Although scientific knowledge is considered by many a universal and context-free product, its producers are often embedded in geographically bounded networks of research collaboration. However, in an age of globalisation these local networks of knowledge production are challenged by pressures to make science more efficient and to align its priorities with problems of global relevance such as climate change and worldwide epidemics. Against this background, the dissertation sets out to examine changes in the contemporary geography of research collaboration and explores how these changes affect the publication of research findings in scientific journals. The dissertation starts with introducing a framework to understand how geographical space structures research collaboration among researchers. The two structuring principles in this framework are a logic of proximity that provides solutions for coordination problems in research practice, and a logic of stratification that provides researchers with differential means to engage in collaborations. The logic of proximity mainly follows from the importance of physical co-presence both for carrying out the complex tasks associated with scientific research and for establishing trust in research results. The logic of stratification is an outcome of the reward system in science which provides differential credit to researchers on the base of their past productivity. Globalisation affects these logics through technological advancements in ICTs and mobility, and through the harmonisation of research policies and practices across territories. It is hypothesised in this dissertation that these changes have implications for geographical patterns of research collaboration, and also for the way research findings are communicated in scientific publications. The empirical validation of this framework centers around two main themes that very much bear the imprint of globalisation in science. The first theme concerns the research policies of the European Union that are focused on the harmonisation of regional and national institutions in Europe in order to create an integrated ‘European Research Area’ (ERA) which should make the European research system more efficient and competitive. The Framework Programmes are explicitly designed to facilitate this integration process and in doing so they fund thousands of transnational research projects making it the largest transnational funding scheme in the world. Against this background, Chapter 2 evaluates the extent to which European research collaboration networks are already spatially integrated based on publication and patent data with multiple addresses. The results indicate that research collaborations in Europe are structured by geographical proximities as the choice for collaboration partners is impeded both by the kilometric distance between researchers and by national borders separating them. The chapter also presents some evidence that research collaboration networks are stratified on the base of similarity in productivity and access to resources. This logic of stratification operates irrespective of the location of researchers vis-à-vis each other. The main conclusion that follows from this analysis is that the present efforts towards the creation of ERA are well justified. The empirical study in Chapter 3 develops a dynamic approach to the geography of research collaboration by studying whether the logic of proximity is changing over time. The main argument of this chapter holds that one should make a conceptual distinction between a possible changing effect of geographical distance and a possible changing effect of territorial borders when studying proximity dynamics in research collaboration networks. When making such a distinction in the context of the European research system, the chapter shows that it is primarily the importance of regional and national borders that is decreasing over time, but that the role of geographical proximity in structuring research collaborations is remarkably stable. The findings indicate that globalisation in science is mainly realised through the harmonisation of territorial institutions, but that physical co-presence remains an important coordination device for exchanging complex forms of knowledge that cannot be easily communicated over large distances. The objective of Chapter 4 is to study to what extent the Framework Programmes (FPs), as the main funding instrument of the European Commission, are affecting the geography of European research collaboration. It is hypothesised that, in case the FPs indeed render territorial borders less important, they are likely to create (new) stratified networks of research collaboration that disproportionally consists of high-performance researchers located in Europe’s core regions. Contrary to the expectation no evidence for this hypothesis is found. The presented analysis indicates that the FPs indeed have a substantial effect on promoting international scientific collaboration networks which are still relatively uncommon in comparison to national collaboration networks. However, it is also shown that acquisition of FP funding is rather equally distributed over Europe and that the FPs are more effective in establishing ties between poorly connected researchers than in further strengthening existing ones. When stimulating already existing networks the FPs run the risk of being a substitute for other funding sources. This implies that current EU research policy is in line with the cohesion objective of the European Union. The second theme of this dissertation concerns the global standardization of medical experiments on human subjects. In recent years, proponents of an evidence-based medicine have pushed for standards concerning the conduct of clinical trials and subsequent publication of research findings in clinical trial registers and scientific publications. This standardization process is closely linked to an increase in the number and size of clinical trials that involve scientific researchers and patients from across the globe. The empirical chapters address whether this standardization process has an effect on several aspects of scientific publishing including the constitution of authorship on publications, the communication of evidence after study completion, and the presence of error in scientific publications. In order to analyse these questions, a database is created that links information on registered clinical trial projects (www.clinicaltrials.gov) to scientific publications of the main findings after study completion. Chapter 5 focuses in this respect on the standardization of good clinical practice (ICH-GCP) which has made the exchange of clinical data between geographically dispersed research sites less complicated. This has resulted in a process of global outsourcing with increasing enrolment of patients from emerging economies, especially in Central and Eastern Europe, Latin America and Asia. The chapter describes this globalisation tendency and studies whether worldwide patient involvement in clinical trials is reflected in the geographical composition of scientific management teams in those trials. The chapter develops an empirical strategy to determine the geographical distribution of management teams by using authorship data from publications reporting on primary outcomes. On the basis of this data it is shown that, given patient involvement, authorships are disproportionally granted to researchers in a few leading countries. The chapter discusses possible adverse consequences of this situation especially concerning the monitoring of clinical trial quality and (the lack of) interactions between researchers that are in immediate contact with patients and researchers that design trials and interpret their results. Chapter 6 concentrates on the publication behaviour of pharmaceutical companies who are well known for their strategy to withhold negative research findings from the scientific literature. To remedy this situation several authorities have recently mandated both registration of clinical trials before study onset and publication of major research findings after study completion. The main question holds under what conditions pharmaceutical companies decide to publish their clinical trial findings either in scientific journals or on the web. The main hypothesis is that under the new institutional context pharmaceutical companies will continue to highlight positive results in the scientific literature as it provides them with certification that their research findings are scientifically sound, methodologically rigourous and thus credible. Negative results, by contrast, are expected to be published on the web. This hypothesis is tested against a sample of clinical trials that assess the efficacy of glucose lowering agents in diabetes patients. The results indicate that firms continue to highlight positive results in scientific journals which results in an ongoing and persistent bias of evidence in the literature. Finally, Chapter 7 studies the production and detection of error in scientific publications on the basis of published errata and retractions. It is derived from earlier chapters that geographical proximity remains an important coordination device in research practice. This begs the question whether researchers operating in geographically dispersed research projects are also more likely to produce error because effective peer-control may be lacking and the establishment of mutual understanding hindered. The chapter addresses this question by making a conceptual distinction between modes of coordination that influence error production, and the prestige of research findings that influences error detection. With respect to prestige of research the chapter shows that editorial policies of scientific journals may actively steer the process of error detection by organising impact around particular findings and by enforcing strict publication guidelines. After controlling for these factors the analysis finds that geographically distributed research results in less accurate scientific publication. Globalisation tendencies thus put increasing responsibility on the publication system to correct errors in publications. Based on the findings of the empirical chapters, the overall conclusion of the dissertation is three-fold. The first conclusion holds that changes in the contemporary geography of research collaboration are mainly visible in institutional harmonisation across territories, rather than in a tendency towards a ‘death of distance’ per se. This paradoxical process provides new prospects for worldwide research collaborations, but limits at the same time the possibilities to make these prospects work in actual research practice. Second, the presented analysis indicates that in an age of globalisation, science does not become a global level playing field where chances of success level off. Rather, stratified structures are reproduced at different spatial scales via the creation of new reward systems and global research collaboration network that exhibit high entry barriers. Third, globalised science reveals new publication practices that concern authorship norms, the prevalence and correction of error in scientific publications and the conditions under which disclosure of research findings takes place. In this respect, new global contexts have often been cited as contributors to the quality, impact and practical application of research findings. The results presented here do not support the argument and at least point to some potential side-effects of geographically distributed research. These effects require a rethinking of science’s institutions in light of globalisation.
|Qualification||Doctor of Philosophy|
|Award date||18 Oct 2012|
|Place of Publication||Eindhoven|
|Publication status||Published - 2012|