Introduction to research data management
The nature of research data can vary across disciplines. As a result, defining research data can be challenging. Deakin broadly defines research data as:
All data collected, observed, or created by researchers in the course of their work, for the
purposes of analysis to produce original research results.
Read more about Deakin's Research Data Management Procedure.
Research data formats
Research data can come in a range of formats. Click on the sections below to explore some examples:
Primary vs. secondary data
Research data can be broadly divided into primary and secondary data.
Primary data is the original data derived from your research endeavours. Secondary data is data derived from your primary data. Often the distinction between primary and secondary data may be less than clear.
You will collect and create both types of research data when conducting research. It is essential you have plan for the management of all types of data and primary materials used in your research.
Common examples include:
| Primary materials | Primary data | Secondary data |
|---|---|---|
| Interview schedules | Interview audio recordings | Nvivo interview transcripts |
| Purchased laboratory reagents | Investigational product | Product analyses |
| Research animals | Tissue samples | Stained slides |
| Validated questionnaires | Completed paper and pencil questionnaires | SPSS data files containing raw data and calculated variable summary scores |
In the above examples, the creation of secondary research data is an intermediary step between the collection of primary research data and the dissemination of research findings.
Digital vs. non-digital data
A research project may include data that is both digital and non-digital, or data which may originally be non-digital that is later digitised. For example, printed self-report questionnaires that are later entered into a data analysis program such as SPSS, R or Excel.
Other examples include:
Digital data:
- Digital chemical analyses (e.g. spectroscopy)
- Digital photographs
- Digital medical images (e.g. DEXA scans)
Non-digital data
- Biological materials
- Original artworks
- Completed paper-based questionnaire responses
Your research data management plan should account for all types of data.
Research data lifecycle
Research data often outlives the projects that generate them, and can hold long-term value for future research, collaboration, and innovation. The research data lifecycle outlines the key stages involved in managing data effectively from the initial planning phase through to long-term preservation and discovery.
Click on the plus (+) icons below to explore each part of the research data lifecycle.
Interactive research data lifecycle image overview
An image of the research data lifecycle is displayed which is a circle with 6 headings on it representing the stages of the lifecycle. There are 6 icons (1 beneath each heading) that users can click to reveal information about that stage in the cycle.
Revealed information
- Data management planning:
- Design research
- Plan data management
- Plan consent for sharing (Ethics application)
- Plan data collecting, processing, protocols and templates
- Explore existing data sources
- Active data management:
- Collect data
- Capture data with metadata
- Acquire existing third-party data
- Enter, digitize, transcribe and translate data
- Check, validate, clean and anonymise
- Derive data
- Describe and document data
- Manage and store data
- Analyse and interpret data
- Produce research outputs
- Cite data sources
- Appraisal and risk assessment:
- Research data security classification
- Cybersecurity risk management
- Preservation:
- Migrate data to best format/media
- Store and backup data
- Create preservation documentation
- Preserve and curate data
- Manage legal retention periods
- Secure disposal of data if appropriate
- Access and publishing:
- Establish copyright/license
- Create user documentation
- Create discovery metadata
- Select appropriate access to data
- Publish/share data
- Promote data
- Discovery:
- Conduct secondary analysis
- Undertake follow-up research
- Conduct research reviews
- Scrutinise findings
- Use data for teaching and learning
Research Data Lifecycle. Adapted from UK Data Service (2019) YouTube video (1m 40s) and Rans, J and Whyte, A. (2017). 'Using RISE, the Research Infrastructure Self-Evaluation Framework' v1.1 Edinburgh: Digital Curation Centre. Available online: www.dcc.ac.uk/guidance/how-guides used under Creative Commons Attribution license.
Why RDM matters
It's good practice to have a clear plan before you begin collecting or working with research data. A research data management plan (RDMP) also helps you meet a range of compliance requirements.
University policies
- Deakin University's Research Conduct Policy (clause 19-37)
- The Australian Code for Responsible Conduct of Research
- The National Statement on Ethical Conduct in Human Research
- Copyright and licensing conditions
Funder requirements
Many national and international funding bodies now require a formal data management plan as part of their application or reporting processes. These include:
- The Australian Research Council (ARC)
- The National Health and Medical Research Council (NHMRC)
- The US National Science Foundation (NSF) – Data Management Plan Requirements
- The UK Medical Research Council (MRC) – Data Management Plan FAQs and Guidance.
Benefits of good RDM
There's a growing expectation that data generated from publicly funded research should be made openly available. Governments and research organisations worldwide are increasingly adopting open data by default.
Planning for data sharing and publication from the beginning of a project makes it easier to meet these expectations. It also allows you to take advantage of the benefits of openness, including increased impact, visibility, and potential for collaboration.
