#SNAc week 1: what are networks and what use is it to study them?
I’m finally getting to grips with social network analysis via the #SNAc MOOC. See My first MOOC for thoughts about the course environment (and Coursera) in general – posts here are my notes on the content.
Week 1 consisted of an overview of concepts and an introduction to software tools. I’m generally not a video fan, but these were the exception that proves the rule. Watched them once, then scanned the slides and script afterwards for the detail. Crucially, you can speed them up (something I often want to do!) or, more useful in this case, slow them down for crucial bits. Definitely still a question of blindly following instructions rather than knowing how to do things, but hopefully that will come with time.
— Alessandro Zonin (@AlessandroZonin) September 24, 2012
#SNAc using Gephi for visualisation and basic network metrics, plus NetLogo for modelling network dynamics. Programming types will also be using iGraph.
Not using Pajek (free, extensive functionality via drop-downs), UCINet (costs) or NodeXL, all of which are Windows based. Or NetworkX (open source, uses Python). Specialised tools not using include the SNA package for R (statistics heavy) and SoNIA (Social Network Image Animator, good for longitudinal analysis. From Twitter: posts on Python for SNA and R for SNA.
The recommended readings look interesting, but it’s the curse of the netbook again – there’s no way I’m going to read a 20 page PDF on a screen. Some highlighted resources from Twitter and the forum look a bit more possible:
- Wikipedia | brief intro | glossary | glossary2 | 116 page book
- How to conduct a social network analysis (slides)
- Interactive visualisation of influence in programming languages
- Mathematical overview of small world models and experiments
- Moviegalaxies - social graphs in movies, eg Pulp Fiction, Fargo(of course) - mouse over nodes to find a character’s degree and betweenness
- Network Science - Barabási’s book, with slides and other resources
- Visually detecting communities (NetLogo)
- Why your friends have more friends than you do
- Analysis of the Faculty of Psychology in Seville – from participant (Spanish)
Any two people in the Facebook graph are connected with an average of 4.7 hops.
My Facebook network is small and, although I had to refer back to my notes for most of the tasks, I got 10/10. Hurra! I might give it another go with my partner’s rather larger network.
Lots of people have tweeted their networks, and some people have also tweeted LinkedIn maps. On 26 September an email was sent out re collaboration and Coursera’s honor code – not quite sure what that’s about, but some people have been oversharing their short answers and/or code, apparently. Tsk tsk.
Subscribe to comments with RSS.