(Post copied from Danegeld blog, 4 Feb 2015.)
Update, July 2015: see Hazel Hall on DREaM Again (again), investigating the long term impact of the project. Splendid! May 2016: not much on #sna lately, apart from a snippet on R4’s Digital Human: Are you more likely to find what you’ve lost using online social networks? Are we as connected as we think we are? Or does it make more sense to step out of the digital world and search with the help of physical social networks? A larger network of weaker/looser ties is more effective in finding something lost – these ties have information you don’t have. Other factors also come into play, eg how navigable is the network? The same processes go on IRL, with the Lost and Found Office now also online.
Over the last couple of years I followed the work of the DREaM project, aimed at building a community of LIS researchers in the UK. Effective event amplification provided me with an introduction to social network analysis (SNA; nearly two years ago now!) and a host of other research methods.
The DReAM project SNA’d themselves, specifically a cadre of 33 individuals who attended all the f2f events and created the network ‘core’. In the first workshop the participants provided data on (1) individuals’ awareness of the research expertise and knowledge of other participants, and (2) social/ interactional links across the network, data which was collected again at the final workshop. The hypothesis was that analysis of the two sets of data would reveal changes in levels of integration among the DREaM cadre and network density among the group as a whole over the series of workshops – ie that integration and network density would increase.
Initial findings were presented at the final DREaM event and a paper finally published in the Journal of Documentation in October – see Hazel Hall’s post for full details and to download the manuscript. The paper offers a potential model for nurturing and assessing network and community (of practice) development, specifically a developing, or emergent, network based on spontaneously formed ties, which could also be applied to NSMNSS , the legal education community, Danish literary translators, walking types, etc. As well as a useful overview of the development of SNA from the 1930s it provides a model for moving forward from the presentation of network diagrams, discussing features of network articulation and measurement, relational ties and network roles.
Methodology and findings:
- data were input manually into Ucinet v.6 and visualised network diagrams (sociograms) were produced using Netdraw; measures of density and degree centrality were calculated using Ucinet
- the sociograms highlighted the centrality of position of certain participants, prompting speculation as to their identity and the reasons behind this centralisation as well as discussion on the meaning behind some of the more isolated positions occupied by some of the outliers
- the findings from the first round of data collection demonstrated that the participant networks were not very highly connected, and heavily centralised around a small number of actors from one role
- analysis of data collected in the course of the final workshop reveals a demonstrable increase in network density, indicating a much more closely linked and robust network; more evenly linked, with less dependence on two or three very densely networked actors, when analysed by role several categories had moved to a more central position, one category had formed a clique and one category seemed particularly adept at network building, with most members moving towards the centre of the network
- not all the key players were those one might have expected to play such roles; a small number of relatively novice researchers proved to be particularly strong networkers and were central to the network structure (this was not explored further due to ethical concerns)
- greater change in the density of the network with regard to expertise awareness than for interaction, suggesting that even if participants had not had one-to-one interaction with another participant they were still more likely to know of their area of research expertise – ie who knows what, typical of a work related rather than ‘social’ network
- note of caution: in an information sharing network, for example, an actor with a high degree of betweenness centrality may be playing the role of either broker or a bottleneck – for most network patterns multiple interpretations are possible, and it is therefore appropriate to follow up such analysis with qualitative research that seeks to explore likely explanations (data from other sources included a ‘before and after’ audit of skills and feedback on face to face events)
- the results suggest that network density and integration can be increased by structured and informal social and work based interaction; a model of combining workshops with social events and the use of social media reduces the isolation often experienced by the researcher, in particular the solitary, novice or practitioner researcher
- increased network density and integration reduces the dependence of the network on a couple of actors, making the sustainability of the network more likely and increasing network capital – more likely that participants will be able to leverage potential benefits
- potential drawbacks – a higher density of network structure and the formation of cliques may pose a barrier to incomers and increased homogenisation – homophily; it is critical to ensure that barriers to entry to the network remain low with a network of loose ties; individuals should be encouraged to play an active role in boundary spanning, ensuring innovation, opportunity and diversity of viewpoint
- the challenge is to maintain the existing links and further develop the network so that it evolves into a self sustaining and continuously developing supportive community
Specific interventions used to increase and strengthen network ties over the course of the project included pre-event social meetups, a Twitter list, curation over the full event lifecycle, a Spruz community, participant led sessions, event reporters.
The role of event amplification in particular is interesting, an issue which keeps popping up and perhaps has potential in proving its ROI. Effective event coverage can in fact change the nature of an event, ensuring that participants can make the most of f2f interaction and are better able to reflect after the event. Alan Cann touches on this issue too in his recent post on the way forward for #solo13 – the conference as aggregator, building an online community of mutual support. The same goes for MOOCs, but the role of aggregation and curation is often overlooked.
Some #sna bits n bobs picked up from the paper:
Commonly measured network features:
- size – at the actor level: the number of linkages an actor has; at network level: the total number of linkages in the network
- reachability – the accessibility of points of the network based on a notion of path, ie the connected sequence of linkages by which it is possible to move from one point to another in the network; a point is reachable when there is a path between points
- density – the degree to which actors are linked to one another; parts of a path are dense if each of its points is reachable from every other
- centrality – the degree to an individual actor is near others in the network and the extent to which the person lies on the shortest path between others and thus has potential for control over their communication
Examples of relational ties:
- evaluation of one person by another – friendship, liking, respect
- transfer of material resources – business transaction, lending, borrowing
- association/affiliation – jointly attending the same social event, belonging to the same club
- behavioural interaction – talking together, sending messages
- movement between places or statuses – migration, social or physical mobility
- physical connection – co-location at work
- formal relations – authority
- biological relations – kinship, descent
- communication relations – sharing of publications, discussion of ideas
Example of network diagrams from Martin Hawksey: