Print

Print


*****  To join INSNA, visit http://www.sfu.ca/~insna/  *****

In most cases, you can replace real names with numbers/ids.  But sometimes
(as was pointed out re contact tracing and other types of organizational
research) it is essential to know the real identities to make sense of the
graph/connections.

In this case, you could invent codes (e.g., Colombo) to represent the real
identities; a locked and/or encrypted hard-copy can be used to jog your
memory if you forget who "Colombo" really is.

There are other variations in types of codes to use, depending on the type
of network you are analyzing (whether egocentric, sociocentric), your
knowledge about attributes/tastes of network members, the context for the
research, level of security required for relations underconsideration
(who-has-sex-with-whom relation will usually require much more safeguard
than who-socializes-with-whom, though in terrorist networks, socialization
could equally be a sensitive measure), etc.

Hope this helps.

/Em


On Mon, 19 Jan 2004 12:16:57 -0400, Richard Rothenberg <[log in to unmask]>
wrote:

> *****  To join INSNA, visit http://www.sfu.ca/~insna/  *****
>
> Steve/Glen:
>
> One circumstances in which names on the graph have been important to us
> is in the ongoing activity of STD, HIV or TB control using contact
> tracing.  Sequential graphs allow the investigator to know who is
> involved, where they are in the process, and possibly what to do next.
> Such investigations use public health acquired data, usually not subject
> to IRB scrutiny, but are increasingly so because use of network methods
> differs from classical or traditional contact tracing procedures.  It's
> fuzzy at the moment, since network ideas are gradually being adopted by
> programs.  Be that as it may, in most other circumstances, names aren't
> needed, and an short ID usually produces a nicer looking graph.
>
> I'm not sure that IRBs are devoting special scrutiny to networks (though
> everyone on this list probably thinks they are).  I think it's part of a
> much more generalized phenomenon.  IRBs are increasingly identification
> averse.  The guiding principle is that if something can go wrong, it
> must be defended against, even if improbable.  The IRB parlor game is to
> discover a pathway to malfeasance, however tortuous, and insist that the
> PI erect a barrier (often at the expense of the study).   The process
> has intensified with the advent of HIPAA, and certainly affects network
> research in a major way, because of the perceived underlying threat to
> the autonomy of named network partners.  If you are willing to believe
> (as I do, but apparently many IRBs do not) that the vast majority of
> investigator are ethical and a minority are not, than the obvious
> illogic of such an approach is evident (...taken your shoes off at the
> airport lately?).  But it seems to be Inescapable Illogic that powers
> much of public policy these days.  For many public decisions, the
> underlying rationale is that the probability may be miniscule, but the
> payoff can be catastrophic.  It isn't clear that such a situation
> obtains in research    But this is where we seem to be heading, so avoid
> names.
>
> Rich Rothenberg
>
> Steve Borgatti wrote:
>
>> *****  To join INSNA, visit http://www.sfu.ca/~insna/  *****
>>
>> Glen, is there some reason why you need to enter names into the network
>> data
>> files? Why not use id codes in the ucinet/netdraw files?
>>
>> steve.
>>
>> ----- Original Message -----
>> From: "Glen Murphy" <[log in to unmask]>
>> To: <[log in to unmask]>
>> Sent: Wednesday, January 14, 2004 7:41 AM
>> Subject: Entry of data - common procedure
>>
>>
>>
>>
>>> *****  To join INSNA, visit http://www.sfu.ca/~insna/  *****
>>>
>>> Hi everyone,
>>>
>>>
>>>
>>> Happy New Year and all that !
>>>
>>>
>>>
>>> I've finally begun collecting network data for the PhD (supervisors
>>> very
>>> relieved) and have spent the last couple of days explaining to
>>> participants the extraordinary lengths I'm going to to protect their
>>> identity (separate code books, unique code identifiers and numbers on
>>> the questionnaire etc.).  This is all wonderful and everyone's happy
>>> but
>>> as I sat down to key it in tonight I've realised that all of those safe
>>> guards really protect against unauthorised access to the hardcopy
>>> interview records.  If (as I had unthinkingly planned to do) I key in
>>> participant names into a matrix (using UCINET) then theoretically that
>>> data file is highly vulnerable.
>>>
>>>
>>>
>>> I'm aware that NETDRAW has some nifty little features allowing you to
>>> change labels & attributes - all good for post hoc diagrams, but it
>>> does
>>> leave a problem with UCINET reports and the raw data file itself.  What
>>> is the usual procedure ? How do people deal with this issue ?  I'm
>>> aware
>>> that US IRB boards are a little touchy regarding network analysis - has
>>> anyone had any stipulation from their IRB about this issue ?
>>>
>>>
>>>
>>> So far no-one I've dealt with has twigged to this problem, but I
>>> thought
>>> I'd be proactive just in case.
>>>
>>>
>>>
>>> Interested to hear thoughts on this.
>>>
>>>
>>>
>>> Cheers, Glen.
>>>
>>>
>>>
>>>
>>>
>>> Glen Murphy
>>>
>>> PhD Candidate
>>>
>>> School of Management
>>>
>>> Queensland University of Technology
>>>
>>> 126 Margaret St
>>>
>>> Brisbane 4001
>>>
>>>
>>>
>>> Ph.     3314 8061
>>>
>>> mob.    0403 001 623
>>>
>>> _____________________________________________________________________
>>> SOCNET is a service of INSNA, the professional association for social
>>> network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
>>> an email message to [log in to unmask] containing the line
>>> UNSUBSCRIBE SOCNET in the body of the message.
>>>
>>>
>>>
>>
>> _____________________________________________________________________
>> SOCNET is a service of INSNA, the professional association for social
>> network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
>> an email message to [log in to unmask] containing the line
>> UNSUBSCRIBE SOCNET in the body of the message.
>>
>>
>>
>
> --
> Richard Rothenberg, MD
> Professor
> Department of Family and Preventive Medicine
> Emory University School of Medicine
> 69 Jesse Hill Jr Drive, SE
> Atlanta, GA 30303
> Tel: 404-616-5606
> Fax: 404-616-6847
> Pager: 404-866-8575
> Email: [log in to unmask]
>
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.



--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.