This project explores an innovative interdisciplinary approach for studying the dynamics of scientific collaboration. Over the last decade, science policy has increasingly focused on the benefits of knowledge sharing, openness, and collaboration, a policy theme that has motivated a number of large recent investments by the NSF and other funding agencies in cyberinfrastructure development and deployment. The success of these investments however depends on attention to field-specific practices and cultures that may influence or even block adoption of cyberinfrastructure. This exploratory project sets out to advance a new interdisciplinary methodology that integrates ethnographic field studies with the analysis of large-scale publication networks to strengthen the empirical basis for understanding field differences in scientific collaboration.
Methodologically, this project offers an innovative integration of quantitative network analysis of large publication networks with qualitative ethnographic field studies. Theoretically and substantively, it illustrates on the multi-level temporal dynamics that shape, and sometimes frustrate, efforts at scientific collaboration and innovation. The methodology developed in this project supports field-level comparisons of scientific communities and fills a critical gap in the research capabilities of scientometrics, innovation studies, and science policy, which are too often split between macro-scale (countries, disciplines, journals) or micro-scale (individuals, specific research localities) analyses, neglecting meso-scale dynamics that are formative to the shape and outcome of scientific knowledge production within and across fields.
Specifically, this project investigates the addition of a temporal dimension to a network ethnographic approach to study and compare temporal dimensions of scientific collaboration across research fields. The test cases are provided by two research fields (in ecology and at the boundary between physics and chemistry) that exhibit significant variation in the temporal rhythms that underlie the structure of short- and long-term collaborations within a research group and between this group and its collaborators. Adding temporality to the analysis increases the resolution of collective structures in publication networks to support the strategic sampling of ethnographic field sites; enables the study of the emergence and evolution of collaborative structures at the team, sub-community, and field level and the detection of field-specific historical trends; and supports the search for field differences in the timing and rhythms of collaborative activities. This EAGER projects serves to explore and test the suitability of a network-analytic approach to capture scientific collaboration dynamics at various scales before deploying it in larger-scale empirical studies.
Broader Impacts: Through strengthening the empirical base to inform science and innovation policy and to guide investment decisions, this exploratory project develops a methodological approach with a wide re-use potential for the comparative study of collaboration dynamics in scientific communities. Understanding the field-specific dynamics, tensions, and challenges of collaboration practices and their evolution over time is critical when the potential benefit and impact of policy and socio-technical interventions (such as cyberinfrastructure, data sharing mandates, and new incentive schemes), and, by improving metrics to quantify the impacts of these investments, complements existing investments and informs future investments in these specific areas.
This project set out to explore an innovative interdisciplinary approach for studying the dynamics of scientific collaboration. It used both quantitative and qualitative research methods to explore temporal dimensions of collaboration in scientific fields. The combination of these two methods is innovative and extremely useful. The quantitative methods (e.g. network analysis) reveal laege-scale macroscopic features. These then provide guidance on where to target more labor intensive ethnographic studies that are more nuanced and reveal why in addiition to what. In addition, there is a feedback loop where the qualitative analysis informs new rounds of network analysis. Our project partner at Cornell University used ethnographic field research to study in depth how various aspects of time shape scientific collaborations. In our part of the project we developed software to visualize how scientific research groups who collectively generate new knowledge, change over time, namely their collaborative links, participation in the community, and their focus on specific research topics. Such maps of scientific communities allow to track their evolution and can support the ethnographic study of a scientific fields by highlighting specific field sites and interview partners of interest. Further, they can provide an important tool for science policy analysts to understand the past impact and anticipate future impacts of funding decisions.
