Last updated: September 2026
This guide focuses on admissions-stage research plans for research-based master’s and doctoral programs. Document names, required sections, and the role of prospective advisors vary by university, department, and field. If your application provides a fixed form or prompt, that form always takes priority. (The thesis proposal you submit after admission, usually before your thesis semester, is a different document with its own departmental rules.)
If you have already contacted a prospective advisor, you may have written a miniature version of this document: a one-page research sketch attached to your email. If the professor asks to see more, you can expand that sketch into a fuller proposal. If you have not contacted a professor — or your department does not use pre-application contact — you can use the same process to build the proposal required by the application itself. Either way, the document is the 연구계획서, and it is where the committee sees whether your interests can become a viable research project. The hub guide explains why it matters; this post is about how to actually build one.
Before the structure, the warning that every section below exists to serve: one of the most common proposal failures is the one written to impress a committee instead of to answer a skeptical specialist reader. Your proposal may be read by a prospective advisor, a departmental committee, or both — often alongside many other applications. It therefore needs to make its question, contribution, and feasibility visible quickly. Everything that follows is engineered for that reader.
First, Know Which Document You’re Being Asked For
Korean applications use labels that international applicants constantly confuse:
- 연구계획서 (research proposal) — usually research-centered: a question, its context, a method, a timeline. This is what this post covers.
- 학업계획서 / 수학계획서 (study plan) — often covers broader degree goals: why this program, what courses and skills, what you’ll do after. More autobiographical, less technical.
These labels are useful, but they are not standardized across Korean universities. Some universities combine study, motivation, and research plans in a single form — Seoul National University’s graduate form is a “Personal Statement & Study Plan,” and Yonsei’s “Study and Research Plan” asks for self-introduction, motivation, and study-and-research plans in one document. Read the actual prompts, not just the document title, and keep autobiographical material in its lane so it doesn’t leak into the research half.
For scholarship applicants, be precise. For the 2026 GKS-G application, master’s and doctoral degree applicants submit Form 3, the Study Plan. The separate Form 4 Research Proposal is required only for applicants to the designated Research Program. The Study Plan may still need to explain your research direction, so follow the questions in the current form rather than relying on the document label alone — confusing the two can leave important parts of the prompt unanswered. The GKS side is covered in the GKS guide.
One more formatting reality: many departments issue a fixed form (양식), sometimes with page or character limits, occasionally with prescribed headings. The form always wins. The structure below is for when you’re given freedom, and a mental checklist for when you’re not.
From One-Page Sketch to Full Proposal
If you sent a sketch during 컨택, the full proposal expands its three claims — here is a question, here is why it matters, here is why it belongs in your research group — exactly at the points where a careful reader would push back, and nowhere else.
Two rules govern the relationship:
- Explain substantial changes. Refinement is expected. If the question, data, or theoretical direction has changed substantially since the sketch, explain briefly what changed and why — a sentence in the covering email like “after reading further in X, I’ve narrowed the question from A to B” is enough. A reasoned revision signals development; an unexplained replacement can make the project look unstable.
- Expansion means depth, not padding. Every added page should be answering a harder question than the sketch could. If a section exists only because proposals are supposed to have that section, the reader can tell.
The Structure: Five Sections, Five Silent Questions
Here is the perspective that makes the whole genre click. A research proposal is not a container you fill with information — it is a sequence of answers to questions your reader is silently asking, in a predictable order. Write each section as the answer to its question and the document organizes itself.
A rough allocation, for when the form gives you no guidance (form and length limits always take priority):
| Section | What the reader needs | Suggested share |
|---|---|---|
| Background | Problem and significance | 15–20% |
| Literature review | Existing work and the gap | 20–25% |
| Questions / objectives | The exact research target | 10–15% |
| Methods | Data, procedure, feasibility | 30–35% |
| Contribution / timeline | Output and schedule | 10–15% |
1. 연구의 배경 및 필요성 — “Why does this matter, and why now?”
Open with the phenomenon, not with your biography. The weakest openings recount how the applicant became interested in the topic; the strongest establish, in a paragraph or two, that something in the field is unresolved, newly urgent, or newly possible to study. Your personal path can appear in one sentence — it belongs in the study plan, not here.
The functional job of this section is to stake out territory the reader recognizes as real. If your prospective advisor works in this territory — and if you chose them properly , they do — this is also your first, quietest demonstration of fit.
2. 선행연구 검토 — “Do you know the conversation you’re entering?”
This is a literature review with a specific mission, and it is not the mission most applicants assume. You are not proving you’ve read a lot. You are drawing a map of the existing conversation in order to point at the empty spot on it. Every study you cite should be doing work: either supporting the ground your question stands on, or marking the gap your question fills. A review that summarizes ten papers without arriving anywhere is the most common way strong applicants produce weak proposals.
Practical mechanics: RISS and KCI for Korean-language scholarship, Google Scholar for the international literature. If your advisor has published in the area, cite the work that is genuinely relevant — as scholarship, in the same register as everything else you cite. Citing it as tribute is transparent and mildly embarrassing for everyone; omitting it when it’s directly on point suggests you didn’t do the reading. The test is simple: would this citation survive if the author weren’t your prospective advisor?
3. 연구 문제 및 목적 — “What exactly are you asking?”
The center of gravity. State your research questions explicitly — numbered, if there are several — and make each one answerable: a question whose answer could in principle turn out one way or another. “This study explores aspects of X” is a topic wearing a question’s clothes. “Does X differ between A and B, and if so, in what respects?” is a question a thesis can actually answer.
Here is the difference on the page.
Too broad: This study will explore social media and identity among international students.
Answerable: How do Vietnamese graduate students in Korea use Korean, Vietnamese, and English when presenting academic and personal identities in university-based KakaoTalk groups?
Two or three well-formed questions generally beat five ambitious ones. Each question you add is a promise the methods section now has to keep.
4. 연구 방법 — “Can you actually do this?”
The section where a promising idea either becomes credible or remains only an idea. You need to name your data (what, from where, how much), your method of analysis (what you’ll do to the data, and why that method answers these questions), and — the part applicants skip — the feasibility constraints: access, language, tools, time.
Continuing the example above, the methods paragraph that keeps the question’s promise looks like this:
Data: 3–5 consenting university KakaoTalk group chats and follow-up interviews with 12–15 participants.
Method: interactional discourse analysis of language choice and switching, plus thematic coding of the interviews.
Scope: one university, one semester.
For empirical projects, also specify how cases or participants will be selected and how you will make the analysis credible — for example through a coding protocol, triangulation, inter-coder checks, or transparent analytic procedures appropriate to your field. If human participants are involved, say how you will handle consent, privacy, data storage, and institutional ethics review, rather than gesturing at “approvals.”
Precision here signals more than diligence. Many full-time master’s programs are structured around four semesters, but the usable thesis timeline may be shorter — a methods section scoped like a five-year project tells the reader you haven’t yet imagined the degree from inside. It is far more impressive to propose something modest and demonstrate you know exactly how to execute it than to propose something sweeping on faith.
5. 기대 효과 및 연구 일정 — “Does this fit the clock, and what will exist at the end?”
Short section, two jobs. First, a semester-by-semester timeline: coursework, data collection, analysis, writing, defense. It will be wrong — everyone’s is — but its presence shows you’ve thought in calendar terms, and its realism is read as a proxy for your general judgment. Second, expected contribution: one honest paragraph on what the field, however narrowly defined, gains. Resist the temptation to claim your thesis will transform the discipline; the reader has advised enough theses to know the genre’s actual ceiling, and inflated claims here retroactively cheapen everything above them.
Close with the reference list — complete, consistently formatted in whatever style your field uses, and containing only works the text actually cites and you have actually read.
For doctoral applicants: The same five questions apply, but the burden of originality is higher. You should position the project more precisely within the literature, demonstrate stronger methodological readiness, and explain how the proposed work extends — not merely repeats — your master’s research.
Showing Fit Without Saying “Fit”
The hub guide makes the judgment call: a solid proposal aimed at the right research group beats a brilliant one aimed at the wrong group. What does aiming look like on the page? Not a paragraph declaring admiration. Fit is shown structurally: your question sits in the territory the advisor works in, your literature review includes the conversation they’re part of, your method is one their group can plausibly supervise. Nothing announces it; nothing pulls.
The corollary is the failure mode I see most often from strong applicants: proposing, in effect, the advisor’s own recent paper back to them. It feels like the safest possible fit. It reads as either flattery or a lack of independent ideas, and it leaves the advisor nothing to supervise. The sweet spot is adjacency — visibly in their territory, visibly not their existing work. Your angle, their neighborhood.
Korean or English?
First, follow the application’s language requirement. If both Korean and English are accepted, choose the language in which you can present your research most precisely and naturally — reviewers are evaluating your research thinking, and machine-translated academic Korean obscures exactly that. English-track and GSIS programs expect English. Some Korean-track programs accept English from international applicants, while others impose department-specific requirements; confirm with the department office or admissions office rather than assuming. If you do write in Korean, have a native academic reader check register, not just grammar: academic Korean is hedged and formal in ways that language school doesn’t teach, and unhedged declarative prose can read as overconfident even when the content is sound.
The Failure Modes, Collected
- The recycled undergrad thesis. Submitting your graduation paper’s abstract with the verbs changed to future tense. Advisors recognize the genre instantly — a proposal looks forward from a gap, not backward from a result.
- The literature review with no destination. Ten summaries, zero gap. See section 2.
- The five-year question in a two-year degree. See section 4.
- The wrong genre. A study plan where a 연구계획서 was requested, or vice versa — or a combined form answered as if it were only one of them. Reread the prompts.
- The citation you never opened. AI-generated or secondhand references that do not say what the proposal claims they say can undermine the credibility of the entire document.
- The unsupervisable proposal. Perfectly executed, aimed at no one in the department. The document is only half the aim; the advisor is the other half.
FAQ
How long should a Korean graduate research proposal be? There is no universal number — departments differ, and fixed forms (양식) override everything. As a working default when no length is specified, think “long enough to answer the five silent questions, short enough to be read in one sitting,” which for most applications means roughly 2–5 pages as a planning default. When in doubt, ask the department office; the question is normal.
Will I be held to my proposal after admission? Usually not as a fixed contract. Research questions commonly evolve through coursework and supervision. However, a major change may affect advisor fit, funding, data access, or ethics approval, so discuss substantial changes with your advisor rather than assuming they are purely private decisions. The proposal is evaluated as a demonstration of how you think — you are not choosing your thesis today, you are showing you know what choosing one involves.
Is the 연구계획서 the same as the GKS study plan? Not by default. GKS degree applicants submit a Study Plan (Form 3 in the 2026-G cycle); a separate Research Proposal (Form 4) applies only to the Research Program. Some applicants need both a university 연구계획서 and a GKS Study Plan, written separately for their respective prompts. Details: the GKS guide.
Should I send the full proposal with my first email to a professor? Not automatically. Follow the professor’s or department’s instructions. If pre-application contact is permitted and no format is specified, a concise email with a CV and a short research summary is often enough to begin. Send a full proposal when it is requested or when the department explicitly asks for one. The contact sequence, with templates, is in How to Contact a Korean Professor Before Applying.
Research proposals are the documents I write most, read most, and rewrite most in my own graduate life — for funding applications, for projects, and alongside the international students I work with as they draft theirs for admission. The five silent questions above are the checklist I actually use. If your form or your field bends the genre in a way this post didn’t cover, tell me — reader questions have a habit of becoming the next post.
Leave a Reply