Research Data Management: What is RDM?

1. What is Research Data Management?
2. FAIR Principles
3. What Are Research Data?
   1. Types of Research Data
4. Data Related to Research
5. Definitions

Research data management (RDM) covers practices associated with the collection, documentation, storage, sharing, and preservation of all research data. It involves the active organization and maintenance of data throughout the entire research process which includes the periods before, during, and after the active research phase. RDM eases the facilitation of the research process during active research, while also promoting greater data accessibility and supporting research excellence.

Research data management strives for data that meets the FAIR Guiding Principles, which means that data that is Findable, Accessible, Interoperable, and Reusable. In practice, this means asking a series of questions about your research data:

• Can I find the data?
• If I can find the data, is it accessible?
• If it's accessible, can I open the data file?
• If I can open the data file, can I understand the data well enough to reuse it? Does the data have a license that allows for reuse?

A common misconception is that research data are limited to the sciences. In reality, they span all academic disciplines but vary widely depending on the project, domain and context. Research data can include spreadsheets and numbers, text, videos, images, historical artifacts, code, and more. See formal definition below.

Research data are also not a mere by-product or a steppingstone to publication. They often outlast the outputs they support. They serve as evidence to validate and verify findings and can be reused for future research, teaching, and other purposes. 

That’s why research data deserve to be treated as first-class research outputs and why RDM is essential. 

Types of Research Data 

Research data can be classified in different ways, for example based on their: 

• Content: numerical, textual, audiovisual, multimedia, etc. 

• Format: spreadsheets, databases, images, maps, audio files, (un)structured text, etc. 

• Mode of data collection: experimental, observational, simulation, derived/compiled from other sources 

• Digital (born-digital or digitized) or non-digital  (analogue) nature (e.g. paper surveys, notes…) 

• Primary (generated by the researcher for a particular research purpose or project) or secondary nature (originally created by someone else for another purpose) 

• Stage in the research data lifecycle: raw (directly collected or generated), cleaned (errors corrected, standardized, de-identified), or processed (transformed or analyzed) 

In addition to managing the research data itself, RDM also involves managing documentation that ensures data can be understood and reused effectively. 

Examples of what to document include: 

• Methodology and protocols: Descriptions of research methods, procedures, and workflows 

• Data structure and organization: Dataset structures, database schemas, data dictionaries, and codebooks 

• User support materials: User guides, notebooks, and questionnaires or interview schedules 

• Technical specifications: Equipment settings and instrument calibration details 

• Contextual references: Links to related datasets, sources, software, or code 

UWaterloo researchers should include documentation in a README file.