Print

Print


*****  To join INSNA, visit http://www.insna.org  *****

Balazs,

One possibility would be to, first, conduct the subgroup analysis on the 
collapsed data, and, second, calculate the degree to which those 
subgroups are realized at each point in time. That would distinguish 
your two scenarios. An alternative second step might be to graph the 
number of consecutive observations against the part of the collapsed 
structure (maybe calculated as proportion of ties or proportion of 
cohesive subgroups) cumulatively realized.

The question of tie concurrency appears important here. The paper I 
cited obviously deals with it, and cites other relevant work (by Morris 
and Moody).

David

Balazs Vedres wrote:
> *****  To join INSNA, visit http://www.insna.org  *****
>
>  David,
>
> Your question about what cohesion means is interesting, because it leads to
> what time means. Assuming that the points in time are sparse enough to see
> changes in group composition, I think analyzing the time-collapsed dataset
> is an interesting idea. 
>
> Although in that case two different processs would look the same: 
>
> 1. one set of nodes are connected only as dyadic components, at different
> points in time, that looks as a cohesive group in the collapsed data
>
> 2. another set of nodes are connected into a cohesive group at one point in
> time (at the same point in time).
>
> Thanks!
>
> Balazs
>
>   
>> -----Original Message-----
>> From: Social Networks Discussion Forum 
>> [mailto:[log in to unmask]] On Behalf Of David Gibson
>> Sent: Tuesday, December 20, 2005 8:57 AM
>> To: [log in to unmask]
>> Subject: Re: cohesion in a dynamic network
>>
>> *****  To join INSNA, visit http://www.insna.org  *****
>>
>> Balazs,
>>
>> First you have to ask what cohesion means in dynamic data. 
>> Your formulation "common history of cohesion" would probably 
>> lead most of us to simply sum instances of dyadic contact (I 
>> take your data to be in some way episodic) to create a 
>> time-collapsed snapshot, on which standard subgroup analyses 
>> can be performed. The interesting complication is that a 
>> group can be "actually" cohesive and yet not act that way 
>> over shorter or longer periods, if you permit a distinction 
>> between latent or propensity ties and enacted ties and 
>> recognize that the latter are subject to constraints such as 
>> those considered in
>>
>> Gibson, David R. 2005. "Concurrency and Commitment: Network 
>> Scheduling and its Consequences for Diffusion." Journal of 
>> Mathematical Sociology
>> 29:295-323
>>
>> David
>>
>> Balazs Vedres wrote:
>>     
>>> *****  To join INSNA, visit http://www.insna.org  *****
>>>
>>> Dear Socnetters,
>>>
>>> I wonder if anyone had experience with identifying cohesive 
>>>       
>> groups in 
>>     
>>> dynamic network data.
>>> How would you identify a group as a common history of 
>>>       
>> cohesion rather 
>>     
>>> than a cohesive subset in a network snapshot?
>>>
>>> Any comments are appreciated,
>>>
>>> Best
>>> Balazs
>>>
>>>
>>>       
>> _____________________________________________________________________
>>     
>>> 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.
>>>   
>>>       
>> -- 
>>
>> David Gibson
>>
>> Assistant Professor
>>
>> Department of Sociology
>>
>> University of Pennsylvania
>>
>> 3718 Locust Walk
>>
>> Philadelphia, PA 19104-6299
>>
>>  
>>
>> http://www.soc.upenn.edu/~gibsond/
>>
>> _____________________________________________________________________
>> 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.
>   

-- 

David Gibson

Assistant Professor

Department of Sociology

University of Pennsylvania

3718 Locust Walk

Philadelphia, PA 19104-6299

 

http://www.soc.upenn.edu/~gibsond/

_____________________________________________________________________
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.