Print

Print


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

Reposted to Stocnet discussion list as suggested.


On Thu, 24 Apr 2008 17:00:04 +0200
  "T.A.B.SNIJDERS" <[log in to unmask]> wrote:
> Dear Shan,
> 
> Your query might be more appropriate for the Stocnet 
>discussion list
> which is specialized in issues concerning the programs 
>included in
> Stocnet, among which Siena. See 
>http://groups.yahoo.com/groups/stocnet/
> 
> But let me try to answer it here, as far as I can.
> 
> 1. I do not understand your problem in running Siena and 
>have not heard
> of such problems earlier. The van de Bunt data supplied 
>with Siena, for
> example, have 7 waves, and these can be used in Siena 
>without any
> problem. When you follow exactly the steps in the 
>Stocnet manual
> (assuming that you run Siena from within Stocnet) and 
>the Siena manual,
> it should work.
> 
> 2. Investigating groups of 6 actors with 10 attributes 
>strikes me as
> being a bit risky - too many variables for so few 
>actors.
> 
> 3. Too analyze multiple groups you may need to use 
>adapted specifications.
> 
> 4. What you write about structure still seems too vague. 
>It may be a
> good idea to develop your thoughts further on this 
>aspect.
> 
> Best regards,
> 
> Tom Snijders
> 
> 
> Shan Cook wrote:
>> ---------------------- Information from the mail header 
>>-----------------------
>> Sender:       Social Networks Discussion Forum 
>><[log in to unmask]>
>> Poster:       Shan Cook <[log in to unmask]>
>> Subject:      Waves in siena problem & Idea request
>> -------------------------------------------------------------------------------
>> 
>> *****  To join INSNA, visit http://www.insna.org  *****
>> 
>> Question 1 of 2)
>> I have been trying to run Siena but have encountered a 
>> problem:
>> More than 3 waves causes in the program freezing/failing 
>> to respond etc.
>> And no, this is not just that the program needs more 
>>time 
>> to process the data - it is an actual freeze that 
>>requires 
>> a restart of the program.
>> 
>> I have tried:
>>   using the supplied data.
>> 
>> using no attributes (actor waves only).
>> 
>> removing all model options (density etc) so that the 
>> program has few functions to perform.
>> 
>> using a different PC.
>> 
>> uninstalling the program and reinstalling a new 
>>download.
>> 
>> 
>> The data I wish to analyse: 12 groups, mean 6 actors
>> in each group and mean 6 waves for each group, 10 
>> attributes.
>> My exploration: if a group achieve a structure by wave X 
>> (density,centralisation, etc)
>> and they have attributes >=Y (motivation, social 
>> familiarity, etc)
>> then their performance (group and individual) is more 
>> likely to be >=Z
>> 
>> I lack the necessary knowledge in IT, mathematics and 
>>SNA 
>> to understand the problem.
>> Is it simply that STOcNET can only run 3 waves at any 
>>one 
>> time?
>> If there is not 'fix' is is possible to run multiples of 
>>3 
>> with overlaps? i.e. waves 123, then 345, then 567?
>> 
>> 
>> 
>> Question 2 of 2)
>>   I would appreciate thoughts on which 'structures' to 
>> examine. I have found a few papers that touch upon 
>>similar 
>> topics but if anyone has any ideas that they think would 
>> be worth trying out but haven't had the time to try, 
>>then 
>> I would be more than happy to accomodate.
>> 
>> 
>> Thank you for your thoughts
>> Ms Shan Cook
>> 
>> _____________________________________________________________________
>> 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.
> 
> -- 
> Tom A.B. Snijders
> Professor of Statistics in the Social Sciences
> University of Oxford
> 
> Professor of Statistics and Methodology
> Department of Sociology, University of Groningen
> 
> for addresses and telephone numbers: see
> http://stat.gamma.rug.nl/snijders/

_____________________________________________________________________
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.