RDM & Top 5 Preventable Research Disasters

Next, on this section of the website you will find a carefully curated selection of support and guidance according to each stage of the research and rdm cycle.

What is Research Data Management (RDM)?

In short, RDM is taking good care of your data during and after your research. The importance of good RDM becomes clear when you look at the TOP 5 research disasters that could have been easily be prevented by good RDM:

Legal problems
Money issues
Journals refusing to publish your article
Data loss
Wasting a lot of time & being frustrated

The following info slide deck illustrates how poor RDM can lead to those kind of research disasters and, as a result, how you can prevent such research disasters regarding your own research projects: <info slide deck = under construction>

One aspect of good RDM is be sure that your data will be as "FAIR" as possible. During your research you should already take measures to make your data FAIR. In additon, RDM is also about managing your data during your research project, like proper folder naming and management of acces rights for these folders (view only/ edit rights), making use of well-known commenly used standards (device settings, annotations, software codes, etc.), well regulated version control of files during your research, and for example documenting your metadata, sharing.

Similar to a research project, RDM also can be visualized as a continuous life-cycle.

Next, on this page you will find the following RDM specific content:

SHORT and SWEET

What is Research Data Management (RDM)?

In short, RDM is taking good care of your data during and after your research. The importance of good RDM becomes clear when you look at the the TOP 5 research disasters that could have been easily be prevented by good RDM:

Legal problems
Money issues
Journals refusing to publish your article
Data loss
Wasting a lot of time & being frustrated

Similar to a research project, RDM also can be visualized as a continuous life-cycle.

On this section of the website you will find a carefully curated selection of support and guidance according to each stage.

kick-starting your research data management

RDM contributes to Open Science, FAIR data, and reproducible data and software. In the following tabs you will find quick help on how to improve your data management, research data exposure, as well as some pointers on legal matters and FAIR data.

How to your data files

1. Create a coherent folder and file system

Step 1: Create a coherent folder and file system

Step 2: Establish version control

Step 3: Write readme files

Creating a logical folder structure with a few top-level folders to divide between research and administrative content, and establishing a number of sub-folders at the beginning of a new project helps with managing, sharing, and saving files over time.
Adding versioning numbers on higher level folders, and all files helps keeping the overview and statuses of their contents independent from time-stamps in the file-directories.
Finally, adding simple text-files with a minimum set of information (aka readme file) about all subordinate folders and files for each top-level folder will increase the comprehensibility for you and your colleagues in the course of the project and beyond.

Example folder structure with naming conventions

click to enlarge

Image reference: https://www.wur.nl/en/Value-Creation-Cooperation/WDCC/Data-Management-WDCC/Doing/Organising-files-and-folders.htm

2. Establish file naming conventions

Step 1: Start adding the last modification date in a standardised syntax in every new file and folder title

Step 2: Add version numbers in in every new file title

Step 3: Create and add abbreviations and identifiers to every new file and folder title

From now on, add the last modification date to every new file title you write. The international standard to write a date is YYYY-MM-DD (year, month, date), out of convenience the dash can be left out.
Example: Test_190517
Additionally, adding a version number with integers will quickly reveal what the latest version of a file is.
Example: Test_V5_190518
Creating abbreviations and identifiers for processes or applied methodologies enhances the quick comprehension of the file contents in the directory without the need to open the actual file.
Example: <analysis_name>.filtered_variants.snpEff_GoNLv5.vcf

example file naming convention and folder structure

click to enlarge

Image reference: https://www.ukdataservice.ac.uk/manage-data/format/organising

3. Improve the project documentation

Step 1: Write about the relationship and context between files within the folder and file structure.

Step 2: Add information about vocabulary, terminology and definitions that you have applied.

Step 3: Continue with explaining the chosen methodology.

For starters, by offering some contextual information how files and folders within a project directory relate to each other, users of these files can grasp their relationship to each other more easily.
By adding explanation to applied vocabulary, terms used, and domain specific definitions, the framework in which your research happened is more transparent and reproducible.
Lastly, specifying which methodologies were applied for which data-files will complete the picture of your research that will enable authentic re-usability.

click to enlarge

Image reference: https://weblog.wur.eu/openscience/documenting-research-data-along-way-tips-tools/

4. Create a regular data back-up routine.

Step 1: Decide and adhere to a regular back-up habit.

Step 2: Consider automating this regular back-up routine.

Step 3: Store your back-up data ideally in 3 different locations.

Depending on the uniqueness of your data and the sensitively level, the intervals of transferring your important files to a back-up location might be different. At least a weekly back-up is a good start.
By applying back-up management software, the effort of taking care of regular back-ups can be reduced.
Ideally, your regular back-up will be spread across different and independent and secure storage locations. Besides your working computer, institutionally provided storage drives are a good option for another back-up location.

Do you know about SurfDrive?
To all researchers and students of the UMC Utrecht SurfDrive is accessible right away; it offers an online interface and a desktop-client with inter-device synchronisation.

Image reference: https://www.flaticon.com/authors/flat-icons

How to

Sign up for your freely available unique authors ID. In addition to most of the publishers these days requiring this ID-type already, the ORCID-ID helps you being individually identified as creator and will list all outputs that is related to this ID.

Publishing other scientific material such as poster, presentation slides, graphics, software, and videos will also count as publication and will accumulate in your publishing record.

Read more about the ORCID-ID and register: ORCID.org

Image reference: https://orcid.org/

Reconsider sharing finished data-sets and other files via email upon request, or placing it as individual file on your project website: rather self-publish it on a general data repository such as Zenodo, to receive a persistent link and therefore all benefits of citations and discoverability with it.

Another advantage: you can make the publication underlying data available as its own entity with a DOI. Set-up properly, both the publication and the data-set will have their own persistent links and refer to each other. That can possibly lead to a scenario, where your data is independently cited and re-used as much and perhaps more than the related publication.

Image reference: https://en.wikipedia.org/wiki/Digital_object_identifier

The best way to cite your own data is to refer to it in-text (not in the footnote!) or in the reference section. That way it can be detected and picked up by citation tools used by publishers and library indices that feed into your citation score.

Instead of sharing your data files via a website with a URL that might be unstable over time, choose to publish your data via a dedicated repository or archive that will attach a persistent link to that data-set. The most common persistent identifier is the DOI.

Any type of publication with a persistent link will accumulate more citations over time due to accessibility and discoverability on the web.

The way science is conducted and documented is changing, so is the academic reward system. A future is on the horizon where data and software publications, as well a plethora of other output forms of scientific work next to the classic and prestigious publications are recognised as valuable and impactful body of works.

One popular employed by many universities across the globe is called ALMETRICS. Besides the classic citation tracking it enables the attention and impact detection and tracking of all sorts of outputs ranging from tweets, over opinions in comment sections, to policy papers. The impact of your work on the wider society can now be traced and real time interaction with peers facilitated.

Due to ORCID-ID’s and DOI's this transparency and historical hashing of interactions is possible.

Image reference: https://www.altmetric.com/products/altmetric-badges/

RESEARCH DATA MANAGEMENT KNOWLEDGE CLIPS

a few short introductory videos

In the five tabs below you can watch a knowledge clip in video format that will quickly introduce you to the main topics per category:

1. 1. data management planning (DMP)
  2. budgeting for research data management (BUDGET)
  3. research data management essentials (RDM)
  4. scientific exposure and impact (EXPOSURE)
  5. GDPR and informed consent (GDPR).

Your research project from start to Finish

the ins and outs of research data management and data stewardship alongside your research project

take me back to the top

Consultancy service

For expertise advice on your research and the possibilities UBEC offers to get the most out of your data: request a meeting by emailing us. Experimental design, data management, bioinformatics analysis, results and follow-up experiments are discussed. The facility manager ensures that experts from participating organizations are present during this meeting.

You can request a meeting at bec@umcutrecht.nl.

This Global Block does not have any content.