Re: [Air-L] a question about privacy protection and copyright in Internet research
Hey, all, I'm late in adding a couple of points about this discussion. From a US-human subjects perspectives (talking more regulatory), I disagree with Jeremy's hard line "if you are dealing with texts, you are not dealing with research subjects." I disagree with it from disciplinary and philosophical perspectives, but, that would take us down the road of another discussion. But--from a US regulatory perspective,(f) Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains (1) Data through intervention or interaction with the individual, or (2) Identifiable private information. So, I would argue that to patently reject "texts" as subjects is a risky choice and not a line that holds a lot of ethical weight; disciplinary specificity is very important for this discussion--these decisions are not, should no,t be based on only one set of criteria; with online research, as we know, there are community (or venue) norms, ethics, practices; disciplinary norms, ethics, practices; extant research ethics guidelines at the federal/country/commission level; and, increasingly, third party (eg, Facebook, etc etc) norms, "ethics", practices. All of those must be considered in making these decisions about research practices. Regarding the issue of public v private (which I agree with Michael is a fluid and moving dichotomy), again, from a regulatory perspective, there are boundaries: From 101 (b)(2) and 101(b)(4) in the CFR 45: Regarding online observation and exemptions for existing data, the issue of researcher "entry requirements" and "threshold of access" is significant. The OHRP standard is "what is readily ascertainable." If there are costs, restrictions, or other measures that prevent data from being "readily ascertainable, "the exemption does not apply and the data should be considered private. (This comes from a report in draft form, written with a colleague from the Office for Human Research Protections here in the US.If you are looking for more regulatory guidance, I can share this report). Recently, I gave a few talks with the lawyer who advises OHRP, and while they have issued no formal guidance on Internet research in particular, some interpretations have been presented. I can share our powerpoints on Internetresearchethics.org for those interested. Best regards, e. Elizabeth A. Buchanan, Ph.D. Endowed Chair in Ethics Director, Center for Applied Ethics University of Wisconsin-Stout PO Box 790 Menomonie, Wisconsin, 54751-0790 buchanane@uwstout.edu Principal Investigator, Internet Research Ethics Commons internetresearchethics.org (NSF Project # 0924604) elizabeth@internetresearchethics.org
I think i already mentioned the private information exception. But past that, as I said, instead of complicating it, i argue that we we need to simplify it. Texts are not people. Representations of people are not people. Whatever claims people have on those objects are not claims against their nature, they are claims to ownership or copyright. That is the line I think we have to take here. Things like documents are not research subjects. They are research objects, research quasi-objects, but they don't rise to the level of being a research subject no matter what exterior subjective claims people put on them. If they have private information embedded in them, that is still an object, the issue with private information is not that the object becomes a research subject, but that the claim of a subject overrides the interests of the researcher in that case. So my argument is that the private information claim is not that it transforms the document, but it creates a claim by one person upon another based on the claim of privacy, however... very few things once published rise to that claim of private, especially once they are published, for instance a person's name on a document published is not private, it isn't private in the newspaper, it isn't private on public documents, nor are other identifying characteristics, height, demographic features, even salary might be public. So what rises to the claim of private in a public document? Medical records are one... though even then medical records can be made public by any number of sources and once public, might not be made private again. so... this line... seems really to be far, quite far away. Treat a published source as a published source, it is not a research subject. On May 10, 2011, at 8:58 AM, Elizabeth Buchanan wrote:
Hey, all,
I'm late in adding a couple of points about this discussion. From a US-human subjects perspectives (talking more regulatory), I disagree with Jeremy's hard line "if you are dealing with texts, you are not dealing with research subjects." I disagree with it from disciplinary and philosophical perspectives, but, that would take us down the road of another discussion.
But--from a US regulatory perspective,(f) Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains
(1) Data through intervention or interaction with the individual, or (2) Identifiable private information.
So, I would argue that to patently reject "texts" as subjects is a risky choice and not a line that holds a lot of ethical weight; disciplinary specificity is very important for this discussion--these decisions are not, should no,t be based on only one set of criteria; with online research, as we know, there are community (or venue) norms, ethics, practices; disciplinary norms, ethics, practices; extant research ethics guidelines at the federal/country/commission level; and, increasingly, third party (eg, Facebook, etc etc) norms, "ethics", practices. All of those must be considered in making these decisions about research practices.
Regarding the issue of public v private (which I agree with Michael is a fluid and moving dichotomy), again, from a regulatory perspective, there are boundaries:
From 101 (b)(2) and 101(b)(4) in the CFR 45:
Regarding online observation and exemptions for existing data, the issue of researcher "entry requirements" and "threshold of access" is significant. The OHRP standard is "what is readily ascertainable." If there are costs, restrictions, or other measures that prevent data from being "readily ascertainable, "the exemption does not apply and the data should be considered private. (This comes from a report in draft form, written with a colleague from the Office for Human Research Protections here in the US.If you are looking for more regulatory guidance, I can share this report).
Recently, I gave a few talks with the lawyer who advises OHRP, and while they have issued no formal guidance on Internet research in particular, some interpretations have been presented. I can share our powerpoints on Internetresearchethics.org for those interested.
Best regards,
e.
Elizabeth A. Buchanan, Ph.D. Endowed Chair in Ethics Director, Center for Applied Ethics University of Wisconsin-Stout PO Box 790 Menomonie, Wisconsin, 54751-0790
buchanane@uwstout.edu
Principal Investigator, Internet Research Ethics Commons internetresearchethics.org (NSF Project # 0924604)
elizabeth@internetresearchethics.org
_______________________________________________ The Air-L@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/
Jeremy Hunsinger Center for Digital Discourse and Culture Virginia Tech Computers are useless. They can only give you answers. --Pablo Picasso
I find this logic troubling, Jeremy. In our earlier exchange about Facebook suddenly making something public that was originally shared to a limited group, you seem to be placing strong preference on what a platform might afford at a given time (does it make something publicly available) rather than what the subject expected/intended/currently desires. That's not how I wish to have research ethics determined. And below you again now that "Texts are not people. Representations of people are not people", and appear to state that unless you are directly engaging with a human, you're not engaging in human subject research. Elizabeth has replied on this, but just to add: your logic seems to indicate that I can do research using a database of personally-identifiable and sensitive information, publish my results, etc without ever having to worry about ethics related to "human subjects" since I'm merely using a database, a "representation of people". Is this your position? -mz -- Michael Zimmer, PhD Assistant Professor, School of Information Studies Co-Director, Center for Information Policy Research University of Wisconsin-Milwaukee e: zimmerm@uwm.edu w: www.michaelzimmer.org On May 10, 2011, at 8:33 AM, jeremy hunsinger wrote:
I think i already mentioned the private information exception. But past that, as I said, instead of complicating it, i argue that we we need to simplify it. Texts are not people. Representations of people are not people. Whatever claims people have on those objects are not claims against their nature, they are claims to ownership or copyright. That is the line I think we have to take here. Things like documents are not research subjects. They are research objects, research quasi-objects, but they don't rise to the level of being a research subject no matter what exterior subjective claims people put on them. If they have private information embedded in them, that is still an object, the issue with private information is not that the object becomes a research subject, but that the claim of a subject overrides the interests of the researcher in that case. So my argument is that the private information claim is not that it transforms the document, but it creates a clai m by one person upon another based on the claim of privacy, however... very few things once published rise to that claim of private, especially once they are published, for instance a person's name on a document published is not private, it isn't private in the newspaper, it isn't private on public documents, nor are other identifying characteristics, height, demographic features, even salary might be public. So what rises to the claim of private in a public document? Medical records are one... though even then medical records can be made public by any number of sources and once public, might not be made private again. so... this line... seems really to be far, quite far away. Treat a published source as a published source, it is not a research subject.
On May 10, 2011, at 8:58 AM, Elizabeth Buchanan wrote:
Hey, all,
I'm late in adding a couple of points about this discussion. From a US-human subjects perspectives (talking more regulatory), I disagree with Jeremy's hard line "if you are dealing with texts, you are not dealing with research subjects." I disagree with it from disciplinary and philosophical perspectives, but, that would take us down the road of another discussion.
But--from a US regulatory perspective,(f) Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains
(1) Data through intervention or interaction with the individual, or (2) Identifiable private information.
So, I would argue that to patently reject "texts" as subjects is a risky choice and not a line that holds a lot of ethical weight; disciplinary specificity is very important for this discussion--these decisions are not, should no,t be based on only one set of criteria; with online research, as we know, there are community (or venue) norms, ethics, practices; disciplinary norms, ethics, practices; extant research ethics guidelines at the federal/country/commission level; and, increasingly, third party (eg, Facebook, etc etc) norms, "ethics", practices. All of those must be considered in making these decisions about research practices.
Regarding the issue of public v private (which I agree with Michael is a fluid and moving dichotomy), again, from a regulatory perspective, there are boundaries:
From 101 (b)(2) and 101(b)(4) in the CFR 45:
Regarding online observation and exemptions for existing data, the issue of researcher "entry requirements" and "threshold of access" is significant. The OHRP standard is "what is readily ascertainable." If there are costs, restrictions, or other measures that prevent data from being "readily ascertainable, "the exemption does not apply and the data should be considered private. (This comes from a report in draft form, written with a colleague from the Office for Human Research Protections here in the US.If you are looking for more regulatory guidance, I can share this report).
Recently, I gave a few talks with the lawyer who advises OHRP, and while they have issued no formal guidance on Internet research in particular, some interpretations have been presented. I can share our powerpoints on Internetresearchethics.org for those interested.
Best regards,
e.
Elizabeth A. Buchanan, Ph.D. Endowed Chair in Ethics Director, Center for Applied Ethics University of Wisconsin-Stout PO Box 790 Menomonie, Wisconsin, 54751-0790
buchanane@uwstout.edu
Principal Investigator, Internet Research Ethics Commons internetresearchethics.org (NSF Project # 0924604)
elizabeth@internetresearchethics.org
I find this logic troubling, Jeremy. In our earlier exchange about Facebook suddenly making something public that was originally shared to a limited group, you seem to be placing strong preference on what a platform might afford at a given time (does it make something publicly available) rather than what the subject expected/intended/currently desires. That's not how I wish to have research ethics determined. And below you again now that "Texts are not people. Representations of people are not people", and appear to state that unless you are directly engaging with a human, you're not engaging in human subject research. Elizabeth has replied on this, but just to add: your logic seems to indicate that I can do research using a database of personally-identifiable and sensitive information, publish my results, etc without ever having to worry about ethics related to "human subjects" since I'm merely using a database, a "representation of people". Is this your position? -mz -- Michael Zimmer, PhD Assistant Professor, School of Information Studies Co-Director, Center for Information Policy Research University of Wisconsin-Milwaukee e: zimmerm@uwm.edu w: www.michaelzimmer.org On May 10, 2011, at 8:33 AM, jeremy hunsinger wrote:
I think i already mentioned the private information exception. But past that, as I said, instead of complicating it, i argue that we we need to simplify it. Texts are not people. Representations of people are not people. Whatever claims people have on those objects are not claims against their nature, they are claims to ownership or copyright. That is the line I think we have to take here. Things like documents are not research subjects. They are research objects, research quasi-objects, but they don't rise to the level of being a research subject no matter what exterior subjective claims people put on them. If they have private information embedded in them, that is still an object, the issue with private information is not that the object becomes a research subject, but that the claim of a subject overrides the interests of the researcher in that case. So my argument is that the private information claim is not that it transforms the document, but it creates a clai m by one person upon another based on the claim of privacy, however... very few things once published rise to that claim of private, especially once they are published, for instance a person's name on a document published is not private, it isn't private in the newspaper, it isn't private on public documents, nor are other identifying characteristics, height, demographic features, even salary might be public. So what rises to the claim of private in a public document? Medical records are one... though even then medical records can be made public by any number of sources and once public, might not be made private again. so... this line... seems really to be far, quite far away. Treat a published source as a published source, it is not a research subject.
On May 10, 2011, at 8:58 AM, Elizabeth Buchanan wrote:
Hey, all,
I'm late in adding a couple of points about this discussion. From a US-human subjects perspectives (talking more regulatory), I disagree with Jeremy's hard line "if you are dealing with texts, you are not dealing with research subjects." I disagree with it from disciplinary and philosophical perspectives, but, that would take us down the road of another discussion.
But--from a US regulatory perspective,(f) Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains
(1) Data through intervention or interaction with the individual, or (2) Identifiable private information.
So, I would argue that to patently reject "texts" as subjects is a risky choice and not a line that holds a lot of ethical weight; disciplinary specificity is very important for this discussion--these decisions are not, should no,t be based on only one set of criteria; with online research, as we know, there are community (or venue) norms, ethics, practices; disciplinary norms, ethics, practices; extant research ethics guidelines at the federal/country/commission level; and, increasingly, third party (eg, Facebook, etc etc) norms, "ethics", practices. All of those must be considered in making these decisions about research practices.
Regarding the issue of public v private (which I agree with Michael is a fluid and moving dichotomy), again, from a regulatory perspective, there are boundaries:
From 101 (b)(2) and 101(b)(4) in the CFR 45:
Regarding online observation and exemptions for existing data, the issue of researcher "entry requirements" and "threshold of access" is significant. The OHRP standard is "what is readily ascertainable." If there are costs, restrictions, or other measures that prevent data from being "readily ascertainable, "the exemption does not apply and the data should be considered private. (This comes from a report in draft form, written with a colleague from the Office for Human Research Protections here in the US.If you are looking for more regulatory guidance, I can share this report).
Recently, I gave a few talks with the lawyer who advises OHRP, and while they have issued no formal guidance on Internet research in particular, some interpretations have been presented. I can share our powerpoints on Internetresearchethics.org for those interested.
Best regards,
e.
Elizabeth A. Buchanan, Ph.D. Endowed Chair in Ethics Director, Center for Applied Ethics University of Wisconsin-Stout PO Box 790 Menomonie, Wisconsin, 54751-0790
buchanane@uwstout.edu
Principal Investigator, Internet Research Ethics Commons internetresearchethics.org (NSF Project # 0924604)
elizabeth@internetresearchethics.org
_______________________________________________ The Air-L@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/
Jeremy Hunsinger Center for Digital Discourse and Culture Virginia Tech
Computers are useless. They can only give you answers. --Pablo Picasso
_______________________________________________ The Air-L@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/
participants (3)
-
Elizabeth Buchanan -
jeremy hunsinger -
Michael Zimmer