Reference
Core Terms
- RDM (Research Data Management): Planning, organising, storing, documenting, sharing, and preserving research data across the project lifecycle.
- DMP (Data Management Plan): A living plan describing how data will be collected, stored, documented, shared, retained, and preserved.
- FAIR: Data principles that aim to make data Findable, Accessible, Interoperable, and Reusable.
- Metadata: Descriptive information that helps data be understood, located, and reused.
- Data lifecycle: The sequence from planning and collection through analysis, sharing, preservation, and reuse.
- Personal data: Information relating to an identified or identifiable person.
- Special category data: More sensitive personal data, such as health, genetic, biometric, race, religion, or sexual orientation data.
- Lawful basis: The legal justification for processing personal data under UK GDPR.
- DPIA: Data Protection Impact Assessment used to identify and reduce data protection risks.
- Data controller: The organisation that decides why and how personal data are processed.
- Processor: A party processing data on behalf of the controller.
- Ethics review: Formal review of research involving humans, personal data, or other significant risks.
- Retention: How long data are kept.
- Preservation: Longer-term keeping of data so they remain usable and understandable.
- Version control: A way of tracking changes to files, code, documents, or datasets over time.
- Reproducibility: The ability to understand, check, and rerun a workflow or analysis.
Institutional Platforms
| Platform | Typical use |
|---|---|
| Microsoft 365 | Day-to-day collaboration, communication, writing, and spreadsheet work |
| OneDrive | Individual working storage with sync and version history |
| Teams | Group communication, meetings, and shared project working space |
| SharePoint | Collaborative document storage with permissions and longer-lived team spaces |
| RIS (Worktribe) | Research lifecycle tracking, outputs, grants, and linked researcher information |
| UniCore | Finance, HR, procurement, and operational research administration |
| ORCID | Persistent researcher identifier used to maintain a consistent research identity |
| TRE Secure Storage | Restricted environment for sensitive or confidential data |
| CPS Research Storage | Group-managed research storage for backed-up project data |
Quick Decision Guide
Where should this work live?
| Situation | Usually the best fit |
|---|---|
| Personal draft notes or early working files | OneDrive |
| Shared project documents used by a team | Teams / SharePoint |
| Sensitive or confidential research data | Approved secure storage such as TRE Secure Storage |
| Proposal, output, or institutional research record | RIS (Worktribe) |
| Finance or procurement workflow | UniCore |
| Public researcher identity and linked outputs | ORCID plus institutional systems |
Practical Good Practice
File and folder naming
Use names that are consistent, sortable, and readable.
2026-03-23_interview-theme-summary_v01.docx
project-alpha/
raw-data/
cleaned-data/
documentation/
outputs/
Helpful patterns:
- use dates in
YYYY-MM-DDformat - avoid spaces if files move across systems regularly
- use meaningful version labels such as
v01,v02,final-reviewed - separate raw, cleaned, and derived data
Minimum documentation to keep
For most projects, try to retain:
- who created or updated the file
- when it was created or changed
- what the file contains
- how values, variables, or categories should be interpreted
- what processing or transformation has already happened
- where the authoritative version is stored
Metadata prompts
If you are not sure what metadata to record, start with:
- title
- creator or owner
- date
- format
- location
- description
- methods or source
- access restrictions
- reuse conditions
Governance and Compliance Reminders
- Ethics approval and data protection are related but not identical.
- Consent is important, but it is not automatically the lawful basis for research processing.
- Personal data can include indirect identifiers, not just names and email addresses.
- Sensitive data requires both stronger justification and stronger safeguards.
- A good DMP should match your real storage, access, sharing, and retention practices.
Data Quality and Analysis Prompts
When reviewing a dataset or chart, ask:
- What exactly does each row represent?
- Are categories used consistently?
- What is missing, ambiguous, or mixed together?
- What would a collaborator misunderstand without explanation?
- Does the chart help interpretation or distort it?
- What action is the audience expected to take from this output?
AI Use in Research
Use AI most safely when you define:
- the task
- the context
- the source boundary
- the output format
- the verification step
A useful prompt pattern:
Help me with [task] for [context]. Use [source or boundary]. Return the result as [format]. Flag uncertainty and do not invent evidence.
Before using AI on research work, ask:
- Is the material safe to paste into this tool?
- Do I need to use only approved or institutionally permitted services?
- What part of the answer must I verify manually?
- Am I asking for support, or am I trying to outsource judgement?
End-of-Course Checklist
By the end of the course, you should be able to:
- choose more appropriate tools for collaboration and storage
- explain key governance and RDM concepts in practical terms
- design a clearer folder structure and documentation approach
- critique charts and data summaries more confidently
- turn an analysis result into a clearer action or recommendation
- use AI in a more selective, source-aware, and defensible way