***** To join INSNA, visit http://www.sfu.ca/~insna/ *****
If you look at the small world research, Milgram (20-30% will pass on a
high-priority message, depending on race and other demographics) Russ
Bernard and Nan Lin (to high status individuals), Keith David (11-20% will
pass on a message to 3.3-6.9 people depending on the topic) you can get the
empiricial research for transitivity. The precise model for the question is
Hn = L^n(Ho)
Ho = 1.0
L =1-a (fidelity is less than perfect by definition from Information Theory
a = dH/dT or the slope of the fidelity over time
therefore H--> 0.0, as n--> infinity
The precise values of course depend on a number of demographic and message
factors.
-----Original Message-----
From: Social Networks Discussion Forum [mailto:[log in to unmask]]On
Behalf Of Valdis
Sent: Thursday, October 31, 2002 11:35 AM
To: [log in to unmask]
Subject: Re: noise in IEN
***** To join INSNA, visit http://www.sfu.ca/~insna/ *****
Interesting question...
I remember discussing this in my Org Comm classes [Prof. Vince Farace]. For
human networks the numbers 50% through 80% were used -- for every
re-transmission of a message only 50-80% of the original content is
retained.
Of course one re-transmission may retain 90% while the next one may retain
only
20% depending on the particular node/link in the path.
For A-->B-->C-->D ...
at 80%: D only gets about 1/2 of the original content
at 50%: D only gets 1/8 of the original content!
And what about multiple paths from A to D that distort the original message
differently?
Makes 6 degree of separation quite a large world, eh?
Valdis
P.S. With computers/routers as nodes this is usually not a problem -- most
of
the transmission protocols have built-in error checking that makes sure what
is
being received is what was originally sent.
Marko Pahor wrote:
>
> ***** To join INSNA, visit http://www.sfu.ca/~insna/ *****
>
> Dear SocNet-ers,
>
> I have a question. I'm trying to think of a network of contacts that is
> not an actual information exchange network in terms of a such. But I'm
> afraid that I'll have a lot of noise. By noise I mean that when A gets
> information from B and B gets information from C (A and C are not
> connected), the information A gets from C (through B) will not be
> correct. Think for example of gossip. Someone cuts his finger and after
> a couple of steps the story will be that he cut off his hand - this is
> what I mean by noise.
>
> Does anyone know about some research on noise in the information
> exchange network? Maybe it's called differently, as I couldn't get
> anything under this search (I got some information exchange network
> concerning noise, noise emitters and stuff like that).
>
> Thank you for your help.
>
> Marko Pahor
>
> _____________________________________________________________________
> 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.
_____________________________________________________________________
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.
|
