Print

Print


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


With missing data, you have three choices...
1) find out what is missing, and figure out a new way to get it (or get proxy data)
2) choose different network metrics, ones not as sensitive to missing data
3) infer missing links

I usually go in the above order and only infer as a last resort.  I am usually working with bounded networks inside task organizations -- which makes the problem easier.  If I do infer, I check the inferences with people who have “local knowledge/” of the organization/people.  Algorithms sometimes find "logical links" that are completely laughable to the locals!

Valdis Krebs
http://orgnet.com
http://thenetworkthinkers.com



On Apr 17, 2014, at 6:03 PM, Peter Carrington wrote:

> ***** To join INSNA, visit http://www.insna.org *****
> ***** To join INSNA, visit http://www.insna.org ***** 
> Hi.  I’m looking for articles on how to handle missing data in network research.  How should network measures be modified?  What good ways are there of imputing values for the missing data?  Thanks in advance. 
>   
>   
> Phillip Bonacich 
> Professor Emeritus of Sociology 
> University of California, Los Angeles 
> ...................................................................
> 
> some more:
> 
> Gill, Jeff and John R. Freeman (2013). Dynamic elicited priors for updating covert networks. Network Science, 1, pp 6894 doi:10.1017/nws.2012.6
> 
> Handcock, M. S., & Gile, K. J. (2010). Modeling social networks from sampled data. Annals of Applied Statistics, 4, 5–25.
> 
> Koskinen, Johan H., Garry L. Robins, and Philippa E. Pattison. "Analysing exponential random graph (p-star) models with missing data using Bayesian data augmentation." Statistical Methodology 7.3 (2010): 366-384.
> 
> Koskinen, Johan, G. Robins, and P. Pattison. "Missing data in social networks: Problems and prospects for model-based inference." MelNet Social Networks Laboratory Technical Report (2011): 08-03.
> 
> Kossinets, G. (2006). Effects of missing data in social networks. Social Networks, 28, 247–268.
> Rhodes, C. J., and P. Jones. "Inferring missing links in partially observed social networks." Journal of the operational research society 60.10 (2009): 1373-1383. 
> 
> Stork, Diana, and William D. Richards. "Nonrespondents in Communication Network Studies Problems and Possibilities." Group & Organization Management 17.2 (1992): 193-209.
> 
> Peter Carrington
> 
> 
> 
> -------- Original Message --------
> Subject:	Re: miissing data in networks
> Date:	Thu, 17 Apr 2014 23:31:05 +0200
> From:	Anja Znidarsic <[log in to unmask]>
> Reply-To:	<[log in to unmask]>
> To:	<[log in to unmask]>
> 
> Dear Prof. Bonachic, 
> 
> Few papers on how to handle missinga data in social network research:
> Huisman, M. (2009): Effects of missing data in social networks. Journal of Social Structure, 10. Available at:http://www.cmu.edu/joss/content/articles/volume10/huisman.pdf.
> Huisman, M. and Steglich, C. (2008): Treatment of non-response in longitudinal network studies. Social networks, 30, 297-308.
> Žnidaršič, A, , Ferligoj A., and  Doreian P.Non-response in social networks: The impact of different non-response treatments on the stability of blockmodels .Social Networks 34 (4), 438-450
> Žnidaršič, A., Doreian, P., and, Ferligoj A. Absent Ties in Social Networks, their Treatments, and Blockmodeling Outcomes. Advances in Methodology & Statistics/Metodoloski zvezki 9 (2). Available at: www.stat-d.si/mz/mz9.1/znidarsic1.pdf
> 
> 
> Best, 
> Anja Znidarsic 
> Assistant Professor of Quantitative Methods 
> University of Maribor, Slovenia 
> 
> 
> 
> From:        Phillip Bonacich <[log in to unmask]> 
> To:        [log in to unmask] 
> Date:        17.04.2014 23:14 
> Subject:        [SOCNET] miissing data in networks 
> Sent by:        Social Networks Discussion Forum <[log in to unmask]> 
> 
> -- 
> Professor Peter J. Carrington
> Editor, Canadian Journal of Criminology and Criminal Justice
> Department of Sociology and Legal Studies
> University of Waterloo
> Waterloo, Ontario, Canada N2L 3G1
> email: [log in to unmask]
> phone: +1 519 888 4567 ext. 33961
> fax:   +1 519 743 1126
> URL:   http://sociology.uwaterloo.ca/peter_carrington.html
> CJCCJ home pages:
> http://www.ccja-acjp.ca/en/cjc.html and
> http://www.utpjournals.com/Canadian-Journal-of-Criminology-and-Criminal-Justice.html
> _____________________________________________________________________ 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.