Protocol for RDR Research Administrators

Scope  

This document aims to prepare research administrators to fulfil their role in the Radboud Data Repository (RDR). It assumes that the research administrator has access to the RDR with administrator rights and has basic knowledge of the features of the RDR as described in the help page (https://data.ru.nl/doc/help).  

Responsibilities 

A research administrator is a person holding the administrative responsibility in one of the repository’s Organisational Units (OUs). A research administrator must be employed by Radboud University and must use a u- or z-number account to log in on the RDR. There are three main responsibilities of the research administrator role:  

• Authorizing users within the Organizational Unit  

• Creating, managing, and removing collections 

• Access to all collections in the Organizational Unit for scientific integrity purposes  

A fourth responsibility can be fulfilled by the research administrator, but also by a data steward or director of the research institute:  

• Developing a user policy for the RDR  

Keep in mind that these responsibilities do not necessary overlap with the responsibilities of a data steward. The research administrator’s responsibilities are administrative in nature, those of a data steward are related to policy and content of Research Data Management. The role of research administrator can be fulfilled by a data steward, but also by someone from the administrative staff of the research institute. 

Authorizing users in the Organizational Unit 

Description of the task 

The research administrator can add users as a member of their OU in the RDR. Users can be authorized for different roles: 

• Regular users of the RDR This role is automatically assigned to all users with a user profile in the RDR. Users can create their own user profile by logging in once to the RDR: https://data.ru.nl/doc/help/quickstart.html#step-1-create-your-repositoryname-profile.   

• Members of an OU Only users that are employed by Radboud University (RU) or Radboudumc can be selected to become members of an OU. Prior to authorizing a user as an OU member, the research administrator must check whether the user is employed at the relevant research institute (i.e., OU)  

• Metadata viewers for collections in your OU This user can view the metadata of all your OU’s collections in the RDR, even of the internal collections (DACs and RDCs). Only OU members can be selected to become metadata viewers of that OU. It is up to your research institute to decide whether it is desirable if (and if so, which) researchers of the institute can view metadata of their colleagues. Some research might be too sensitive to allow even direct colleagues to view metadata of internal collections. If this is the case for your research institute, it is best not to make any researchers “Metadata viewer”.

• Eligible managers This user is eligible to become a manager of a data collection in your OU. Make sure that this person is employed by your OU, and -depending on your policy- a group leader/senior researcher.

• Research administrator Research administrator rights can only be granted by the system administrator. Contact rdmsupport@ubn.ru.nl to request that a user is added as research administrator of your OU.  

User manual 

Find the overview of OU members

Open

Add users to the OU and provide rights 

The system automatically ensures that only users with an account that is linked to the RU or Radboudumc (e.g., with a u- or z- number, but not an s- number) can be selected to become OU members. Prior to authorizing a user as an OU member, the research administrator must check whether the user is employed at the relevant research institute (i.e., OU).

Open

Remove users from the OU 

A user's authorization as an OU member may also be revoked. The typical reason for this is the termination of that user's contract with the RU or Radboudumc. When a user is no longer employed by the RU/Radboudumc, his/her u- or z- account becomes inaccessible and the authorization to access the RDR is automatically blocked. Nevertheless, we recommend that the research administrator removes that user as member of the OU to prevent unwanted future access in case the researcher would get a new appointment elsewhere at the RU or Radboudumc and the u- or z-account were to be reactivated.

Creating, managing, and removing collections 

Description of the task 

Data collections can only be created by a research administrator. Prior to initiating a collection, the research administrator checks the financial, legal, and organisational aspects, decides on the project identifier that will link the collection to a project (e.g., a BASS project number, another number that uniquely identifies the project in the OU, or a keyword that identifies the project), and ensures that the proper manager is made responsible for the collection.   

A researcher is only eligible to be a collection manager if he or she is employed by the RU/Radboudumc and has the corresponding credentials. Within these restrictions, it is up to the research institute to decide who is eligible to become collection manager. Some research institutes require that (at least one of) the collection manager(s) is a group leader or senior researcher to guarantee longer availability of the collection manager -since PhD students or postdocs tend to leave the university within a couple of years. The research administrator is responsible for ensuring that all collections in the OU have a manager that can be reached, including collections that are archived/published. This means that if a collection manager leaves the University or loses eligibility to be a manager in another way, the research administrator must assign a new manager (e.g., another group leader, the director of the research institute, or the data steward). To support the research administrator in this task, the RDR sends email notifications if there are collections without a manager.   

User manual 

Creating a data collection 

Log in to the RDR. Next, click the tab “Create Collection”. A new screen will appear where you can fill out information of the new collection (obligatory fields are indicated with an asterisk): 

1. Title of the collection. This is usually provided by the researcher that requested the collection. Please make sure that a title properly describes the subject of the data collection (for example, “Maria’s Thesis Collection” is not very descriptive, but “Genetic and epigenetic networks in Intellectual Disabilities” is). The researcher can still change the title later 

2. Project identifier. This forms the last part of the collection identifier: a unique code describing a collection. In the WebDAV client, collections appear only with their collection identifier (the title will not be visible). Project identifiers can be any combination of numbers, signs, and letters and the research institute can decide how to implement a project identifier (see the section “Developing a user policy for the RDR” below). Note that the project identifier, contrary to the collection title, cannot be changed anymore after collection creation 

3. Organisational Unit. The OU of the collection. If you are research administrator of just one OU, only that OU will be visible to you 

4. Managers: assign one or more managers to the collection. Click the button “Add manager” and find the user in the search bar. Only users that are eligible to be manager in the OU can be selected from this menu. Then press “select user”. You can still add/remove managers after collection creation and even after publication or archiving of the collection.  

5. Collection type: a DAC or RDC for internal archiving and/or sharing, or a DSC for external public sharing. Usually, the researcher specifies which type the new collection should be; a research administrator or data steward can provide help on this decision based on the research institute’s policy 

6. Audience: you may fill out one or more target audiences of the collection. You may also leave this field empty and let the collection manager choose one or more target audiences after collection initiation. The first option could be helpful if researchers have difficulties identifying a proper target audience themselves   

7. Collection quota. The size (in GB) that is reserved as the maximal size for the collection. Note that this has no direct consequences for costs: costs in the RDR are based on actual used storage space, not storage space that is only reserved for a collection. You can always increase the collection quota later, if the researcher has started working on the collection and needs more space. Your research institute may choose to fill out a standard collection quota for each collection -and increase it later if the researcher needs more space- or to let the researcher request a specific amount of quota when he or she requests the collection. Either way, it is important for the research administrator to keep track of extremely large collections, and encourage researchers that request such large collections to consider cost as well as research integrity and scientific value of the data when choosing to preserve substantial amounts of data for the long-term 

8. Preservation time. How long (in years) the collection should be kept as a minimum. Note that RU’s Research Data Management (RDM) policy dictates that research data underlying a publication must be stored for at least 10 years for research integrity purposes. Therefore, do not reduce the preservation time below 10 years unless there is a valid reason to do so. The RDR will never remove a collection’s metadata even after the retention period, but the data may be removed. The data will never be removed without notifying the collection manager. The collection manager can therefore always request that the research administrator extends the preservation period, should he or she find this desirable. Your research institute may choose to set a standard preservation time (for example of 10 years based on RU’s policy) or allow the researcher to choose a preservation time when he or she requests the collection 

9. Embargo until. For DSCs, an embargo can be requested by the researcher. The research administrator can set the embargo date at collection initiation, but also afterwards 

10. Make Metadata Public. If you select this option, the metadata of the collection (those that are visible in the “metadata” tab of a collection) will be made publicly available and registered in RIS as soon as the collection is archived or published. Note that for DSCs, this is always the case, and you cannot unselect this option. It is up to your institute to decide whether it is desirable that researchers register al their output, and therefore also the metadata of DACs and RDCs, in RIS. You can still edit this field after collection creation, but not anymore as soon as the collection is archived. Note that the collection manager cannot edit this field but should contact you instead if he or she want the field to be edited.  

11. Demo collection. To demonstrate the RDR to researchers and/or students of your institute or to external parties, it could be useful to create a collection for demonstration purposes only. By clicking this checkbox, you will create a demo collection. Demo collections never show up as a “featured” collection on the RDR’s home page, can be deleted at all times, are not findable via Google Dataset Search, and their metadata are never registered in RIS. Otherwise, demo collections have the exact same functionality as “real” RDR collections.   

Click “next”. Your collection has now been created. The collection manager(s) have automatically been notified by email about the collection creation (with a link to the newly created collection).   

Managing data collections 

The responsibility of managing data collections is highly variable and depends -in part- on the policy of your institute (see the section “Developing a user policy for the RDR” below).  

Some activities regarding collection management include: 

1. Creating new versions of collections. If a researcher needs to edit or fix an error in a published or archived collection, you can create a new version of the collection. The collection manager is not allowed to do this, but is informed that they can make a collection “editable” again by contacting their research administrator.

2. Ensuring that each collection of your OU has a collection manager that can be reached (see the description above).  

3. Helping researchers fill out the Data Use Agreement (DUA) for their restricted access collections. When researchers choose to use a restricted access collection, a DUA must be filled in with information from the data provider (i.e. the collection manager) and the data user (i.e. the data access applicant). Radboud University provides a DUA (RU-RA-DUA-1.0) but you could also choose to provide a more specific DUA for your research domain. Contact rdmsupport@ubn.ru.nl to add your OU-specific DUA to the RDR. The filled DUA must be signed by an authorized person of your research institute and that of the data user. This is usually the director of the research institute. To allow online signing of this document, you can make use of ValidSign (if you do not have an account yet, contact rdmsupport@ubn.ru.nl to obtain one). Start by entering all required information in the DUA together with the researchers in question. Then upload the document in ValidSign and provide contact information of those that need to sign the document as well as that of the collection manager. They will receive an email with a URL to sign the document. Once the document is signed, it will be send as a secured PDF to those people that were entered as contacts in ValidSign. Therefore it is important to enter contact information in ValidSign of not just the authorized persons that need to sign the DUA, but of the collection manager also (and perhaps yourself): in that way the collection manager (and you) will automatically receive the signed document and can store it appropriately. Unfortunately, the RDR does not offer support to store the signed DUA.

4. Keeping track of “dormant” collections: collections that have not been edited for more than a year but are still in the “editable” state. “Dormant” collections are undesirable, since the RDR serves to preserve research data on the long-term for internal re-use and scientific integrity and to publicly share research data. If the collection manager never closes a collection, these goals cannot be achieved. To identify “dormant” collections, you could use the excel or csv files containing the metadata and metrics of all collections of your OU (see below).  

5. Keeping track of usage of the RDR at your OU to develop/improve your research institute’s Research Data Management policy. To do this, you can download a csv or excel file containing metrics of all collections of your OU. This task could also be executed by a data steward of your research institute. 
   To help you with these activities, the RDR allows you to download a csv or excel file containing the metadata and metrics of all selected collections. In terms of metrics, the RDR tracks:  

   1. Views: a user (human or machine) visiting the collection detail page 

   2. Downloads: a user (human or machine) downloading a file from a collection 

   3. Data volume: the total data volume that has been downloaded for all files in a collection by users 

   4. Viewers: the number of unique users (human or machine) that visited the collection detail page 

   5. Downloaders: the number of unique users (human or machine) that downloaded a file from a collection 

Removing a data collection 

Data collections can be removed in two situations: 

To remove a collection, log in to the RDR. Then navigate to the landing page of the collection in question. On the top of that page, click the “Delete collection” button: 

If the collection contains data files or is not in the “editable” state, the delete button is disabled. Request the collection manager to remove any files and change the collection collection to editable. 

Access to data collections 

The research administrator carries the responsibility to have access to all data collections in the corresponding OU. This is required for scientific integrity, so that security checks with respect to fraud, plagiarism, and data construction can take place if needed. The research administrator is only allowed to access collections if there is a valid reason to do so, observing GDPR compliance and sensitivity of the data. All research administrators are accountable for keeping up confidentiality when processing data, in particular personal data.  

Developing a user policy for the RDR 

The research administrator may take up the responsibility to (help) develop an RDM protocol within their research institute that incorporates the RDR and guides researchers on how to use the RDR. This responsibility can, however, also be taken up by a Data Steward or director of the research institute. This responsibility can be carried out with help from the central Data Steward of the RDR.   

The RDR should be incorporated within the research institute’s RDM policy. This policy helps to direct researchers on what RDM solution (for example “werkgroepmappen”, SURFDrive, discipline-specific repositories, etc.) to use in a specific occasion. Please note that the RDR has been designed as a repository for research data. Data generated by students as part of a research internship do not belong in the RDR unless they are of high scientific value. In line with this policy, the research institute may decide that the RDR is only needed to publicly share datasets (i.e., to only use the RDR’s DSCs), to internally archive datasets (i.e., to only use DACs, RDCs, or both), or that both functionalities of the RDR may be used (i.e., to use DACs and/or RDCs as well as DSCs). The research institute’s policy should also specify where privacy sensitive data that are not research data (i.e., administrative data), such as the pseudonymization key and informed consent forms, should be stored. It is highly recommended not to store these files in an RDR collection since they are not research data and may have a very different preservation period than the research data of a collection.   

The institute’s policy on RDR usage should specify who is eligible to become an OU member. Furthermore, it should specify who is eligible as collection manager in an OU. In all cases an eligible manager should be employed by the RU/Radboudumc and by the relevant OU. It is recommended not to initiate a collection with only PhD students or postdocs as collection managers, since they tend to leave the university after only a couple of years, leaving the research administrator with the task to transfer collection responsibility to another researcher.   

Furthermore, the policy should specify whether OU members can be “Metadata Viewers” (see the paragraph Authorizing users in the Organizational Unit above). The advantage of allowing researchers to become metadata viewers is that it creates transparency within an OU on who is working on what and may therefore promote collaborations. However, in the case of extremely sensitive research it may be preferable not to allow researchers to become metadata viewers.   

The research institute should develop a procedure that allows researchers to request a collection. This can be as simple as providing an email address of the research administrator to the researcher. Another option would be that researchers can fill out a form on the research institute’s intranet to request a collection. The advantage of this method is that via a form additional information can be requested from the researcher that is needed to initiate a collection (i.e., project identifier, title, type of collection, manager, collection quota, preservation time, etc.).  

The research institute must decide how to implement the project identifier. This could be a keyword that helps researchers to identify their collection based on the collection identifier or a (financial or organisational) project number assigned to a research project by the administration of the institute. In the first example the researcher gets to choose the project identifier, and therefore must provide that information to the research administrator upon requesting a collection.   

Some researchers may request and use a collection, but leave it in the “editable” state without ever archiving or publishing it. This is not desirable and the research institute may implement a policy to keep track of these “dormant” collections and encourage researchers to archive or publish their data collections. 

The research institute’s policy should make decisions about open access data sharing. Under what circumstances can the researcher request an embargo? What recommendations are made to researchers on when to choose a Restricted Access DSC? Take into account that Radboud university’s policy dictates to share research data as open as possible.

Practicing in the acceptance environment 

For research administrators new to the RDR it might be useful to practice working with the RDR. This can be done in the acceptance environment (https://data-acc.ru.nl) which has the same functionality as the RDR. It is only accessible on the internal network, so use eduVPN if you are working form outside of the RU.  

 Abbreviations & important links  

RDR homepage: https://data.ru.nl 

RDR login page: https://data.ru.nl/login 

RDR help page:  https://data.ru.nl/doc/help 

RDR acceptance environment: https://data-acc.ru.nl

Be aware, you need a VPN connection to access the acceptance environment