GDN relys on 11 'Regional Network Partner', so-called Hubs. These Hubs coordinate the regional windows of GDNet and host the Regional Research Competitions (RRCs). As they have a lot of decisions to take these network partners are a crucial part of GDN. Therefore I searched for a possibility to unveil a collaboration network of these Hubs.
Considering the fact that the Hubs differ severely in terms of their nature, some are networks of organisations (like EADN), some are research institutes (as CERGE-EI), it is very difficult to compare them. I decided to read out the html-links of the (Hubs-)websites and build a network out of these informations. Although one could argue that the links are not 100% comparable to collaborations, I think that this network is as close as I could get to a, let us call it, 'network of friendship'.
Click to enlarge: Figure 7
Source: M. Schlögl
Red circles represent websites of GDN-hubs, blue rectangles sites were the Hubs are referring to. As only Hubs can have outgoing connections (only the Hubs-websites have been scanned), it is due to the analysis method that they dominate the network.
The node duke.edu represents BREAD, the Hub for North America. As it is situated at Duke University I used the main-domain duke.edu for incoming links. Therefore it could be that BREAD itself would be less central to the network than it is now (if a Hub for example links to the main-domain of Duke University it is counted as a link to BREAD). For outgoing links I used only the sub-domain of BREAD.
The bigger the rectangles and circles are, the more central the website is to the network. For making the network a little bit more clearly I deleted every website with (originally) only one link (646). Centrality to the network rises with the number and degree of incoming links.
It is very interesting that nearly all important nodes represent 'northern' organisations. Among the most important ones are (elite-)universities like Harvard, London School of Economics, Yale, Princeton, Stanford and others more, less important but still there are international organisations like Worldbank (worldbank.org is pretty important), OECD, ILO, UNCTAD and IMF. One of the most central nodes is GDN itself (gdnet.org), that shows that at least the Websites of the Hubs are strongly connected to gdnet.org.
To sum up, I can say that the results of this research were far beyond my expectations. Although my working assumption was that the websites of the Hubs are not very well connected to organisations in the South, I found far less connections to 'Southern' Organisations than I had expected. As the task of the Hubs is to foster scientific research in their corresponding 'Regional Window', it is remarkable that I found only very few links with 'Southern' top-level domains.
Conclusio
On the one hand, following the present analysis, GDN can be viewed as an instrument to give southern scientists the possibility to take part in northern scientific discourse. As we saw the majority of registered scientist is from southern countries, they use the advantages GDN is offering them: free access to scientific papers through Jstor and GDNet, the possibility to take part in Global Research Projects or to get their own work granted by GDN. In an online-survey among scientists registered at GDN carried out in 2009 (Schloegl 2009), most of respondents expressed rather positive feelings towards GDN.
On the other hand, the intention to foster scientific output in the south has not been very fruitful up to now. Papers are still produced and organisations still situated in the north. Some of our analysis even gave rise to doubts about the honesty of these intentions: What else could be the reason for Hubs to showcase Harvard and Oxford, rather than southern scientific organisations through their website? Additionally we have to question the severe brain drain from southern to northern countries within GDN. GDN should foster, not derogate research capabilities in southern countries.
We found that GDN-papers do not have a ver y high impact on the discourse. Therefore GDN should think about introducing a peer review process and publish papers on its own. That would make it more easy to support young scientists and maybe even strengthen scienti?c structures in the south, boost the stock of GDN-papers and make GDNet a more common place for research.
--> back to the KEF forum 'Knowledge, Media & Networks'
