‘Sharing data leads to more citations, and I see it as a moral obligation’

Professor Peter Verburg explains why he shares his data in open access via DataverseNL.

05/24/2019 | 10:25 AM

‘Within my Environmental Geography research group, we have recently begun archiving and publishing our data on DataverseNL’, Professor Peter Verburg says enthusiastically. ‘VU researchers can use DataverseNL free of charge for datasets up to 50 GB in size, and the platform is user-friendly. It also allows us to share our data in open access, which I believe is very important.’peterverburgklklein

Data on land use
‘As environmental researchers, we are interested in spatial data on land use. This includes information relating to urbanisation, deforestation and food production, for instance. We collect data by interpreting satellite images and conducting interviews, among other things. We then use this data to create models (scenarios) that show what the land use in these areas might look like in the future.’

People do not have to ask for permission to use our data
‘We use a great deal of data from others. Our models in turn produce data that others can use, for example in research on biodiversity or CO2 emissions reduction. I believe we have a moral obligation to share our data. Our group’s policy has long been to make the data that we collect and produce available through open access publishing. We do this as soon as we have published the first paper on the topic. By “open access”, I mean that people do not have to ask us for permission to (re)use our data. In Dataverse, researchers can indicate whether or not people need to request permission to use their research results.’

Sharing data in open access leads to more citations
‘The biggest advantage of sharing data in open access is that our papers are cited more often. This is good for our careers as well as for the research assessments. My PhD students who publish their datasets have above-average citation records. In addition, some of our data models have been used by institutions such as the EU’s Joint Research Centre and the FAO.’

‘Another important reason to share our data in open access is that we want our research to be transparent and reproducible. Others should be able to inspect it. I occasionally receive messages from researchers expressing how grateful they are that we share our data. People in the academic world respect what we are doing.’

‘One disadvantage of sharing data in open access is that you receive fewer requests to collaborate or co-author. If you make your data openly available, people will start using it themselves without having to ask you to actively contribute. We have also noticed that despite the supporting documentation we provide, our models are sometimes used incorrectly, such as in a context for which they are not intended. We also receive a lot of questions from non-professional users. Unfortunately we can’t help them since we simply don’t have enough time. Finally, there are sometimes people who will make a small change to your dataset and then publish it under their name.’

Produced with public money
‘For me, the benefits of openly sharing data outweigh the drawbacks. I also want it to be truly open. I get annoyed by people who “sit on their data”, even though that data has been produced with public money. Some researchers who publish their data, but not in open access, will only give you permission to use the data if you list them as a co-author. I see that as an abuse of their position. Furthermore, I think that if financial backers make data sharing mandatory, they should also check whether this is done properly. I still regularly see instances where researchers only show a scan of their chart. You can’t use their data that way.’

Self-managed data via DataverseNL
‘The great thing about DataverseNL is that it allows you to manage all your data yourself. If you publish with a publisher or other organisation, you can’t just go and fix small mistakes or improve the documentation after it has been published. With Dataverse, we can always access the documents ourselves and we retain control over our data. We are pleased with the support we receive through the University Library.’

Recording data in PURE
‘We also record as much of our data as possible in PURE/the VU Research Portal, since data is research output, too. We use PURE to generate our research output reports, so it’s important to have everything in one place. In addition, we can use these datasets to promote ourselves on our profile pages in PURE. This shows what we are capable of, which can in turn lead to new collaboration opportunities. PURE also provides you with a permanent link to your data so that you will always be easy to find.’

For more information about archiving and publishing data, see the RDM-Libguide

Dataverse VU University Amsterdam

Please send an email to: researchdataservices.ub@vu.nl