***** To join INSNA, visit http://www.insna.org *****
I'm using Steve Borgatti's KeyPlayer v.2.0 program. I am comparing results that I've calculated by hand using equations from Borgatti's "The Key Player Problem" (the equations contained in this paper are the basis for his KeyPlayer program). In making this comparison, I'm getting inconsistent results that I'm hoping someone can help me understand.
Here's the situation:
I have 201 nodes in my matrix, with a total of 474 "1s" in the matrix, and the rest are "0s". This means there are 237 connections (474 divided by 2, since the 474 is for the FULL matrix). When I plug this information into Equation 2 of the KPP paper, I get a Fragmentation statistic of .988, or 98.8%.
Now when I import this matrix into KeyPlayer and run Disrupt, Maximize Fragmentation (the equivalent of Equation 2), I get a Fragmentation Statistic of 84.6%.
Obviously something is wrong, since taking out key players should INCREASE, not decrease the fragmentation statistic (since closer to 1 is better).
If anyone can shed some light on this curious result, it would be much appreciated! Thanks!
Daniel Schwartz, Ph.D.
Analyst, Government of Ontario
[log in to unmask]
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.