· By Thesisroom
Data Management Plan for a Thesis: Sections, Steps and a Template
What a data management plan covers, when Southampton, Reading, Sheffield and UWE Bristol ask for one, consent wording, storage rules and a template to print.
This guide has a printable version: open the printable data management plan template
A data management plan (DMP), says the University of Maryland’s guide, “usually describes the data, code, and other research products you will produce and how you will format, document, store, and preserve them.” For a thesis, the University of Southampton adds who will own and have access to the data, and it asks every first-year doctoral researcher for a preliminary plan within three months.
What is a data management plan for a thesis?
A DMP is your written answer to one question: what happens to each piece of research material from the day you collect it to the day you graduate. Southampton calls it “a living document” that you update as the research develops.
Southampton’s guide lists what a plan describes:
- What data will be created
- What policies will apply to the data
- Who will own and have access to the data
- What data management practices will be used
- What facilities and equipment will be required
- Who will be responsible for each of these activities
“Data” is wider than spreadsheets. Southampton defines it broadly as “Material intended for analysis”, and its examples include research notes on a text and a notebook describing the context of photographs. Interview audio, archive images and field notes all belong in the plan.
Do thesis students have to write a data management plan?
Some universities, including the four below, require one from doctoral students. Where yours does not, a funder or an ethics committee may still ask for one. Rules differ between universities, so check your own graduate school’s code of practice before you rely on any example below.
Four UK universities that publish a rule:
| University | Who it applies to | When the plan is due |
|---|---|---|
| University of Southampton | All first-year postgraduate researchers | A preliminary DMP in the first three months, as part of the first three-month review, then updated before every progression review |
| University of Reading | Postgraduate researchers registered in or after September 2025, including those using only secondary data | A supervisor-approved DMP as part of confirmation of registration, then updated for each annual review |
| University of Sheffield | New starters from 2019/20 onwards (a mandatory milestone) | DMP submission 9 to 12 months after the start, in the full-time PhD timeline |
| UWE Bristol | All researchers, including doctoral researchers | Before collecting any data or using any data for research |
Reading also fixes the format: “It is not permitted to submit the DMP using any other template.” Its ethics guidance asks for a separate, narrower DMP with ethics applications, so one student can need two plans. Our PhD timeline guide shows where Sheffield’s DMP sits among the other first-year milestones.
For a Master’s thesis, check your department’s handbook first. Reading’s ethics DMP applies to taught postgraduate students only if their research is expected to lead to peer-reviewed publication, though its Schools may still ask for one.
In the US, the National Science Foundation, a funder, says its two-page data management and sharing plan “is a required part of a proposal”. If your funding comes from your advisor’s grant, ask whether that grant’s plan already covers the data you will collect.
If you are still writing an admission proposal, a full DMP can wait unless the department asks for one. Our guide to the PhD research proposal format lists funders that ask for one at the funding stage.
What sections does a data management plan cover?
The Digital Curation Centre’s how-to guide groups a plan into six themes: data types and formats, ethics and intellectual property, access and sharing, short-term storage, deposit and preservation, and resourcing. The table turns each theme into the question a thesis plan has to answer, with the matching item from NSF’s general guidance for US readers.
| DCC theme | The question for your thesis | Closest NSF item |
|---|---|---|
| Data types, formats, standards and capture methods | What will you collect or reuse, in which file formats, and what notes will explain it? | Types of data and materials; standards for data and metadata |
| Ethics and intellectual property | Who owns the data, what did participants agree to, and what must stay restricted? | Part of “Policies for data access and sharing” (protection of privacy, confidentiality, security, intellectual property) |
| Access, data sharing and reuse | Who can see the data during and after the degree, and where? | Policies for access and sharing; reuse, redistribution and derivatives |
| Short-term storage and data management | Where does the working data live, how is it backed up, and who looks after it? | Not a separate item on NSF’s list |
| Deposit and long-term preservation | Which data will you keep after the degree, and in which repository? | Plans for archiving and preserving access |
| Resourcing | What extra storage, software or training will you need? | Not a separate item on NSF’s list |
Use whichever template your university or funder names. If none is named, two planning services build a plan from funder and institutional templates. DMPonline is provided by the Digital Curation Centre. The DMPTool is described in UC Irvine’s graduate research guide as “a free resource” with customized templates.
How do you write a data management plan, step by step?
Write the plan before you collect data and before you apply for ethics approval, then keep it current. Reading tells its research students to start a DMP in the first year, before collecting data or applying for ethical approval.
- Find the template you must use. Check your graduate school’s code of practice, your funder’s terms and your ethics committee’s forms. Reading, for example, accepts only its own PGR template at confirmation.
- List every kind of material. Include primary data, secondary datasets, field notes, consent forms and your reference library. Southampton asks you to check the terms of any secondary data, because they can limit where you store and analyze it.
- Estimate the volume and choose formats. Southampton notes that documents and a reference library usually run to megabytes or gigabytes, while high-resolution images can need terabytes. Prefer open formats, or software that exports to them.
- Set file names and version numbers. Southampton advises against “draft”, “latest” or “final” in file names. Use a date in YYYY-MM-DD form or a version number such as v1.0, raised for major changes.
- Settle personal data and consent. Decide what identifies people, how you will anonymize it, and what your consent form will say about sharing. Southampton says it is important that your DMP and your ethics application agree.
- Choose storage and backup. Name the drive or service for each kind of data, who can open it, and how it is backed up. The storage section below gives the rules to check.
- Decide what you keep, share and destroy. For each kind of data, say whether it will be deposited, kept privately or deleted, and when. Southampton asks you to retain or discard data in line with your ethics approval.
- Name who does what, then review it. Say who is responsible for each task and send the plan to your supervisor. Southampton suggests recording review dates and a version table in the plan itself.
What does a filled-in data management plan look like?
This is a short plan for an invented thesis, written to the six DCC themes. The project, schools and numbers are invented to show the level of detail that makes a plan usable.
The invented thesis is a PhD in education on how secondary-school teachers use written feedback. It draws on interviews with twenty teachers at two schools, plus each school’s published inspection reports.
- Data types and formats: Twenty interview recordings as audio files, transcripts saved as .docx and exported to plain text for analysis, a codebook as .csv, and the two inspection reports as PDF. Files are named in the pattern INT07_YYYY-MM-DD_v1.0, and a readme file explains the codes.
- Ethics and intellectual property: The school research ethics committee approves the study before the first interview. The consent form says anonymized transcripts will be preserved and shared with other researchers under controlled access. The inspection reports are already public, so the deposit cites them instead of copying them.
- Access and sharing: During the degree, only the student and two supervisors can open the project folder. Anonymized transcripts, the codebook and the readme go to a repository under controlled access before the thesis is submitted, embargoed until the thesis is public.
- Storage and backup: Recordings move from the recorder to the university network drive on the day of each interview, then are deleted from the recorder. An encrypted external drive holds a second copy of the anonymized transcripts and codebook only. A test restore from backup is run once a term.
- Deposit and preservation: Audio files and the key linking pseudonyms to names (personal data, not the research data being shared) are deleted once transcripts are checked and anonymized. The thesis cites the deposited data in a data availability statement. Consent forms are kept for the period the university’s retention rules set.
- Resourcing: No extra storage is needed. The student completes the university’s data management training in the first term and asks the library’s research data team to read the plan.
Each answer names a place, a format, a person or a point in time, so a supervisor can check it. “Data will be stored securely” gives a reader nothing to check.
How should a plan handle personal data, consent and anonymization?
Name what identifies people, say how and when you remove it, and make the consent form promise exactly what the plan says. Reading’s ethics reviewers read the DMP against the participant information and consent documents.
Personal data
Reading defines personal data as “any information relating to an identified or identifiable natural person”. It warns that pseudonymized or link-coded data should still be treated as personal data. Its most effective route to fully anonymous data is to remove the linked ID or destroy the link table.
US readers will meet the same questions in an IRB application. Maryland’s guide notes that if you have IRB approval, you may be able to adapt your plan’s data description from your IRB content. If your proposal’s ethics paragraph already covers consent, anonymity and data storage, as our guide to proposal methodology recommends, build on that text.
Consent wording
Your consent form decides what you can share later. Reading advises an “I understand” statement rather than “I give permission”, because permission can in principle be withdrawn. Reading suggests a short statement for open data. “I understand that the research data collected from me in this study will be preserved and made available in anonymised form, so that they can be consulted and re-used by others.”
The University of Leeds suggests wording that begins “I agree” for data shared on a restricted basis. The two universities differ, so use the wording your own ethics committee provides.
Reading also warns against one promise in the participant information sheet: do not undertake to destroy research data, or to keep it within the project. Either promise blocks sharing later.
Anonymization
The UK Data Service says anonymization “should not be treated as a single technical step carried out at the end of a project”. It pairs it with clear communication to participants and with access controls.
Southampton asks for data to be anonymized or pseudonymized as early as possible after collection. Reading suggests an anonymization plan written early, listing each identifier you collect and the action you will take. You then log each action as you complete it.
Where should you store and back up thesis data?
Keep working data on storage your university manages, hold more than one copy, and test that you can restore it. What counts as approved storage differs between universities, so read your own IT pages.
- Personal data: Southampton says personal and special category data generally need university network drives with added protection, not personal computers or cloud storage such as OneDrive. Treat this as Southampton’s rule and check yours.
- Transfers: Southampton says transfers must be secure, with encryption where necessary. It does not recommend USB or external drives (encrypt them if you must use one) and advises against sending files by email.
- Copies: UC Irvine’s storage and backup guide recommends a “Rule of 3”: the original copy, a second local copy and a remote copy.
- Testing: UC Irvine also says to test that you can fully restore your data. It notes that mirroring is not backup, because changes to the main files are copied to the mirror too.
- Priorities: The DCC guide says “Back-up of your unique data is more critical than copies of secondary data.”
What happens to your thesis data when the degree ends?
At Reading and UWE Bristol, you deposit the data that supports the thesis, keep what the rules say to keep, and securely delete the rest. Rules differ between universities, so your plan should name your own university’s rule before you collect anything.
- University of Reading: Data supporting the thesis should be deposited in a repository before you submit the thesis for examination, and cited in a data availability statement. Its archive can embargo a dataset until the thesis is public. Our guide to citing a dataset in APA 7 shows the reference format.
- UWE Bristol: Doctoral researchers should deposit the complete core research data relating to the thesis upon submission and before graduating. UWE normally restricts access to core data for a maximum of two years.
- Consent forms (Reading): Keep them as long as any personal data are kept, and otherwise for at least 5 years after the project ends. Personal data should be destroyed when no longer required.
- What to keep (Maryland): Funding agencies and journals generally do not expect you to retain all your data. Your plan should explain why you keep some data and discard the rest.
- Deleting (Southampton): Digital data may need to be destroyed so that it cannot be forensically recovered. Ask your IT service for the method.
Add the deposit to your thesis submission checklist, next to the final copy of the thesis, so it is not left until after graduation. At Reading, that deposit comes before examination, not after the viva.
Printable data management plan template
Write one answer per row, then tick the checks before you send the plan to your supervisor.
| Section | Your answer |
|---|---|
| Project title, your name, supervisor, version number and date | |
| Template required (graduate school, funder, ethics committee) | |
| Data you will collect: type, source, format, rough size | |
| Secondary data you will reuse, and its terms of use | |
| Documentation: readme, codebook, file naming, versions | |
| Personal or sensitive data, and how you will anonymize it | |
| Consent wording about storage and sharing | |
| Who owns the data and any intellectual property | |
| Working storage location, and who can open it | |
| Backup copies: original, second local, remote | |
| What you will deposit, where, and when | |
| What you will keep privately, and for how long | |
| What you will destroy, how, and when | |
| Training, storage or software you still need | |
| Who is responsible for each task | |
| Next review date |
- The plan uses the template my university or funder names
- Every kind of data, including notes and secondary data, is listed
- File naming and version rules are written down
- The plan matches my ethics application and consent form
- Personal data sits only on storage my university approves
- I have restored at least one file from backup
- Each dataset has a decision: deposit, keep or destroy
- My supervisor has read this version
- A review date is set before my next progress review
Frequently asked questions
What is a data management plan in simple terms?
It is a document saying what data your research will produce, how you will format, document, store and preserve it, and what you will share. That is the University of Maryland’s definition, and its guide says any researcher or team benefits from a plan, not only funded projects.Does a Master’s thesis need a data management plan?
Only if your university, funder or ethics committee asks for one. At Reading, for example, the ethics DMP applies to taught postgraduate students only if the research is expected to lead to peer-reviewed publication, though Schools may still ask. Check your department’s handbook and ethics form.How long should a data management plan be?
Follow the template you are given. NSF’s plan is two pages. The University of Maryland’s guide says plans are typically short but should give more detail when your project is unique, special or especially complex.Do I need a data management plan if I only use secondary data?
At some universities, yes. Reading’s requirement applies to research students using only secondary data. Their plan is simpler: no data collection methods or sharing plan, though Reading says you can use it to cover secondary data, intellectual property and confidentiality.Can I change my data management plan after I submit it?
Yes, and you are expected to. Southampton treats the DMP as a living document reviewed before each progression review. Reading asks for a new version number and a short summary of changes each time you update it.What can I use to write a data management plan?
Start with the template your university or funder names. Without one, DMPonline, provided by the Digital Curation Centre, and the DMPTool, described by UC Irvine’s library as a free resource, both offer funder and institutional templates.Sources
- University of Maryland: Data Management Planning (DMP)
- University of Southampton: Theses, Data Planning
- University of Reading: PGR data management
- University of Reading: Research ethics (research data management)
- University of Sheffield: Progress monitoring and milestones
- UWE Bristol: Research governance (Doctoral Academy Handbook)
- U.S. National Science Foundation: Preparing Your Data Management and Sharing Plan
- Digital Curation Centre: How to Develop a Data Management and Sharing Plan
- DMPonline (Digital Curation Centre)
- UC Irvine Libraries: Managing Research Data
- UC Irvine Libraries: Data Storage and Backup
- University of Leeds Library: Safeguarding data
- UK Data Service: Anonymising data and protecting participants
How this guide was made: Thesisroom drafted this guide with the help of AI writing tools. Every fact, figure and source in this post was checked against its original source before publishing.