The RDM Cycle

???

In this short video, we want to explain you the research data lifecycle and why it is important to consider all steps already before you start your project.

---

???

Data can be defined as any information or materials that are collected, observed, generated or used during the research process. Data can take various forms, including experimental results, numerical data, text documents, images, software code, surveys, and more.

---

???

Here, we want to make you aware of managing your data effectively through the research lifecycle. You need to handle data in a structured, organized and documented manner to ensure its quality, integrity and long-term usability. Make sure, your data complies to the FAIR principles: your data need to be findable, accessible, interoperable, and reusable.

---

???

Research data management includes all data handlings according to the research data lifecycle and the FAIR principles. Funders require data archiving for 10 years and more after a project has ended. If you do a proper research data management, you can find data faster, preserve knowledge, prevent data loss, you can much easier collaborate, share and reuse data, increase your own visibility, trace research results, reference by use of persistent identifiers, there will be overall a social benefit of open data, and specifications in research data policies will be fulfilled.

---

???

The research data lifecycle may consist of a varying number of phases because different disciplines emphasize different aspects of data management based on their specific needs, regulatory requirements, and operational complexity. This research data lifecycle consists of 7 phases. Reuse is the last but also the very first step of the cycle. Before collecting or generating new data, check if data which you can reuse is already available.

---

???

What is data re-use? Reuse means using data for other purposes than it was originally collected for. By reusing data it allows different researchers to analyse and publish findings based on the same data independently, it is a key component of the FAIR principles. To make them reusable, data needs to be well-described, curated and shared under clear terms. By reusing existing data you can obtain reference data, avoid doing new, unnecessary experiments, run analyses to verify that reported findings are correct, make research more robust by aggregating results obtained from different methods or samples and gain novel insights by connecting and meta-analysing datasets.

---

???

The first phase after checking for reusability is plan. Planning means defining the strategy for managing data and documentation. It is a very important step to avoid problems or unexpected costs related to data management. Here, you set the conditions for your research data to achieve the highest possible impact. It helps to identify risks in data handling and finding solutions and it facilitates data sharing, preservation and reuse. Planning your data handling strategy usually is documented in a Data Management Plan. Most funders require such a plan. Data management plans describe aspects of the data management process before, during, and after the project. Therefore they are living documents which are updated over the project run time. There are many open source tools available to help you writing a data management plan, e.g. RDMO or DMPonline.

---

???

During the collection phase, you gather information about specific variables of interest for example instrumentations and parameter of machines, which collects data. Collecting these parameters ensures the quality of your data. Check for re-use existing data sets such as earlier collected datasets and reference data. It is important to capture the provenance and to define experimental design collection plan, e.g. repetitions, controls. If you are working with sensitive or confidential data, check data protection, security issues and permissions. Define in this phase how to store data, for example in which format and which volume. You need to find suitable repositories where to store data and identify suitable metadata standards for your discipline.

---

???

During data processing, data are made compatible for integration with each other. Convert your data into the desired format and prepare them for downstream analysis. Check the quality and discard bad or low quality data in order to create clean, high-quality dataset for reliable results. Sensitive data should be pseudonymised and anonymized which means removing identifying data. Here again it is very important to accurately document every step during data processing because it is key for the reproducibility of your results.

---

???

During this phase of analysis the collected data is explored. Here, try to understand the messages contained in a dataset and apply mathematical formula or models to identify relationships between variables. New knowledge and information is generated. Remember, the analysis workflow applied to a dataset needs to comply with the FAIR principles: it need to be reproducible. Publish your analysis workflow according to the FAIR principles as well as your datasets.

---

???

Preserving data ensures safety, integrity and accessibility of data for as long as necessary. Data preservation is more than just data storage and backup since data can be stored and backed up without being preserved.
Data preservation prevents data from becoming unavailable and unusable. Update your software and hardware as well as the long term data repositories. Ensure that data is organised and described with appropriate metadata and documentation to be always understandable and reusable.

---

???

Data sharing means to make your data known to others. But data sharing doesn’t mean open data or public data. There is the possibility to share data with restricted access or even closed access. Data sharing can be done at any time during the research data life cycle. It is a prerequisite for making research reproducible. It is good research practice to ensure that data underlying research is preserved & made available to the research community and society at large.

---

???

Here we are again at the last and first phase of the research data lifecycle. Reuse data. To reuse data, data need to be stored according to the FAIR principles. Data storage can be done at any time point. For long-term storage there are different option at the university, on state level, Germany-wide, European alternatives and beyond. To decide for a storage option you find some hints in the storage video.

---

???

Thank you for watching the video.