Humanity and the world around us generate an intense amount of data, and that data can provide answers to questions and solutions to problems. However, it doesn’t matter how much data you have access to or how good-quality that data is, if it’s not being managed properly.
This is especially true when it comes to research. A study cannot come to the correct conclusion if it doesn’t manage its data properly. Research data management (RDM) is key to ensuring the success of any study or experiment.
The Stages of Research Data Management
Let’s look at the lifecycle of data and how proper research data management principles can be applied along the way.
- Collection: Data should be collected in a consistent manner, in terms of how it is gathered and what data points are recorded.
- Entry: If data is being recorded manually, it should be entered carefully into the system, ensuring that there are no typos or other errors. All relevant data points should be recorded. Formatting should be consistent.
- Storage: Data should be stored in a secure manner and in compliance with any relevant regulations.
- Masking: In some cases, such as when dealing with human subjects, some data may need to be masked in order to ensure that sensitive personal information is protected, or that results are not skewed based on perception.
- Sharing: If data is to be shared with other researchers within or outside the lab, the recipients should be briefed on any manners of compliance surrounding the data and must demonstrate good research data management practices as well.
Why RDM Matters
Good stewardship of research data pays off in the long run. An initial commitment to properly collecting, recording, and storing data saves time down the road. Without good data management practices, researchers can find themselves wasting time sorting through incomplete, incorrect, and corrupted data. Hours can be lost to normalizing inconsistent entries, tracking down missing information, and more.
Even worse problems can arise if issues with data pass unnoticed, leading to analysis being done on incorrect and incomplete data. This can skew the results of an experiment, leading to researchers coming to the wrong conclusion.
In a worst-case scenario, improperly managed data can be rendered completely useless, effectively hamstringing a research project. Researchers may be forced to start the experiment all over, or even abandon it unfinished if there is no funding to perform a new round of data collection.
Research data management is also important at later stages of the data’s life cycle, however. Data which is improperly stored may be subject to being corrupted, lost, or accessed by unauthorized persons or organizations. Researchers have a responsibility to ensure that any sensitive or proprietary data in their possession is stored safely and securely.
Fortunately, it is not particularly difficult to practice good RDM. Researchers simply need to take the time to learn the best practices, rules, and regulations surrounding the project and then take the time to carefully adhere to them. Issues mainly arise when researchers are rushed or are forced to cut corners due to a changing deadline, new experiment parameters, or budgetary constraints. Sometimes, too, honest mistakes can be made. Researchers are only human.
