Interdisciplinary research data: Challenges & observations from a data steward’s perspective

Throw­back to the Research Data Alliance (RDA) 17th Ple­nary meet­ing in Edin­burgh! On 21st April, Yan Wang, Inna Kouper, Eva Mén­dez and Con­nie Clare co-chaired a Bird of a Feath­er (BOF) ses­sion about the Chal­lenges of Curat­ing Data from Inter­dis­ci­pli­nary and Col­lab­o­ra­tive Research.

Aims of the session:

• To share expe­ri­ences of work­ing with inter­dis­ci­pli­nary research teams and their data.
• To share exam­ples of dif­fer­ent approach­es to dis­ci­pline-spe­cif­ic data stew­ard­ship pro­vi­sion.
• To exam­ine data dis­cov­ery in an inter­dis­ci­pli­nary research con­text.
• To iden­ti­fy the emer­gent themes from the dis­cus­sion and trans­form them into action items for a poten­tial work­ing group or inter­est group to address.

The agenda

Check out the pre­sen­ta­tions:

• Inter­dis­ci­pli­nary and high­ly col­lab­o­ra­tive research (IHCR) by Inna.
• Inter­dis­ci­pli­nary research data — chal­lenges & obser­va­tion from a data steward’s per­spec­tive by Yan and Con­nie.
• FAIRs­FAIR … and the chal­lenge of inter­dis­ci­pli­nary research (when the F of Find­able, become a D of Dis­cov­er­able) by Eva.

A session summary by Yan

Inter­dis­ci­pli­nary research is receiv­ing more and more atten­tion due to the increas­ing com­plex­i­ty of soci­etal chal­lenges and the decreas­ing com­mu­ni­ca­tion bar­ri­er across dif­fer­ent research fields. Man­ag­ing inter­dis­ci­pli­nary stud­ies and data becomes a new and impor­tant issue that pos­es many ques­tions when it comes to research data man­age­ment.

There is still con­fu­sion about the dif­fer­ence between inter­dis­ci­pli­nary and mul­ti­dis­ci­pli­nary research. Hav­ing researchers from var­i­ous dis­ci­plines work­ing togeth­er on one project does not imply inter­dis­ci­pli­nar­i­ty, unless they inte­grate the knowl­edge, meth­ods and data via a syn­the­sized approach.

We had ques­tions about what exact­ly inter­dis­ci­pli­nary data is… 

• Is it a new type of data? 
• Is it the appli­ca­tion of data in a new domain? 
• Is it a new method of pro­cess­ing data? 

From the per­spec­tive of data stew­ards, we made a few obser­va­tions about FAIR inter­dis­ci­pli­nary data which brought us more ques­tions and points for dis­cus­sion with the com­mu­ni­ty.  

Findability

The chal­lenge regard­ing data find­abil­i­ty is two-fold. The first chal­lenge relates to the data col­lec­tion process. Before search­ing for suit­able data, it can be dif­fi­cult to define the scope and bound­ary of data sources in inter­dis­ci­pli­nary research. This is due to the fact that many inter­dis­ci­pli­nary research projects are new and have an explorato­ry nature which are often pro­posed in response to spe­cif­ic fund­ing calls. This means that inter­dis­ci­pli­nary projects are often tem­po­rary, last­ing for a lim­it­ed peri­od of time. In addi­tion, it’s dif­fi­cult to syn­the­size exper­tise and data sources from mul­ti­ple dis­ci­plines. This makes it chal­leng­ing to cre­ate a com­pre­hen­sive overview of data sources.    

Anoth­er sec­ond chal­lenge relates to the data shar­ing process. Due to the explorato­ry and tem­po­rary char­ac­ter­is­tics men­tioned above, it’s often unclear as to where the data should be pub­lished. Who are the tar­get audi­ences for data reuse? Would a gener­ic repos­i­to­ry be suf­fi­cient or would a dis­ci­pline-spe­cif­ic repos­i­to­ry help to reach the rel­e­vant audience(s)? 

Accessibility

Cross-dis­ci­pli­nary data use and access require addi­tion­al efforts. For exam­ple, med­ical data is not nec­es­sar­i­ly lim­it­ed for reuse in med­ical fields, but is becom­ing increas­ing­ly impor­tant for social and human­i­ties research fields. How­ev­er, access­ing such data usu­al­ly requires extra pro­ce­dures to ensure its secure and respon­si­ble reuse. 

It’s also pos­si­ble that data is not avail­able for use out­side its research dis­ci­pline. Rea­sons for this vary from case to case. If an inter­dis­ci­pli­nary project demands data col­lec­tion from mul­ti­ple dis­ci­pli­nary sources, the acces­si­bil­i­ty require­ments and con­sid­er­a­tions for all sources need to be con­sid­ered. This is not an easy task.

Interoperability

Com­pared to dis­ci­pli­nary research, the cre­ation of data from con­duct­ing inter­dis­ci­pli­nary research does not always have a clear approach to fol­low. Fol­low­ing the argu­ments above regard­ing the explorato­ry nature of inter­dis­ci­pli­nary research, there are less stan­dard pro­ce­dures, guide­lines and prac­tices in place. Whilst there may be stan­dards for those dis­ci­plines involved in the inter­dis­ci­pli­nary project, it’s uncer­tain which of these stan­dards should be adopt­ed or inte­grat­ed to ensure the inter­op­er­abil­i­ty of such data. 

Reusability

In gen­er­al, reusabil­i­ty of data can be dif­fi­cult to mea­sure since its added val­ue is often not cer­tain until post-pub­li­ca­tion and so can only be esti­mat­ed at the moment of pub­li­ca­tion. There­fore, the reusabil­i­ty of inter­dis­ci­pli­nary data is based on hypoth­e­sis. Fur­ther­more, men­tioned in the dis­cus­sion about data find­abil­i­ty, the tar­get audi­ence for inter­dis­ci­pli­nary data is not always clear which makes it even hard to assess reusabil­i­ty.  

Ask the audience

In order to clar­i­fy some of our thoughts and per­spec­tives on inter­dis­ci­pli­nary data, we asked the BOF ses­sion par­tic­i­pants the fol­low­ing ques­tions dur­ing a dis­cus­sion using Men­time­ter:

• What is your pro­fes­sion­al role?
• What is your research discipline/field of exper­tise?
• What is inter­dis­ci­pli­nary study/data?
• Do you have spe­cif­ic exam­ples to share?
• Do you create/handle/advise on inter­dis­ci­pli­nary study/data?
• What is the added val­ue of inter­dis­ci­pli­nary study/data?
• What are the chal­lenges of inter­dis­ci­pli­nary study/data?
• do you know of any com­mu­ni­ties that are involved in inter­dis­ci­pli­nary study/data?

The ses­sion brought many new ideas and pos­i­tive thoughts about inter­dis­ci­pli­nary data. We look for­ward to fol­low­ing up on these dis­cus­sions in the future.

Writ­ten by Yan Wang (TU Delft)
Edit­ed by Con­nie Clare (4TU.ResearchData)