***** To join INSNA, visit http://www.insna.org ***** I would like to express my thanks to Dr. Baker for taking the time to investigate this so thoroughly. I have nothing but the deepest respect for him, and I take no small measure of inspiration from his diligent scholarship. Thanks again, Dave Wayne Baker wrote: > ***** To join INSNA, visit http://www.insna.org ***** > > Colleagues, > Recently, a message was posted on SOCNET concerning "The Social > Organization of Conspiracy" (Baker and Faulkner 1993). Subsequently, > the sender of this post contacted me via email. I sent a response to > him, but since he raised concerns in this public forum, I thought it > appropriate to post my response. The relevant portions are reproduced > below. > Best wishes, > Wayne Baker > > > The confusion is caused by a typo in the article, and a little lack of > clarity on our part. Otherwise, our measures, analysis, findings, and > interpretation are correct. Specifically, here are the answers to the > issues you raised: > > (1) Our definition of closeness (farness) point centrality, Eq. 3, p. > 848. There is a typo. The superscript -1 is missing from the equation, > which, unfortunately, we didn't spot in proof reading. This is a > measure of "point decentrality" (Freeman 1979:225), or, as we put it, > farness. > > (2) Comparisons of point centrality across networks of different sizes. > It make sense to standardize measures of point centrality in most > cases--but not all (Freeman 1979). In our context, degree centrality > has a very intuitive meaning: the number of direct eyewitnesses of > ego's participation in a price-fixing conspiracy. In a legal setting, > the number of direct eyewitnesses (degree centrality) makes more sense > than the number of direct eyewitnesses relative to the total number of > participants (relative degree centrality). A direct eyewitness is a > direct eyewitness, whether the network has 21 participants > (transformers), 24 (turbines), or 33 (switchgear). (Indeed, it would be > hard to say how a grand jury would interpret relative degree > centrality.) By extension, the same logic applies to farness and > betweenness. Farness, for example, measures paths (geodesics) of direct > and indirect eyewitnesses. So, in our particular context, it makes > sense to *not* standardize point centrality measures. > > (3) For the measure of graph centralization, we used the standardized > closeness index, according to Eq. 4, p. 850. We did not use farness. > We apparently caused confusion by referring to farness when discussing > graph centralization (e.g., footnote 8). We did so to remind the reader > that our measure of "closeness" point centrality was actually a measure > of distance. In this context, it seemed especially important to do so. > In retrospect, we could have been clearer about this in our article, > making sure there was no ambiguity about our measures of point > centrality and graph centralization. > > _____________________________________________________________________ > 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.