A public administration thesis set inside a real government agency — a city department, a state regulator, a federal programme office — needs something a purely academic proposal does not: a written agreement with the host agency covering what data the student can see, what they can quote, and who owns the resulting analysis. Without a template already sitting in the graduate school’s files, this step gets improvised late, department by department, and a good placement stalls at exactly the point it should be moving fastest. Eight steps below build a reusable access-agreement process, each with the office that owns it and the artefact it produces.
Step 1 — Assign one office to own the agreement template, not the individual supervisor
The graduate school or research-partnerships office drafts and maintains a single access-agreement template, rather than leaving each supervisor to draft one from scratch (or skip the step) for their own advisees. Artefact: a named office, and a version-controlled template document with a review date.
Step 2 — Define the three tiers of access the template needs to cover
Most agency placements fall into one of three access levels, and the template should name which one applies before drafting begins: publicly available records the student could obtain independently; internal but non-sensitive records shared specifically for the thesis; and records involving personal, law-enforcement-sensitive or otherwise restricted information that need a formal data-sharing clause and, in many jurisdictions, a separate privacy or freedom-of-information review the agency’s own legal office runs. Artefact: a three-tier access classification the student and supervisor select from at the outset, rather than discovering the tier only once a records request is denied.
Step 3 — Name what the student can and cannot quote directly
Agencies are often willing to share more than they are willing to have quoted verbatim in a public thesis. The template should separate “data the student can analyse” from “findings the student can publish” as two distinct permissions, with a named agency contact who signs off on the second before submission — not after. Artefact: a quotation and publication clause in the agreement, naming the agency reviewer.
Step 4 — Build in a pre-submission review window for the agency
A common friction point is a completed thesis reaching the agency for the first time at the same moment it reaches the examination committee, leaving no time for the agency to flag a concern. The template should include a fixed review window — a specific number of days, set by working backwards from the graduate school’s own submission calendar rather than assumed — during which the named agency contact can flag factual errors or request a redaction, without gaining editorial control over the thesis’s conclusions. Artefact: a review-window clause with a specific number of days and a named contact.
Step 5 — Separate data ownership from authorship
The agreement should state plainly that the agency retains ownership of its own underlying records, while the student retains authorship of their own analysis and interpretation — a distinction agencies and students both sometimes assume rather than see written down, and a source of dispute later if it is left ambiguous. Artefact: an explicit ownership-and-authorship clause, reviewed by the institution’s own legal or research-integrity office once, then reused as boilerplate.
Step 6 — Route the agreement through the same ethics process as any human-subjects research
Where the placement involves personal data, interviews or case records tied to identifiable individuals, the access agreement does not replace the institution’s own ethics review — it runs alongside it, and the template should say so explicitly so a student does not treat agency sign-off as a substitute for ethics clearance. This is the same discipline covered in this site’s piece on how a psychology programme routes ethics approval, applied here to an agency-hosted placement rather than a campus-based study: two separate approvals, run in parallel rather than treating one as a substitute for the other. Artefact: a cross-reference in the template pointing to the institution’s own ethics-review process, not a duplicate approval track.
Step 7 — Give the supervisor a short checklist, not the full legal document, to review at proposal stage
Supervisors are not usually equipped to review a data-sharing agreement’s legal language line by line, and should not need to. A one-page checklist — access tier confirmed, quotation clause present, review window specified, ownership clause present, ethics cross-reference checked — lets a supervisor confirm the agreement is complete at proposal stage without personally vetting its legal drafting, the same division of labour this site’s look at a nursing faculty’s proposal-review workflow applies to separating clinical-placement feasibility from academic rigor at review stage. Artefact: a supervisor-facing checklist, distinct from the agreement itself.
Step 8 — Log every agency partnership centrally, not department by department
A graduate school that logs which agencies have hosted student placements, on what terms, and how the relationship went, builds institutional memory that individual departments lose when a supervisor changes or retires. The next student proposing a placement at an agency the institution has worked with before starts from a known relationship and a known template, rather than negotiating access terms from zero. The register should note, at minimum, the agency name, the access tier granted, whether the agency used its own paperwork or the institution’s template, and any issue that came up during the placement — a slow review window, a late redaction request, a change of agency contact partway through — so the next placement at the same agency starts with that context already available rather than rediscovering it. Artefact: a central register of agency partnerships, owned by the same office that owns the template from Step 1.
What this looks like when it goes wrong
The failure pattern without a template is consistent across public administration cohorts: a student secures verbal access to agency records, writes most of the thesis, and only then discovers the agency expects sign-off before anything is quoted publicly — at which point the review can take weeks the submission calendar does not have. A written template, agreed before data collection begins rather than negotiated after a draft exists, moves that friction to the start of the process, where it costs a conversation rather than a missed deadline.
Adapting the template across different kinds of agencies
A federal programme office, a state regulator and a city department differ in how formal their own data-sharing processes already are — a federal agency more often has an existing standard agreement the institution’s template needs to align with rather than replace, while a smaller city department may have no standard process at all and will effectively adopt whatever template the university proposes. The graduate school’s own template should be built as the default starting position, with an explicit expectation that a well-resourced agency partner may substitute its own paperwork, and the review-partnerships office should keep a note of which agencies use their own agreement versus the institution’s. Where an agency insists on its own paperwork, the partnerships office should still check it against the same five elements the internal checklist covers — access tier, quotation clause, review window, ownership clause, ethics cross-reference — even if the document’s language and structure look nothing like the university’s own template, so the same protections apply regardless of whose paperwork is actually signed.
How this compares to other placement-based thesis structures
A public administration placement shares its underlying shape with other fields that route students into a host organisation before writing: an internship or practicum-based thesis in a social work programme separates placement performance from academic rigour using a similar structural logic to the access-tier and ownership clauses here, even though the host organisation, the sensitivity of the records and the specific legal framework differ by field. A graduate school running placement-based theses across several departments — public administration, social work, and others with a fieldwork component — can build one underlying template structure (access tier, quotation clause, review window, ownership clause, ethics cross-reference) and adapt only the field-specific detail, rather than designing each department’s process independently.
Why students consistently underestimate this step
Students proposing an agency-hosted thesis tend to focus their proposal-writing effort on the research question and methodology, treating the access agreement as an administrative formality to sort out once the topic is approved. In practice the access terms often shape what the research question can honestly be — a question that assumes access to case-level records the agency will not release cannot be answered as originally framed, and discovering that after the proposal is approved costs more than discovering it before. A template that exists before proposal writing begins lets a student check feasibility against the access tiers in Step 2 while the question is still easy to revise, rather than after a committee has already signed off on a version of it that access rules make unanswerable.
Where Tesify fits
The access agreement itself is a legal and institutional judgment that stays with the graduate school’s own research-partnerships and legal offices — none of it is a writing-platform decision. Once a student has cleared access and moved into drafting, Tesify is the platform candidates use to write their own thesis chapter by chapter: more than 9,000 students have written over 15,000 chapters with it, and the thesis stays 100% written by the candidate. See how Tesify supports a public administration thesis cohort.
Frequently asked questions
Who should own the access-agreement template?
The graduate school or research-partnerships office, not individual supervisors — a template held by one office is maintained consistently and reused across departments, rather than redrafted or skipped case by case.
Does the access agreement replace ethics review?
No. Where the placement involves personal data or identifiable individuals, the agreement runs alongside the institution’s own ethics-review process, not instead of it.
Who decides what the student can quote directly from agency records?
The agreement should separate data the student can analyse from findings the student can publish, with a named agency reviewer signing off on the second before submission.
How long should the agency’s pre-submission review window be?
A specific number of days, set by working backwards from the graduate school’s own submission calendar rather than left open-ended, so the review neither slips past the deadline nor gives the agency indefinite control over the timeline.
Who owns the data collected during the placement?
The agency retains ownership of its own underlying records; the student retains authorship of their own analysis and interpretation. Stating this explicitly in the agreement avoids a later dispute over who controls what.
Should every agency placement use the same template?
The institution’s template is the default starting position, but a well-resourced agency partner, particularly at the federal level, may already have its own standard data-sharing agreement the university needs to align with rather than replace.
What should a supervisor check before approving a placement proposal?
A short checklist confirming the access tier, quotation clause, review window, ownership clause and ethics cross-reference are all present — not a full legal review of the agreement’s drafting, which is not the supervisor’s role.
Can a research question change after the access agreement is signed?
It can, but a question that assumes access the agreement does not grant will need revising, which costs less time before a committee has approved the proposal than after. Checking the question against the agreed access tier before proposal submission avoids that rework.
Does this template work for other placement-based thesis fields?
The underlying structure — access tier, quotation clause, review window, ownership clause, ethics cross-reference — generalises to other fields with a fieldwork or placement component; only the field-specific detail, such as the type of host organisation and the applicable legal framework, needs adapting.