Summary of Carole Palmer's (University of Illinois at Urbana-Champaign) presentation “Fuelling and Transforming Evidential Cultures of Research” at the 2nd RDA Plenary on 16th of September in Washington D.C.
Carol Palmer’s presentation focussed on research data and culture and drew on research from her work at the University of Illinois. Her research has also been informed by the well-known Understanding Infrastructure Report – and is centred on “data as a fundamentally collective shared resource”.
In Palmer’s presentation, disciplinary cultures were listed among barriers to data sharing, suggesting that rather this is something to be harnessed to join up trust, value and context and follows the established notion of reverse salience.
Palmer references the concept of evidential cultures cited from the Collins (1998) paper on the meaning of data, and uses the gravitational wave research exemplar which compares practice in a US lab with that in an Italian lab. The contrasting open and closed cultures represent evidential collectivism versus individualism. This leads to further questions: What qualifies as a publishable result? Who should take responsibility for validity and meaning? (the lab or the international research community?)
Palmer then moved on to relate some experience form her work with geologists, geochemists and microbiologists at Yellowstone National Park. There is an element of a common culture across these fields in that they are all dependent on field data from the Park. This is an example of site-based data curation, and her research group is trying to understand how to bring data together and have common policies, descriptions, data inventories and divisions of labour. The work has included focus groups to identify what are the essential elements from the researchers perspective and how does this align with other community developments, such as repositories.
The discussions demonstrated a surprisingly rapid shift and convergence on collective responsibility (from individualism to collectivisim), and this transition was associated with discussions around a number of factors: validity e.g. the documentation of sampling events and meaning e.g. the organisation of the collective around data from photographs. The group then shifted further and discussed the collective value around broader impacts of data sharing, which in turn transitioned to debate about efficiencies and strategic science.
Some tensions became apparent:
1) Data sharing is about data producers; data reuse is about data consumers
2) Complex sets (producers) versus usable parts (consumers) (work from the from Data Conservancy Project)
3) Version willing to release versus the version best for reuse (work from the Curation Profiles Project at Purdue).
Palmer then explored examples of added value from data reuse and associated value indicators e.g. ocean modellers add richness and verification from complementary evidence for filed campaign data; rainforest researchers add recalibration and feedback from recovered precision for sensor block temperatures.
Suggestions for RDA
Palmer closed her presentation by thinking about how the RDA can put socio-cultural methods into practice and provided some suggestions:
1) Use a rapid reporting framework with a quick turnaround of key observations for the infrastructure team (from the Data Conservancy Project)
2) Use a common curation vocabulary in the workforce (DPCVocab)
3) Identify data mentors and science mentors in data curation education (for NCAR internships in DCERC – Data Curation Education in Research Centers).
In summary, whilst science aims to reshape scientific endeavours; revolutionary science with high functioning data, should be built on the evidential cultures of “normal” science.
Link to presentation: https://www.rd-alliance.org/sites/default/files/attachment/Palmer_RDA.pdf
Written by: Liz Lyon, Digital Curation Centre, United Kingdom