Desktop version

Home arrow Education arrow Evaluating Collaboration Networks in Higher Education Research: Drivers of Excellence

Research Networks Participatory Evaluation

The work of researchers, in a research project, is structured in various formal and informal networks. Such social spaces can be designed and monitored in order to facilitate interactions, knowledge flows, and knowledge processes (Leite et al., 2012; Merton, 1973; Pinho et al., 2012; Winter et al., 2006). Scientists are immersed in different systems—political, cultural, educational, informational, scientific-technological, and innovational. Scientists are challenged by digital convergence as multiple functions converge to multifunction devices and different platforms coexisting with the features of different electronic devices. Hanging or hosting works in the cloud, for example, can increase availability and ubiquity.

A new interconnected world has emerged. New skills and usability guidelines will become part of the repertoire of the members of the research and collaboration networks. How to achieve these levels of innovation moving toward a widely interconnected future that may as well be highly exclusionary without fully knowing the terrain in which we operate? How to understand the micro space of our own research and collaborative networks, the real ground in which it takes place? How the human dimension of research work should be considered by the evaluation and monitoring processes?

Jonathon Mote and colleagues (2007) propose using social network analysis techniques as a tool for research evaluation in order to understand how real research happens. Because Science is a social process, the evaluation of research networks is naturally a way for improving the research performance. As microsystems of research activities, research networks can be complex objects to be evaluated. Despite the difficulties of such evaluation, the literature review shows few guidelines to overcome this complexity as shown in Chap. 5. Integration of those contributions can help linking conceptual and methodological RNE issues (Rogers et al., 2001; Sala et al., 2011; Wixted and Holbrook, 2012).

However, we highlight the question posed by Spinak (1998, p. 146): “If we have so many tools why are they inadequate?” And his answer: “ ... these tools were adequate to analyze the production of S&T of the central countries, science mainstream, but has serious problems of epistemological and instrumental character for analyzing the production of less developed countries. ”

Of course, each research group leader will report differently to different stakeholders and answer some questions such as the goals/ objectives of the RNE at a specific time, the level of analysis (micro, meso, macro), and type of analysis (static or dynamic/snapshot or changes over time). Having answered those questions, it is possible to prepare the evaluation proposal, where it must be clear not only what is (1) the goal of the evaluation, (2) the object of the evaluation (formal or informal research networks), (3) the time period, but also at what (4) scale or level of analysis the network is located (micro, meso, and macro). At micro, macro, and meso level, quantitative indicators are indispensable, but they must be carefully chosen to produce clear information.

At RNE micro level, it is possible to perform a case study or a participatory evaluation inside the network that should be a collective decision of the evaluators and researchers. The appropriate mix of quantitative and qualitative indicators can lead to useful choices to obtain a richer analysis (Adler et al., 2009; Vasconcelos et al., 2009).

RNPE is a network approach to evaluate research in which researchers and other stakeholders actively engage in developing the evaluation and in all phases of its implementation in order to use results to improve learning, increase skills, and knowledge production. In this kind of evaluation, participants share knowledge and learn together to take corrective actions. This evaluation can develop leaders and build teams. We propose an RNPE process, divided into five main phases: (1) sensitization (discussion about evaluation needs’ and criteria); (2) conducting the evaluation by quantitative indicators; (3) qualitative indicators facing quantitative results; (4) results (internal) and dissemination (if needed); and (5) deliberation and planning for excellence.

The micro-level research network evaluation begins by (1) sensitization, a phase in which values and interests are put on the table. A first step is related to the choice of the indicators and they must be discussed with group members. The discussion about evaluation needs and criteria is the start up for the process. If the members of a research network decide about (2) quantitative indicators’ evaluation, an evaluation protocol can be a result of this phase. It is a guiding document to perform a robust evaluation, in our experience.

It is possible to begin by identifying the research network topology, by observing the network structure that includes identification of authors as nodes/members/actors and their connections by institutions, countries, and coauthorship analysis. As we discussed before, research collaboration can be evaluated through coauthorship. So, in our experience, ten research collaboration network indicators are enough to micro-level evaluation. We can depart from the group leader publication and his/her coauthorship profile. Once the protocol

Table 6.1 Protocol for analysis of collaboration in articles

1 Identification data

Researcher name:

Institution name:

Researcher’s field of knowledge: Researcher responsible for analysis: Date of analysis:

2 Network structure from articles (graph analysis)







Network actors

In the country (box)

In the country, but extragroup (diamond)

Outside the country (ellipse) Total network (includes ego)

Identification of clusters and nodes (vertices)

Isolated (linked only to the leader)

With two components (one besides the leader)

With three or more components (two besides the leader)

Actors’ institutions

Universities and

other higher





In the




In the








Egocentric network (degree of centralization and leader power)

Pure (relations only between the leader and the other actors) Interconnected (relations between secondary authors)


Table 6.1 (continued)

3 Characterization of the production from articles (spreadsheet analysis)







Total of articles

Articles by number of authors/actors


  • 1 (Only the ego)
  • 2
  • 3
  • 4
  • 5
  • 6-10 > 10

Articles by location

In the country Abroad

Total of journals

Journals by location

In the country Abroad

Source: Leite et al. (2014a)

is set, see Table 6.1, we obtain a picture of the group performance by micro-level quantitative indicators for RNPE.

< Prev   CONTENTS   Source   Next >

Related topics