I
think that doing things like stripping names and other identifiers would
help approvals, or stressing that you're using the text in ways that
don't report specific comments, then there is less of a potential
problem, and you may be able to not have to track people down for
retroactive consent.
Still, if you have questions about the ethics, or your institution's
procedures, it's good to check with your local research compliance
officer.  

Ben Bates
University of Tennessee

------------------------------

Message: 2
Date: Tue, 28 Oct 2008 22:38:22 -0400
From: Dhanaraj Thakur <dthakur at gatech.edu>
Subject: Re: [Air-L] question about IRB and content analysis of
	mailing list
To: air-l at listserv.aoir.org
Message-ID: <7.0.1.0.2.20081028215920.027d2bc8 at gatech.edu>
Content-Type: text/plain; charset="us-ascii"; format=flowed

hey all,

this might have come up before so pardon me if this is repetitive.

I am interested in doing a content analysis of texts within an email 
archive. and I was wondering how to deal with IRB issues in this case.
specifically, I am focusing on the text and not the authors. if I did 
need informed consent its seems impractical to get given the number 
of participants (several hundred) and age (over 5 years) of some of 
the emails. what should I do in that case?
also, what would be the approach if the archive is available publicly 
or if its private?
finally, are there any other major IRB requirements that I should 
consider in this type of exercise?

thanks
Dhanaraj



Dhanaraj Thakur
Ph.D. Student
School of Public Policy
Georgia Institute of Technology


------------------------------




More information about the Air-L mailing list