Using Networks

links for 2009-10-16

• Maturity model for email communication in knowledge organizations: The case of police investigations (Gottschalk, 2008)
  "... a four stage maturity model for email communication in knowledge organizations is proposed. The stages are labeled person-to-technology, person-to-person, person-to-information and person-to application, respectively."
  (tags: email maturity organization knowledge cognitive social networks)
• A Relational Event Framework for Social Action (Butts, 208)
  Butts, Carter T. (2008). ``A Relational Event Framework for Social Action.'' Sociological Methodology, 38(1):155-200.
  (tags: event longitudinal dynamics)
• Closure, connectivity and degree distributions: Exponential random graph (p*) models for directed social networks (Robins et al, preprint 2008)
  "In this paper, we explain the theoretical and empirical advantages in generalizing to additional closure effects. We propose three new triadic-based parameters to represent different versions of triadic closure: cyclic effects; transitivity based on shared choices of partners; and transitivity based on shared popularity"
  (tags: transitivity cycles triad popularity closure ERGM_p* directed structural homophily equivalence sources sinks)
• Exponential Random Graph (p) Models for Affiliation Networks (Wang et al, preprint 2007)
  "The classic Southern Women data and Australian interlocking director data are used as examples to show that the ERGM with the newly specified statistics is a powerful tool for statistical analysis of affiliation networks."
  (tags: affiliation multimodal social network analysis ERGM_p*)
• Longitudinal Methods of Network Analysis (Snijders)
  "Pp. 5998-6013 in Encyclopedia of Complexity and System Science (editor-in-chief Bob Meyers), part of the Social Networks section (section editor John Scott), Springer Verlag, 2009."
  (tags: longitudinal social network analysis dynamic stochastic selection influence example)
• Dynamic Networks and Behavior: Separating Selection from Influence (Steglich & Snijders, submitted 2009)
  a very good summary - particularly useful for the tables of effects (with graphics) at the back
  (tags: stochastic dynamic social network analysis selection influence longitudinal)
• Applying SIENA: An Illustrative Analysis of the Coevolution of Adolescents' Friendship Networks, Taste in Music, and Alcohol Consumption (Steglich et al, 2006)
  "A family of dynamic actor-driven models for the coevolution process is sketched, and it is shown how to use the SIENA software for estimating these models. We illustrate the method by analyzing the coevolution of friendship networks, taste in music, and alcohol consumption of teenagers."
  (tags: examples social network analysis stochastic dynamic longitudinal selection information)
• Models for Longitudinal Network Data (Snijders, 2004)
  chapter from P. Carrington, J. Scott, & S. Wasserman (Eds.), Models and methods in social network analysis. New York: Cambridge University Press.
  (tags: social network analysis stochastic dynamic longitudinal)
• Statistical Methods for Network Dynamics (Snijders, ????)
  online paper - short, with examples
  (tags: stochastic dynamic social network analysis tips longitudinal)
• "Statistical Methods for Network Dynamics" (pdf)
  Tutorial slides
  (tags: stochastic network social analysis dynamic software tips selection influence longitudinal)
• "Introduction to stochastic actor-based models for network dynamics" (Sniders et al 2009)
  from special issue of SN
  (tags: stochastic social network analysis software tips selection influence dynamic longitudinal)

links for 2009-10-10

• The Derivatives of Centrality and their Applications in Visualizing Social Networks (Correa et al) (application/pdf Object)
  "derivatives not only provide an analysis tool for social networks, and also help us simplify the layout of complex networks in a way that retains the main structural properties. Centrality derivatives also help to visually measure the centralization degree of a network and provide the necessary information for estimating other metrics, such as structural balance and uncertainty"
  (tags: derivatives centrality degree)
• The stability of centrality measures when networks are sampled (Costenbader & Valente, 2003)
  This study uses bootstrap sampling procedures to determine how sampling affects the stability of 11 different network centrality measures. Results indicate that some measures are more stable than others, and that stability is also a function of network and study properties.
  (tags: centrality bootstrap robustness(math.))

links for 2009-10-09

• Chapter 9:  Ego networks
  (tags: structural holes social network analysis ego)
• The Cogito Knowledge Center
  (tags: social network analysis tool)
• Blue Spider Analytics, Inc - Home
  (tags: network analysis visualization tool)

links for 2009-10-08

• Calculating status with negative relations (Bonacich, 2006)
  (tags: negative status social network analysis)

SNA Tools and Formats diagram - updated

16 Nov Update: I've updated the diagram, and made it easier to read - the link now points to a pdf version. The same license applies as for the whole blog - Creative Commons Attribution-Share Alike v3.0. Do let me know, though - it's always interesting to find out how your work is used.

Some time ago, I was struggling to work out how I could evaluate a new social network analysis tool - one that made use of some obscure format. In order to give a new tool a fair test, I always use a specific group of datasets, that I know inside out. Almost inevitably, the new tool will require that they all be translated into some obscure format. But, given the tools already out there, there's always a chance that I can some combination of existing tools will do it for me. In the end, I drew a map to help me navigate this process.

Now, after many requests, I've revised the map. As before, if you have any more information, corrections, even objections, I will endeavour to respond - and quicker, this time. Mea culpa.

Salmon need artificial pigments

I love this so much. Bring on the language games.

A recent TV advertisement in the UK has stirred up the salmon farming industry. The advertisement shows a man in a lab describing how farmed salmon is coloured with the chemical Astaxanthin. The speaker then holds up on of the now famous "Salmon Colour Charts"

- (via a message found at EdwardTufte.com - in a discussion about Salmon colour charts)

And now, some of the text from a press release from the salmon industry. Look out for the language games -- they're not immediately obvious. No, really.

“Captain Birds Eye charts the wrong course in advert about farmed salmon” says Scottish Salmon Producers’ Organisation

“The Scottish salmon farming industry is very disappointed that Birds Eye has chosen to take this rather melodramatic and cynical approach to its advertising when the fact is that salmon, wild and farmed, are pink because they consume carotenoid pigments. In the wild, salmon take up this pigment from such food items as shrimp and other organisms. Our farmers simply provide the same elements in their feed and the same natural process in both wild and farmed fish means that the pigment is absorbed into the body.

[...]

Salmon cannot synthesize carotenoid pigments and so must find them in their diet, to assist the reproductive process and protect the eggs. Colour charts are used to ensure that the salmon are fed sufficient quantities of carotenoid pigments to meet their and the consumers’ needs.

- via the SSPO, but italics added.

I wonder my supermarket is using the same logic for its smoked haddock - do I need it to be pigmented with that acid yellow colour?

links for 2009-07-16

• ' Generalized blockmodeling ' (pdf)
  (tags: blockmodelling generalized)
• wePapers - Creating the world's biggest study group
  (tags: online paper community research learning document)
• The Human Terrain System: A CORDS for the 21st Century | The U.S. Army Professional Writing Collection
  Short artivcel summarising the history, purpose and nature of the US Human Terrain System
  (tags: military social anthropology culture)
• Michelle Paver's favourite books on archaeology and anthropology | Books | guardian.co.uk
  A good list, this, with some favourites, some new.
  (tags: archaeology anthropology books)

Email policy

I spend many working days away from the office, and many office hours in meetings. Like most people in large organizations, I receive hundreds of actionable emails a week. I can only access these while at work. And my working days are usually over-committed.

Consequently, I just cannot give every email equal attention. Email is, of course, a broken tool - and a working culture that expects a rapid, pavlovian response from the recipient is not one in which anyone can concentrate long enough to do high-quality work.

To manage this as best I can, I use a kind of 'triage' system, based on the book 'Getting Things Done': each email is automatically 'read' and categorized, and I re-prioritize and work through these lists and my voicemail a couple of times a day. Incidentally, this means that receiving a read receipt for an email does not mean that I've read it!

So, if I haven't responded to your email, I'm either not in, or I'm focussing on delivering work required urgently by another customer. Sorry, but there it is.
And if I've sent you a very curt email - say, four sentences long - I'm doing my bit to save you time (see http://four.sentenc.es/). Feel free to do the same to me.

-Mark Round

Categories and equivalences (redux)

To revisit a familiar argument: I find categorization to be a useful process, when engaged in exploratory analysis, if the categories themselves are up for grabs (and not just the assignments to those categories). For an example, the process of partitioning a network into sets of structurally-equivalent nodes -- those that share connections with the same others -- is an essential part of most network analyses. But the patterns of structural-equivalent sets are not always easy to interpret, as there isn't always a category label at hand with which I can make sense of the patterns found by the algorithm.

'Regular' equivalence causes me similar problems. Two actors are regularly equivalent if they have similar patterns of ties to equivalent others. A slightly abstruse definition, but an intuitive concept, given a proper exemple. Take a network describing communication behaviour within at least a medium-sized organization: you would expect to see certain patterns recurring within certain subsets - 'team leaders', 'technical managers', 'salesmen', etc. Some will match formal, appointed roles, but others will be categorising emergent niches in the structure. Many of these latter categorizations will not have clear given names. They may differ from equivalence-sets in networks from similar organizations, reflecting instead the peculiarities of the organization under study.  Finding these and naming them -- as a categorization process -- I find to be hugely useful in getting under the skin of an organization.

In the past I've likened the process to early biology. This may be slightly grandiose (and certainly not original - see Morgan, 2006), but it's almost as if each new organization is a new ecosystem, full of new species. We can see some convergent evolution going on, but we shouldn't be blinded by expecting it.

On categories (Marsay)

I had a comment on the post “Friends, acquaintances, and Facebook” from a colleague, Dave Marsay, that I thought worth its own post.

Mark: As a fan of Lewis Carroll, I not only wonder if the categories may not have changed and if (whatever) constrains the richness of human relationships, but if the notion of category as a means of viewing (whatever) has outlived its usefulness and constrains human relationships - maybe even in a 'bad' way. At the same time, tools like SNA are clearly essential. The problem, then (as with much else) is to develop theories of use that can provide the context for routine use. One argument against categories, adapted to this case, is a follows. Categories make sense in the context of stable systems, e.g. stable societies. But social networking is transforming social affairs. In math-speak, categories are the invariants of 'the system' - but 'the system' is changing. The use of the tools subverts their assumptions. Another (controversial) argument is that it would be inhuman to deal with our fellow humans as instances of categories, rather than as individuals. Arguably, then, any theory based on categories would be a part of the anti-social sciences. Of course, in many cultures reliance on categorisation is endemic, and until a few weeks ago it was commonly argued that the economic and military strength of such cultures proved the merit of this approach. If this consensus were to change it would radically affect 'who we would ask for a favour', and SNA, for example, would not be of much help in understanding what was going on. Social Networking tools, on the other hand, could be very useful, and could help drive any change. You will appreciate that I have difficulty in saying anything about the universality of human experience, beyond yes and no - and don't know. I'll maybe ponder. Regards.

I'll leave it to stand for a while, and comment in a post sometime in the next few days.