Ethics of Research on Usenet
Al Black
social scientists should follow when performing research on Usenet
postings(1). This survey has been cross-posted to comp.admin.policy,
news.groups, and news.admin.policy.
Anthropologists, Communication Researchers, Discourse Analysts,
Linguists, Psychologists, and Sociologists are beginning to treat
Usenet postings as data(2, 3). These researchers are obligated by
their professional associations and institutions of employment to
follow ethical research practises that have been developed through
years of experience. The established ethical practises, however, do
not apply well to the study Usenet postings. The following survey is
an initial attempt to obtain some data about the opinions and
preferences in regard to research on Usenet. We will forward the
results of the survey to the appropriate associations, and publish the
results in a variety of places. As seen below, most of the questions
concern preferences about specific practises that researchers could
perform, although there are general questions about research on Usenet
postings along with a few demographic questions(4).
To complete the survey, respond to the following questions, and e-mail
it to a...@debra.dgbt.doc.ca with "ethics survey" as the subject line.
All responses will be kept confidential, and rendered anonymous by
publishing statistical (aggregate) results. A short version of the
results will be posted on news.misc, and a more detailed version will
be available on the Communications Research Centre Gopher server at
debra.dgbt.doc.ca, or from the authors.
This survey is being conducted by Alex Black and Malcolm Parks(5).
Alex Black is Visiting Fellow at a Canadian Laboratory at the
Communication Research Centre, and Malcolm Parks is Associate Professor
in the Department of Speech Communication at the University of
Washington. We appreciate your participation, and welcome any comments
that you have on the survey or the general research project.
Thank you
Alex Black (a...@debra.dgbt.doc.ca)
Malcolm Parks (ma...@u.washington.edu)
(1) For the purposes of this survey, the term Usenet is used in the
inclusive sense, including the alt, K1-12, soc, talk, and USENET
hierarchies. This survey does not apply to public or invited mailing
lists.
(2) Hereafter the term "researcher" will be used to refer to the social
scientists who interests in language, discourse, or culture lead them to
the study of Usenet. Please note that we are limiting the term to
professional researchers or academics (e.g., graduate students or
professors).
(3) For example, a large scale collaborative effort to examine Usenet
postings and other C.M.C. (Project H) is being conducted presently.
Information about the project can obtained by anonymous ftp at
archsci.arch.su.edu.au.
(4) A detailed rationale of the survey and general research project
may be obtained at Communications Research Centre Gopher server at
debra.dgbt.doc.ca or by contacting the authors.
(5) Research biographies and publications lists of Alex Black and
Malcolm Parks can be obtained Communications Research Centre Gopher
server at debra.dgbt.doc.ca.
------------------------------------------------
PART I. ATTITUDES ABOUT USENET POSTINGS AND RESEARCH ON USENET
Please indicate how much you agree or disagree with each statement
below. Select the one number (1-7) that shows show much you agree or
disagree with each statement.
7 = Strongly Agree
6 = Agree
5 = Somewhat Agree
4 = Neutral, Neither Agree nor Disagree
3 = Somewhat Disagree
2 = Disagree
1 = Strongly Disagree
INSERT THE APPROPRIATE NUMBER IN THE SPACE FOLLOWING EACH ITEM.
1. I consider the messages I post to Usenet to be public: _____
2. When I post I am usually responding to the writer of a previous
post more than I am addressing the newsgroup as a whole: _____
3. If someone wants to send a private message, they should send it
directly to the recipient rather than posting it to the newsgroup as a
whole: _____
4. It is acceptable for me to pass along addresses of Usenet posters
to friends or colleagues who do not read or post to the newsgroup:
_____
5. I would be acceptable to me for system administrators to use my
Usenet posts as data in a study: ______
6. It would be acceptable to me for social science researchers to use
my Usenet posts as data in a study: ______
7. It would be acceptable to me for people who read but do not post to
use my Usenet posts as data for an informal study: _____
8. It would be acceptable to me for my Usenet postings to be presented
as in example in a public forum such as a newspaper, presentation, or
lecture: _____
9. It would be acceptable to me for my Usenet postings to be presented as
an example in a public forum such as a newspaper, presentation, or lecture
*only* if it was altered in ways that so that I could not be personally
identified: _____
10. Members of a newsgroup should collectively be able to decline or veto
researchers from analyzing their posts: _____
11. I should be able to read any message posted in a newsgroup: _____
12. It is acceptable to tell my friends or colleagues about what I
read on the Usenet: ____
13. It would be unethical of someone to forward a message I posted to
some third party who might be interested in it: ____
14. To me there is no difference between a researcher using my posts
as data and a lurker reading posts: ____
PART II. YOUR OPINIONS ABOUT ACCEPTABLE RESEARCH PROCEDURES.
In the following section we are asking your opinion about methods that
social scientists might use for contacting people, using posts,
establishing credibility, and informing participants in a study of the
results. Please indicate how much you agree or disagree with each
statement below. Select the one number (1-7) that shows show much you
agree or disagree with each statement.
7 = Strongly Agree
6 = Agree
5 = Somewhat Agree
4 = Neutral, Neither Agree nor Disagree
3 = Somewhat Disagree
2 = Disagree
1 = Strongly Disagree
1. It would be acceptable for a researcher to contact me by e-mail
asking me participate in a study or survey: _____
2. It would be acceptable for a researcher to contact me by e-mail
asking me for permission to include my Usenet postings in a
study: ______
3. It would be acceptable for a researcher to contact me by e-mail to
ask questions regarding the content of my posts to Usenet: ____
4. It would be acceptable to me if a researcher used my posting to a
particular newsgroup for data, if they had previously posted a
declaration that they might be using postings for research. Please
assume that researcher specified a time span, and you were posting
during that time span: _____
5. It would be acceptable to me if a researcher used my posting to a
moderated newsgroup as data, if the researcher previously contacted and
obtained permission from the moderator of the newsgroup: _____
6. I should be able to prevent researchers from using my postings to
Usenet as data by including a statement to that effect in my the body
of my postings or in my signature: ____
7. I would be able to evaluate a researcher's credibility to my
satisfaction if they provided a third party reference (e.g., a system
administrator or university/research organization contact): _____
8. I would be able to evaluate a researcher's credibility to my
satisfaction if they provided their list of publications and profession
associations at a FTP site: _____
9. I would be able to evaluate a researcher's credibility to my
satisfaction if they presented a method of personal contact and
professional affiliations in their signature: _____
10. When being informed about a study, I would be satisfied a *brief*
description of the study rationale, how the researcher will examine my
postings, and what the final product will be: _____
11. When being informed about a study, I would be satisfied with a
*detailed* description of the study rationale, how the researcher will
examine my postings, and what the final product would be: _____
12. When being informed about a study, I would be satisfied with a
brief description of the study rationale, how the researcher will
examine my postings, and what the final product would be, along with
the availability of a more detailed version that is available
by ftp: _____
13. When being notified about the results of a study, I would be
satisfied with a posting to news.misc: _____
14. When being notified about the results of a study, I would be
satisfied detailed posting to news.misc and a more detailed discussion
available by ftp: _____
PART III. GENERAL INFORMATION.
1. Approximately how many newsgroups do you currently read regularly,
that is, once a week or more? _____
2. In years, how long have your been reading newsgroups? _______
3. Which newsgroups do you read most often? (Insert the the names of
the five groups you read most regularly.)
4. How long has it been since you first posted on the Usenet? (Please
insert the number of years and months it's been since you first
posted.) _______
5. To approximately how many newsgroups do you currently post on a
regular basis, that is, once a month or more? _____
6. To which newsgroups do you post most often? (Insert the the names
of the five groups to which you post regularly.)
7. Are you a system administrator or newsgroup moderator? _____
8. What is your gender? _____
9. What is your age? _____
10. What is your occupation?
_________________________________
PART IV. ADDITIONAL COMMENTS
Please feel free to add any other comments or reactions you may have....
--
Alex Black
Communications Research Centre, Ottawa, Canada.
a...@debra.dgbt.doc.ca
r...@max.u.washington.edu
so to speak, I'd like to comment on this questionnaire.
I think the ethical boundary is drawn (in terms of current regulations,
certainly, in the US) between doing research using published texts --
which is what, in my professional opinion, postings to newsgroups are --
and research in which the investigators are eliciting or provoking
responses for research purposes.
Studying newsgroup postings is like studying letters to the editor
published in newspappers, or studying published books of essays.
The only problem arises from quotation -- and that's a copyright problem, not
a research ethics problem.
Another conceivable problem would arise if one went beyond what
had been published to try to discover eg the actual name of a
pseudonymous or anonymous poster.
Thus, the questionnaire itself would count as the kind of research
that a University of Washington investigator, at least, would be required
to submit for Human Subjects review.
Nome too that this does not apply to study of private communications --
e-ail, mailing lists, and so on. For research into such communications,
full IRB review and approval would probably be required.
Richard Brzustowicz Jr
Human Subjects Review Coordinator (Behavioral)
Human Subjects Division
University of Washington (Seattle)
r...@max.u.washington.edu
Al Black
>As someone who is not a professional ethicist, but more of an ethics tech,
>so to speak, I'd like to comment on this questionnaire.
I am pleased to see some follow up discussion about the survey and the
ethics of research on Usenet postings. Your post raises many different
issues about the ethics of performing research on Usenet posts and why we conducting doing the survey.
Neither Mac Parks nor I are professional ethicists. We are
researchers, and we have found through experience that it is best to
follow the wishes of people who are effected by the ethical decisions.
In this instance, we've found that the people who responded to our
survey and other Usenet posters that we have consulted with have
strongly held, and well reasoned opinions. These people can more than
adequately state what standards of conduct are acceptable to them.
>I think the ethical boundary is drawn (in terms of current regulations,
>certainly, in the US) between doing research using published texts --
>which is what, in my professional opinion, postings to newsgroups are --
>and research in which the investigators are eliciting or provoking
>responses for research purposes.
>
>Studying newsgroup postings is like studying letters to the editor
>published in newspappers, or studying published books of essays.
This is the crux of the issue as to whether or not researchers need to
obtain permission from posters. Are Usenet postings equivalent to
"letters to the editor" or "essays in published text"? Certainly,
postings are public and mass distributed, and in this sense the analogy
is true. But are they really a one way broadcast? Any particular
posting may be a follow up to a previous post (or posts), and in this
sense a singular poster may addressing the other person as much as the
entire world. It is this interactive element of Usenet that
distinguishes Usenet posts from newspapers or collections of essays. A
more accurate analogy of what Usenet is might be "a whole lot of
conversations in a public place". No researcher would be allowed to
tape record conversations in a restaurant without consent, but a
researcher studying Usenet postings without consent is doing just
that.
But note this is an empirical question about what model of
communication best describes Usenet postings. In the absense of data
it would be unwise to derive ethical standards of conduct from a moot
assumption. Especially when we can ask the people what conventions or
practises are acceptable to them, and avoid the assumption altogether.
>The only problem arises from quotation -- and that's a copyright problem, not
>a research ethics problem.
Agreed, quotation and copyright is a separate issue.
>Another conceivable problem would arise if one went beyond what
>had been published to try to discover eg the actual name of a
>pseudonymous or anonymous poster.
Yes invasion of privacy and confidentialty the is another issue. It
also applies to people who are not anonymous or pseudo-anonymous.
(Please note that responses to our survey are confidential, and only
questions that require the the disclosure of personal characterists are
age, gender, and occupation. These latter questions are only used for
comparisons between groups. Of course permission for quotations will
be obtained.)
>Thus, the questionnaire itself would count as the kind of research
>that a University of Washington investigator, at least, would be required
>to submit for Human Subjects review.
By your definition, "research in which the investigators are eliciting
or provoking responses for research purposes", any survey conducted on
the Internet would require Human Subjects review. Judging from
responses to the survey and discussion with people on the net and a few
examples I've pulled down from newsgroups, there should be a concern
for H.S. committees. Simply put, grep for "survey" or "questionnaire"
on your newsfeed every now and then over the next year. You'll be
surprised by the number of surveys that originate from the University
of Washington that have not had Human Subjects review. Some of these
may be may even be in violation of the University of Washington's
standards of conduct.
Surveys on the net are yet another issue altogether in regard to
standards of conduct. Hopefully, in the later stages of this research
project we can do some research about surveys.
>Nome too that this does not apply to study of private communications --
>e-ail, mailing lists, and so on. For research into such communications,
>full IRB review and approval would probably be required.
Another a murky issue. Some mailing lists are gatewayed to Usenet
groups. Some public mailing lists are like Usenet groups: the only
difference is in the manner of distribution. Then there are private
mailing lists or Usenet groups with little propogation and a small
number of users. Drawing the line here is difficult, but from my
impressions these are different, and that's why we`ll study them
separately. Again, the people who are using these media to communicate
with one another know what are acceptable standards of research conduct
better than us researchers or ethics techs.
Cheers
Alex Black
--
Alex Black a...@debra.dgbt.doc.ca
Communications Research Centre, Ottawa, Canada.
Try "gopher debra.dgbt.doc.ca"
Wes Morgan
>
>I think the ethical boundary is drawn (in terms of current regulations,
>certainly, in the US) between doing research using published texts --
>which is what, in my professional opinion, postings to newsgroups are --
>and research in which the investigators are eliciting or provoking
>responses for research purposes.
>
>Studying newsgroup postings is like studying letters to the editor
>published in newspappers, or studying published books of essays.
>The only problem arises from quotation -- and that's a copyright
>problem, not a research ethics problem.
Have you considered the possibility of posting messages for the sole
purpose of provoking/eliciting responses for study?
Usenet history is filled with incidents in which provocative posts
generated huge flamewars; in many cases, the author eventually claims
that the original posting was in fulfillment of some academic research
obligation. (For some reason, it's usually explained away as "a term
paper for my psychology class.")
I see no problem with observing the ebb and flow of online discussions;
however, the idea of creating/initiating a discussion for study purposes
(without the knowledge of the participants) seems a bit unethical. Were
I to do such a thing at your university, it seems that it would fall under
your Human Subjects review procedures...
--Wes
--
Wes Morgan ----- University of Kentucky ----- mor...@engr.uky.edu
Mailing list for AT&T StarServer E/S admins - starserve...@engr.uky.edu
GAT d(--) -p+ c++(++++) !l u++ e+ m* s++/++ !n h* f* !g w++ t+(++) r x+
And here's to you, Vicki Robinson; Usenet loves you more than you will know.
r...@max.u.washington.edu
> <r...@max.u.washington.edu> wrote:
>>
>>
>>Studying newsgroup postings is like studying letters to the editor
>>published in newspappers, or studying published books of essays.
>>The only problem arises from quotation -- and that's a copyright
>>problem, not a research ethics problem.
>
> Have you considered the possibility of posting messages for the sole
> purpose of provoking/eliciting responses for study?
>
> Usenet history is filled with incidents in which provocative posts
> generated huge flamewars; in many cases, the author eventually claims
> that the original posting was in fulfillment of some academic research
> obligation. (For some reason, it's usually explained away as "a term
> paper for my psychology class.")
>
> I see no problem with observing the ebb and flow of online discussions;
> however, the idea of creating/initiating a discussion for study purposes
> (without the knowledge of the participants) seems a bit unethical. Were
> I to do such a thing at your university, it seems that it would fall under
> your Human Subjects review procedures...
Yes, exactly. Observing is one thing; interventions are another. Even
a questionnaire is an intervention, designed to elicit behavior rather than
simply observe what people are doing in public.
A provocative intervention (e.g., proposing a controversial newsgroup
to see what would happen) would require institutional-level review --
and might well take some doing if it were to be approvable.
Richard Brzustowicz Jr
r...@max.u.washington.edu
Matthew B. Landry
r...@max.u.washington.edu writes:
> Yes, exactly. Observing is one thing; interventions are another. Even
> a questionnaire is an intervention, designed to elicit behavior rather than
> simply observe what people are doing in public.
>
> A provocative intervention (e.g., proposing a controversial newsgroup
> to see what would happen) would require institutional-level review --
> and might well take some doing if it were to be approvable.
Ah, but there's an obvious difference. If you post a questionnaire,
it's pretty obvious what it will be used for, even if you don't specify
in detail. If you propose a new newsgroup, you are intervening in the
process secretly, and messing around with other people's lives.
People who answer a questionnaire generally do so with the knowledge
that it is a research project, and that it will have no impact on them
directly. People who argue over a newsgroup usually don't assume
anything of the sort.
--
Matthew B. Landry
President of Project SAVE
m...@ml7694a.leonard.american.edu