If you are finishing a thesis chapter or preparing a first manuscript, you have probably seen two similar-sounding requirements: a data availability statement and a data management plan. They are not the same document, they are not interchangeable, and they show up at different points in your research timeline. The short answer is that a data management plan (DMP) describes what you will do with your data before or during a project, while a data availability statement tells a journal reader where the data behind a finished paper can be found. In practice, many students and early-career researchers only discover the difference when a journal desk-rejects a submission for a missing statement, or a supervisor asks for a plan after the data is already collected.
This guide walks through both documents on the same research task so you can see what each one asks for, what it protects, and where it fits into your workflow. The running example is a student thesis project moving from a January start through revision and submission readiness.
Why Do Journals, Institutions, and Funders Ask for These Documents?
Both documents exist because good research data practices have become part of how science is assessed and shared. The FAIR principles describe research objects as findable, accessible, interoperable, and reusable, with persistent identifiers, rich metadata, provenance, and clear usage conditions. A data management plan is the forward-looking document that shows how you intend to apply those ideas to your own project. A data availability statement is the backward-looking record that tells a reader how you actually made the data behind a specific result available.
Funding policies reinforce the forward-looking side. The NIH data management and sharing policy expects planning for data management and sharing, with attention to tools, software, standards, preservation, access, oversight, and responsible reuse. That is difficult to do retroactively, which is why the plan belongs at the start. Journals and publishers, by contrast, increasingly want the statement at the end, attached to the manuscript itself.
When Do You Need a Data Management Plan?
You need a data management plan when the work is still being shaped. That may be a January project start, a thesis proposal, a grant application, or the moment you realize an existing project has accumulated files faster than you can describe them. A DMP asks questions you can answer well only while the work is still moving: Which files will you keep? How will you name them so they stay meaningful? Which formats can still be opened in five years? Who can access the data, and under what conditions?
The UK Data Service's technical guidance recommends organized files, meaningful names, documented versions, open or suitable formats, and supporting documentation so research data can be understood and reused. That list is a useful backbone for a student-scale DMP. You do not need a 20-page institutional template to get value from it; a two-page document can cover data types, file organization, naming, versioning, formats, storage, and access.
When Do You Need a Data Availability Statement?
You need a data availability statement when the paper or chapter is approaching submission. The statement is short, specific to that manuscript, and anchored to the results it accompanies. It typically answers where the data are located, how a reader can get them, and whether any access conditions apply. A thesis chapter may carry one per chapter if the chapters are built on separate datasets; a journal article carries one statement for the whole paper.
A common failure is treating this as a generic sentence such as "data available on request." If you are preparing a manuscript for a journal, the statement should point to a depository, a permanent institutional record, or a clearly described access route. If the data cannot be shared, the statement needs to say why in a way that still gives the reader as much of a path as possible. The statement is a reader-facing artifact: it exists so someone who has just read your result can find the material behind it.
What Are They Actually For? Comparing Both Documents on the Same Thesis
Imagine a student starting a mixed-methods thesis in January. The project collects interview transcripts, survey results, and a cleaned quantitative dataset. The DMP, written that month, addresses how consent forms will be stored, how transcripts will be pseudonymized, which survey export format will be preserved, and where versions will live. None of those decisions can be made well in April.
The following September, the student finishes a chapter and begins preparing a journal submission based on the survey data. Now the data availability statement is the relevant document. It says whether the survey data are deposited in a repository, whether the analysis code is linked, and what a reader must do to see materials that cannot be public. The DMP said what would happen; the statement says what did happen and where everything can be found.
The tradeoff is simple to see once you place the documents side by side. A DMP that is postponed past the data collection phase loses most of its preventive value. A data availability statement that is drafted too early becomes guesswork, because you cannot point to where data are until the deposit decision is real.
| Criterion | Data Management Plan | Data Availability Statement |
|---|---|---|
| When in the timeline | Start of project, proposal, or major revision | End of project, at submission |
| Primary audience | Supervisor, funder, institutional reviewer | Journal reader, reviewer, future researcher |
| Core question | How will data be handled? | Where are the data behind this result? |
| Length | One to several pages | One short paragraph to a few sentences |
| Main risk if ignored | Disorganized data, inaccessible files, rework | Desk rejection or unclear reader access |
| Best time to draft | Before or during data collection | After data are cleaned and deposited |
Why Most Students Start With the Wrong Document
The wrong-document problem usually begins when the pressure of submission starts. The student has a finished chapter, a reference list, and an inbox of supervisor feedback. The journal template asks for a data availability statement, but all the training materials the student remembers are about data management plans. It is tempting to reuse the DMP text. The result is a statement that reads like a plan: it describes intentions instead of locations, and it sends the reader to a project folder that may not be stable or public.
The reverse error is less visible but just as costly. A student who has only seen data availability statements may enter a new project thinking of data sharing as an end-of-paper checkbox. When the supervisor asks for a plan in January, the student has no clear answer on formats, naming, or preservation. The data gets collected first and organized later, and the cleanup becomes a winter-break archaeology project.
How Do You Build a Student-Scale Data Management Plan Without Overengineering It?
Start with the smallest version that still covers the project's actual data. For the thesis example, write one section for each distinct data type: interviews, survey data, and analysis scripts. Under each, note the format you will work in, the naming convention, how versions will be distinguished, and where the final preservation copy will live. If the data include personal information, describe access controls in plain terms.
Use the FAIR principles as a checklist rather than a compliance test. Can you point to a stable identifier for the dataset? Does the metadata say enough about how the data were produced? Is the format one a future colleague could open? Are the usage conditions stated? If the answer to any of these is no, that is a planning gap to close now, while the project is young. If you are using an AI research assistant to structure your early literature work, keeping the sources you assemble for the DMP in one place helps later. You can save or upload papers and organize them in the WisPaper My Library, which makes the planning conversation with your supervisor easier to reference.
How Do You Make a Data Availability Statement Submission-Ready?
Treat the statement as a specific, checkable answer to one question: if I read this result, how do I get the data? Use the name of the repository or archive if the data are deposited. If the project is a thesis and the data live in the university's institutional repository, say so. If a paper's data cannot be shared publicly, the statement should still tell the reader what exists and what access route applies, rather than leaving the line empty.
Before you mark the statement done, test it from the reader's side. Take the location named in the statement and check whether a colleague could actually reach the data from your wording. If the statement mentions a repository, confirm the deposit is complete and not just planned. If it mentions an access process, confirm you can describe that process in one sentence without a backstory.
How Do You Know the DMP or the Statement Is Ready?
A data management plan is ready when you can hand it to someone who knows nothing about your project and they can tell you, from the document alone, where each type of file lives and why. The plan should not depend on your memory of the project. If a section says "we will organize files later," it is not ready. The UK Data Service's emphasis on supporting documentation is the practical test: can someone understand and reuse the data from what you wrote?
A data availability statement is ready when it survives a single-sentence audit: the named location exists, the access conditions are stated, and the wording matches what is actually in place. If the paper or chapter changes during revision, re-check the statement in the same pass in which you re-check the references. The statement is part of the submitted artifact, not a note to yourself.
Combining Both Documents Into One Workflow
Use the DMP as the map and the statement as the receipt. In January, write the DMP before the first transcript is created. During the project, follow the naming and format decisions from the plan rather than improvising. When a chapter or paper is ready, copy the relevant parts of the plan into a draft statement, then replace every future-tense commitment with a present-tense location. If the plan said "survey data will be deposited in a university repository," the statement must say "survey data are deposited in the university repository at [record name]." If the deposit has not happened, the statement is not ready, no matter how polished the plan was.
For students using digital research workflow tools, the same principle carries across. When you assemble the background reading for the DMP, keep it in one accessible place so the plan can cite the sources behind your format and sharing choices. When you later screen the references in your manuscript for accuracy, check them with a tool built for that task, such as WisPaper TrueCite, which checks BibTeX against real academic databases to flag hallucinated references. The DMP and the statement serve different readers, but both depend on records you can verify.
What Do You Do Next With a Paper That Is Already Written?
If the paper is already written and you are only now seeing the data availability requirement, your first action is not to write the statement. It is to locate each dataset behind each result and confirm where it currently lives. Make a short inventory: for every figure, table, and analysis, note the file or data source and its current storage location. Then tackle the gaps. If a dataset has no stable home, deposit it in an appropriate repository or archive before you draft the statement. If the file names are meaningless to anyone but you, apply a minimal naming and documentation pass now, focusing on the materials a reader of this specific paper would need.
This is also a natural moment to make the paper's internal consistency dependable. Check the manuscript's references and data records together, because a submission that points to data that cannot be found creates the same trust problem as a hallucinated citation. If you have been keeping the paper's sources in a digital library, you can ask targeted questions about the uploaded material in WisPaper Library QA. That helps you confirm what your own reference set says before the reviewer does.
How Do You Use This for a January Project Start?
Start the January project with the DMP, but keep it light. Write the plan against the actual files you expect to create, not an idealized research lifecycle. Then hold one review point mid-project: your winter-break cleanup should not be a rescue operation but a check that the plan and the reality still match. At submission time, the statement should be derivable from the plan with minimal new decisions. If it is not, the likely cause is a plan that was too abstract in the first place, and that is worth fixing before the next project.
For researchers who are moving from one finished project into a new January start, the same reset applies. Move the old project's sources into an organized location, extract the format and naming decisions that actually worked, and carry those into the next plan. You can use WisPaper Research Projects to create a fresh project area for the next piece of work rather than letting everything blur into one folder.




