Skip to content

The stable partnership graph

Jump to bottom
Alberto Cottica edited this page May 20, 2016 · 2 revisions

We are interested in intermediated access to European research funding. The European Commission wishes for its research programmes to be accessible to all, but the popular myth is that writing a successful proposal (and running the project once it is approved) is rather difficult. Not all the action is in the quality of the proposal. Insiders are supposed to be better at it than newcomers. Is this true?

We address this question in a network perspective. To do so, we:

  1. Aggregate data in the form of a stable partnerships graph.
  2. Explore the graph and its internal structure.
  3. Look for the signature of intermediation in the graph.

European research funding accrues not to individual organisations, but to consortia. Therefore, the population recipients of grants can be represented as a network of projects and organisations. Projects are connected to organisations when they are members of that project's consortium.

For meaningful analysis, we next:

  1. Project this graph on to a one-mode one, where all nodes represent organisations. Edges represent partnership in projects. Two organisations are connected if they participated in at least one project together. If they participated together in n projects, they are connected by n edges.
  2. "Stack" edges, so that any two organisations can be connected by either zero or one edge. Edges still represent partnerships. Given an edge from organisation A to organisation B, the weight of their connecting edge is equal to the number of projects they have been partners in.
  3. From the resulting graph, we next look at what we call stable partnerships. These are partnerships of minimum weight equal to 2. We believe this to be relevant, because stable partnerships are more conducive to innovation and innovation diffusion. They mean that organisations have found R&D partners that they value enough to go back for more after one project. In a sense, they are an indicator of outcome for European funded research.
  4. From the resulting graph, we isolate and study the giant component. This represent the "scene" of European funded research; the web of organisation engaging in joint projects, along which information and knowledge can diffuse.

We do all for both FP7 and H2020. For comparability (H2020 is ongoing), we only take the first 28 months of each framework programme.

Clone this wiki locally