PSA2016: The 25th Biennial Meeting of the Philosophy of Science Association

Full Program »

Investigating Data Journeys Across Scientific Fields

This poster presents the empirical work under way within the European Research Council Project “The Epistemology of Data-Intensive Science”, which investigates and compares data practices across three areas of science that have long dealt with large datasets, and have transformed their research activities to take advantage of methodological and technological innovations in the ways in which data are produced, disseminated, stored, retrieved, visualised and analysed: biology, biomedicine and oceanography. Digital networks and information technologies have recently supported attention to data themselves as key research outputs that are valuable in their own right, and related changes in the ways in which data are handled and interpreted. The epistemological implications of such changes have yet to be sufficiently recognised, understood and discussed by philosophers of science. To provide empirical grounding for such philosophical work, the project investigates how online databases and infrastructures for data sharing and analysis are developed and managed to make data widely accessible, visible, and usable. Data processing is not a trivial operation: as recently argued by Leonelli, the epistemic significance of data, and indeed which objects are being recognised by scientists as data in the first place, typically changes in course of the “journey” of data from their site of creation to different situations of processing, use, and storage. The poster presents the fieldwork conducted by project members, and compares data journeys across different scientific fields in terms of the conceptual, social, and material conditions that shape the movements of data across sites. In biology, data production, dissemination and integration has often been anchored on specific (“model”) organisms, for which data are gathered at molecular, cellular and ecological levels, and related to material specimens stored in stock collections. We document such practices particularly in fission yeast and plant science, while also noting the rise of cross-species areas such as phenomics, where the collection and visualisation of complex morphological images from a highly diverse species pool generates opportunites for - and challenges to - established ways of integrating data. In biomedicine, data linkage projects repurpose public health datasets by linking them to other health (e.g. GP practice data, hospital databases, and project-specific) and interdisciplinary (e.g. environmental and metereological) datasets. Here, auspice to support research of both statistical relationships of population-level magnitude and granular observation of the health of individuals is challenged and often frustrated by complexities related to data ownership and governance, and lack of standard solutions to methodological problems. In oceanography, the size of the oceans and the high economic demands of producing data have led to data production systems that involve non-scientific actors such as sailors, fishermen, and even animals. We focus particularly on a long-term survey dating back to the 1950s aiming to create a database of spatio-temporal plankton distributions using sampling devices towed by commercial ships. Building on this work, this project is developing an analytic framework to evaluate the impact of digital technologies for data dissemination on research and understand the conditions under which data found online are used to generate new scientific knowledge.

Author Information:

Gregor Halfmann    
Exeter Centre for the Study of the Life Sciences (Egenis)
University of Exeter

Niccolo Tempini    
Exeter Centre for the Study of the Life Sciences (Egenis)
University of Exeter

Sabina Leonelli    
Exeter Centre for the Study of the Life Sciences
University of Exeter

 

Powered by OpenConf®
Copyright©2002-2015 Zakon Group LLC