***** To join INSNA, visit http://www.insna.org *****
I certainly agree that it typically makes sense to log distances.
But, I'm curious about your rationale for using log10 rather than loge
(natural log), which seems more common. I don't think I've ever seen
a comparison or an argument about which transformation is more
appropriate when predicting interaction as a function of distance, but
perhaps this work is out there.
Zachary Neal, Assistant Professor
Michigan State University
Department of Sociology & Global Urban Studies Program
316 Berkey Hall
East Lansing, MI 48824
Quoting Barry Wellman <[log in to unmask]>:
> ***** To join INSNA, visit http://www.insna.org *****
> As the subject of distance networks has come up, let me take the
> opportunity to suggest that the distances be logged for regression
> types of analysis. (I use log10). That is because the increment of a
> unit at greater distances isn't as important as at closer distances.
> That is, a unit increase to 2 km from 1 is much more important for
> contact, collaboration and support than a unit increase to 2001 im
> from 2000.
> The proof is in the pudding and the writing. Ever since the
> Community Question paper (1979), the regressions are mor powerful
> when logged. This includes studies of internet communication and
> networked organizations.
> Nevertheless, YMMV (literally LOL)
> Barry Wellman
> S.D. Clark Professor of Sociology, FRSC NetLab Director
> Department of Sociology 725 Spadina Avenue, Room 388
> University of Toronto Toronto Canada M5S 2J4 twitter:barrywellman
> http://www.chass.utoronto.ca/~wellman fax:+1-416-978-3963
> Updating history: http://chass.utoronto.ca/oldnew/cybertimes.php
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). 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.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.