What are typical data sharing policies in research consortia?

Question

Since I am very new into academia I am really unaware of typical regulations in research consortia. Right now I am working in a research consortium in which many different research teams will do different surveys. On the other hand, I will be responsible for a more holistic point of view and will also do some upscaling. Also, I am doing my PhD in this project.

Right now some of the regulations sound really awkward and unfair to me:

• Primary data is property of the one who collected it.
  • This is alright of course
• The collector of data has the first right to publish about it.
  • Of course the collector of data shall have the fame for collecting it. I could see a case arising where I just want to validate my results by cross-referecing this data. I honestly do not see why I shouldn't be able to publish a paper where I use this data in a side study. It shouldn't grab any fame from the collector because I will not describe this data in detail nor will it (in this case) be a big part of my publication. This is not such a big deal though, because I can still ask if it is possible when an actual use case arises and of course I understand that they want to protect themselves for their work.
• The collector of data is allowed to share the data with other researchers when they are granted co-authorship and intellectual property rights on the results.
  • This one is really where I have trouble with. I am basically obliged to use the data but do not have any right to access it. I rely on my partners to allow me to use the data and I don't know when/if I get access. That is really not in the interest of the project and is really unfair. Also they demand co-authorship for using their data without contributing anything else to my publication, which again I can somehow understand if the data enables my publication but not if I only wanna use it in a side study e.g. for cross-validation, though.
• An agreement with the data collector is needed if a PhD student or other students want to use data for their theses
  • Of course I understand the need of such an agreement. There is no regulations if and when I should get data though and thus again I am relying solely on the good will of the data collector without any regulations. Since I wanna do my Phd on this project, this could really result in problems for me (and/or worsen my research), if they deny me data. Also there is a critical imbalance here, because every other Phd candidate will be working in a group, which collects data while I am not.

The biggest problem I have with the draft of the data policy is, that I do not have any rights on the data although the project is designed in a way that I should be using this data and thus I always depend on the good will of my collaborators to fullfill my obligations defined in the proposal. They can in theory hide their data untill the project is finished and it shall become open access. Also I probably need to publish solely with my professor to do my PhD and the data policy sounds really rigid and even harmful to me. Also it will only benefit data collectors and basically hinders my research. Why shouldn't I be allowed to publish on a different topic with the same data?

Why do I need to give co-authorship to someone who didn't even work on the paper and does not want to work on the paper? If the data was essential for my research I could somehow understand it, but if it is not? I found this for co-authorship and that sounds reasonable to me.

So my question is: Which of these points are common practice or are they as rigid as I think they are?

Answer 1

These all seem fairly usual, except perhaps the mandate for coauthorship (more below), and in fact, it's excellent that your consortium has set up guidelines for data-sharing because it can get messy when done casually.

You seem to have the perspective that data collection is trivial -- and compared to many other forms of data collection, surveys are -- and that PIs should be willing to give you the data and let you run off with it to do your models.

But designing materials, recruiting participants (paying them), and getting the funding to do all this is not trivial. Most people would agree that someone who collects data should get first crack at it because they put in the hard work to get it. The paper that first releases data is usually what is cited, and so labs want that first paper to have their name on it.

The collector of data is allowed to share the data with other researchers when they are granted co-authorship and intellectual property rights on the results. ... An agreement with the data collector is needed if a PhD student or other students want to use data for their theses

It might be a tad onerous to start off with listing them as authors, but on the other hand, this would also oblige them to discuss your results, and help you interpret data. IP rights don't mean anything here. An agreement is fantastic!

Why do I need to give co-authorship to someone who didn't even work on the paper? If the data was essential for my research I could somehow understand it, but if it is not? I found this for co-authorship and that sounds reasonable to me.

If there couldn't be a paper without their data, it seems essential, no? I agree that once data have been made public, they no longer need to be made authors, but when you are collaborating before data has been released, this all seems like standard and good practice to me

---

That said, if you've been accepted to this consortium, they have an obligation to help you with your needs. It seems like you need to have a good talk with your advisor and the other stakeholders and understand what you need to do to start working with these people and their data. Drafting up the aforementioned agreement sounds like a great place to start!