← AcademiaOne PhD Hub · 返回首页
AcademiaOneEDUCATION
PhD Research Proposal Guide · HK · Social Sciences 香港社会科学研究计划书指南
AcademiaOneEDUCATION
Your PhD application, from idea to offer你的博士申请,从构想到录取
DOCTORAL APPLICATIONS · HONG KONG · SOCIAL SCIENCES   香港社会科学博士申请

PhD Research
Proposal Guide · Social Sciences

香港社会科学博士研究计划书写作指南

A section-by-section model of a social-science PhD research proposal for Hong Kong, covering psychology, sociology, politics, economics, social policy and communication. Each part shows the structure, a fully worked example from psychology, and the difference between a weak and a stronger version, with the current requirements of HK departments taken from their official pages.

逐部分拆解一份香港社会科学博士研究计划书,涵盖心理学、社会学、政治学、经济学、社会政策与传播学。每一部分都包含结构说明、一个完整的心理学范例,以及“较弱”与“更好”写法的对比,并附香港各院系官网上的最新要求。

Start here 开始之前

What Hong Kong universities ask for香港各校要求

Hong Kong universities ask for a research proposal or a research plan in almost every field, but the form differs: HKU faculties set their own format, PolyU uses Form GSB/1A, HKUST takes a research plan typed into the online form, and CityU calls the document a research plan in several schools. The Hong Kong PhD Fellowship Scheme (HKPFS) and the main round at HKU and CUHK all close on 1 December 2026, the earliest deadline of the year, so plan backwards from that date.

香港各校几乎所有学科都要求提交研究计划书(有的称为“研究计划”),但形式各不相同:港大各学院自定格式,理大使用 GSB/1A 表格,港科大要求在网申表格中直接填写研究计划,城大多个学院称之为“研究计划”。香港博士研究生奖学金计划(HKPFS)与港大、港中文的主轮申请都在 2026 年 12 月 1 日截止,是一年中最早的截止日期,请以此倒推时间表。

Social-science departments in Hong Kong set their own format, from a faculty form to a document of 12 pages. HKU uses a faculty proposal form, and its Hong Kong Institute for the Humanities and Social Sciences (HKIHSS) allows up to 8 pages plus a writing sample; HKUST Social Science asks for 10-12 pages; PolyU for Form GSB/1A with a personal statement; and CityU for about 3 single-spaced pages, sent to a staff member before you apply.

香港社会科学院系各自规定格式,从学院表格到 12 页的完整文件不等。港大使用学院计划书表格,其香港人文社会研究所(HKIHSS)允许不超过 8 页并要求写作样本;港科大社会科学部要求 10-12 页;理大要求 GSB/1A 表格加个人陈述;城大要求约 3 页(单倍行距),并须在申请前发给一位教员。

Social Sciences: what each programme asks for社会科学:各项目要求

UNIVERSITY 大学WHAT THE PROGRAMME ASKS FOR 项目要求UNIVERSITY-WIDE RULE 全校规定OFFICIAL PAGE 官方页面
HKU
香港大学
QS 2027: 11
Faculty proposal form; HKIHSS max 8 pages
HKIHSS: writing sample (paper ≤15 pages + master's thesis abstract)
References: 2
Proposal or statement of research interest in faculty format
gradsch.hku.hk
gradsch.hku.hk
CUHK
香港中文大学
QS 2027: 18
Research proposal
Psychology page says 2 referees; Graduate School says 3
References: 2-3
Not stated centrallypsy.cuhk.edu.hk
HKUST
香港科技大学
QS 2027: 33
PhD proposal 10-12 pages
References: 2+
Research topic and plan in form; extra proposal for Humanities and Social Science
fytgs.hkust.edu.hk
fytgs.hkust.edu.hk
PolyU
香港理工大学
QS 2027: 50
Form GSB/1A + personal statement
Applied Social Sciences: proof the master's had a research component
References: 2
Compulsory on Form GSB1a
polyu.edu.hk
polyu.edu.hk
CityU
香港城市大学
QS 2027: 52
Proposal ~3 pages single-spaced; send it to a staff member before applying
References: 2
Required for Liberal Arts & Social Sciences, Creative Media, Energy & Environment, Law
cityu.edu.hk
cityu.edu.hk

Requirements are quoted from the official page linked in each row, in English as published. The third column is the university-wide proposal rule from the graduate school page. 表中要求按各行所附官方页面的英文原文摘录;第三栏为研究生院页面上的全校计划书规定。

Golden rule 核心原则

The programme page always beats this guide. Use this guide to decide what to write; use the programme page to decide how long it is and what form it takes.

项目官网的要求永远优先于本指南。用本指南决定“写什么”,用项目页面决定“写多长、用什么格式”。

HKPFS comes first HKPFS 最先截止

The HKPFS has two steps. Get an HKPFS reference number on the Research Grants Council (RGC) system by 12:00 noon HKT on 1 December 2026, then apply to up to two programmes at one or two universities, quoting it, by 11:59 pm the same day. Universities pre-screen and nominate candidates, and HKPFS applicants are considered for each university’s own awards. Later rounds exist: CUHK clearing closes on 31 March 2027 and HKU Medicine on 30 April.

HKPFS 分两步:先在研究资助局(RGC)系统取得 HKPFS 编号,截止时间为 2026 年 12 月 1 日香港时间中午 12 点;再在当天晚上 11:59 前,凭该编号向一至两所大学的最多两个项目提交申请。各校先行筛选并提名候选人,HKPFS 申请人也会自动参与各校自设奖学金的评选。之后还有补录轮:港中文补录轮于 2027 年 3 月 31 日截止,港大医学院为 4 月 30 日。

Referees 推荐人

HKU, PolyU and CityU ask for two academic referees, HKUST for at least two (the form takes up to five), and CUHK for three for research doctorates. At PolyU your proposed supervisor cannot be one of your referees.

港大、理大和城大要求两位学术推荐人,港科大至少两位(表格最多可填五位),港中文研究型博士要求三位。理大规定拟定导师不能担任推荐人。

Contact first, and check the tests 先联系导师,并核对考试

CityU asks you to send your proposal to a staff member before you apply. HKU asks applicants to psychology without a psychology degree for the GRE Psychology test. CUHK Psychology lists two referees while the Graduate School asks for three, so prepare three.

城大要求申请前先把计划书发给一位教员。港大要求非心理学本科背景的心理学申请者提交 GRE 心理学专项考试成绩。港中文心理学系页面写两位推荐人,研究生院写三位,建议准备三位。

Table built from 5 official programme pages (each checked on the programme page itself; data checked 2026-11-18). Rules change each cycle; the programme page always wins. 表格依据 5 个官方项目页面整理(均在项目页面本身核对;数据核对日期 2026-11-18)。规则每年可能调整,以项目页面为准。

Start here 开始之前

Four caveats by major各专业的四个注意事项

The social sciences share one proposal model, but four situations change what you write or where you apply. Read the one that applies to you before you plan your timeline.

社会科学各专业共用一套计划书写法,但有四种情况会改变你写什么、在哪里申请。规划时间表之前,请先读与你相关的那一条。

1Psychology: clinical programmes are separate

Clinical psychology training is a professional route, and it is separate from a research PhD on a clinical topic. In Hong Kong, professional clinical training is a separate programme from the research MPhil and PhD, with its own entry rules, so check the department page. If you want to research a clinical topic without training as a clinician, apply to a research PhD and say so in the proposal.

临床心理学培训是一条职业路径,和“研究临床题目的科研博士”是两回事。在香港,临床专业培训与研究型 MPhil、PhD 是不同的项目,入学要求各自独立,请查看院系页面。如果你想研究临床问题但不打算成为临床执业者,请申请科研型博士,并在计划书中写明这一点。

2Economics: GRE and a coursework stage

Economics PhDs are organised like business PhDs: one to two years of coursework in microeconomics, macroeconomics and econometrics, qualifying examinations, then a thesis of about three papers. Many ask for the GRE or GMAT: in Hong Kong, HKU requires one of them for Business and Economics. Panels often read a short statement of research interests and your mathematics preparation more closely than a long proposal.

经济学博士的结构与商科博士相似:先修一到两年的微观、宏观与计量经济学课程,通过资格考试,再完成约三篇论文组成的博士论文。许多项目要求 GRE 或 GMAT:在香港,港大商科与经济学要求其中之一。评审往往更关注简短的研究兴趣陈述和你的数学基础,而不是长篇计划书。

3Politics, IR and sociology: longer and theory-heavy

Proposals in these fields are often longer and are expected to argue from theory. HKUST Social Science asks for a PhD proposal of 10-12 pages, HKU’s HKIHSS for up to 8 pages plus a writing sample, and CityU for about 3 single-spaced pages sent to a staff member before you apply. Give the theoretical framework its own subsection, define your concepts, and say which side of a debate your project tests.

这些专业的计划书通常更长,并要求从理论出发展开论证。港科大社会科学部要求 10-12 页的博士计划书,港大人文社会研究所要求不超过 8 页并附写作样本,城大要求约 3 页(单倍行距),并须在申请前发给一位教员。请为理论框架单独设一小节,界定核心概念,并说明你的项目检验的是哪一方的观点。

4Funding forms have their own limits

The HKPFS is a separate step from the university application. You first get a reference number on the RGC HKPFS system by 12:00 noon HKT on 1 December 2026, then apply to up to two programmes at one or two universities, quoting it. The university application then has its own format: HKU faculty forms, PolyU Form GSB/1A or HKUST’s online research plan. Cut the master version to fit each form, and add the training and impact paragraphs the form asks for.

HKPFS 与大学申请是两个步骤:先在 RGC 的 HKPFS 系统取得编号(2026 年 12 月 1 日香港时间中午 12 点前),再凭编号向一至两所大学的最多两个项目提交申请。大学申请本身另有格式:港大学院表格、理大 GSB/1A 表格或港科大网申研究计划。请把母版按每份表格裁剪,并补上表格要求的培训与研究影响段落。

Start here 开始之前

How long, and how the words are shared篇幅与字数分配

Write one full master version of about 2,500 words, then cut it to size. Social-science panels spend most of their time on the theory and the design, so those parts get the most words. References are usually extra, but some departments count them. Each section page then gives a blueprint, a worked example, a weak versus stronger comparison and lines from other fields. In Hong Kong the limit is often set in pages or by a form (HKU Arts 4-5 pages double-spaced, HKUST Social Science 10-12 pages, HKU Education up to 20 pages), so check the spacing and whether references count before you cut.

先写一份约 2,500 词的完整母版,再按要求裁剪。社会科学评审把大部分时间花在理论与研究设计上,所以这两部分字数最多。参考文献通常不计字数,但部分院系计入。之后每个部分都包含结构说明、范例、“较弱”与“更好”的对比,以及其他专业的示例。在香港,篇幅上限常以页数或表格来规定(港大文学院 4-5 页双倍行距,港科大社会科学部 10-12 页,港大教育学院不超过 20 页);裁剪之前请确认行距要求以及参考文献是否计入。

150650200800250250200
0 Title & abstract1 Background & theory2 Aims & questions3-4 Methods & design5 Ethics & access6 Contribution & fit7 Timeline

What panels look for in each field各专业评审关注点

FIELD 专业WHAT PANELS LOOK FOR 评审关注
Psychology
心理学
Testable questions, validated measures, a reason for the sample size, pre-registration for confirmatory studies, and ethics for human participants.
可检验的问题、经过验证的量表、样本量依据、验证性研究的预注册,以及涉及人类参与者的伦理安排。
Sociology · Social policy
社会学 · 社会政策
A defined concept and theory, a population or case chosen for a reason, and a survey, interview, ethnographic or mixed design; access to the field is read closely.
界定清楚的概念与理论、有理由的研究对象或个案,以及问卷、访谈、民族志或混合设计;评审会仔细看你能否进入田野。
Politics · International relations
政治学 · 国际关系
A puzzle placed in a theoretical debate, justified case selection, and the sources you will use: archives, interviews, datasets or texts.
置于理论争论中的研究困惑、有依据的个案选择,以及你将使用的资料:档案、访谈、数据集或文本。
Economics
经济学
A causal or structural question, named data, an identification strategy or model, and mathematical preparation; often a short statement of research interests.
因果或结构性问题、具体数据、识别策略或模型,以及数学基础;常常只需简短的研究兴趣陈述。
Communication
传播学
Experiments and surveys for media effects; content analysis, platform data or interviews for media systems, with a lawful plan for collecting and storing platform data.
媒介效果研究用实验与问卷;媒介系统研究用内容分析、平台数据或访谈,并说明如何合法采集与存储平台数据。

The five questions every social-science panel asks社会科学评审必问的五个问题

#PANEL QUESTION 评审问题WHERE YOU ANSWER IT 在哪里回答
1Is this a real problem placed in a current debate? 这是一个置于当前学术争论中的真实问题吗?Background · Abstract 背景 · 摘要
2Does a theory or concept organise the project? 是否有理论或概念统领整个项目?Background & theory 背景与理论
3Is the question sharp enough to finish in 3-4 years? 问题是否足够聚焦,能在 3-4 年内完成?Aims & questions · Timeline 目标与问题 · 时间安排
4Can the method answer it, and can you reach the people or data ethically? 方法能回答问题吗?你能合乎伦理地接触到研究对象或数据吗?Methods · Design · Ethics & access 方法 · 设计 · 伦理与进入
5Why here, and why this supervisor? 为什么是这里,为什么是这位导师?Contribution & fit 贡献与匹配度
Before you write 动笔之前

The golden thread逻辑主线

A proposal works when each section follows from the one before: the gap leads to the question, the question decides the method, and the method produces the contribution. In the social sciences a theory runs underneath all four and tells the reader why the gap matters. If you cannot write these four sentences, narrow the topic before drafting anything else.

好的计划书每一部分都顺着前一部分推进:研究空白引出研究问题,问题决定方法,方法产出贡献。在社会科学中,还有一条理论贯穿这四步,告诉读者这个空白为什么重要。如果写不出这四句话,先把课题缩小再动笔。

GAP研究空白

Belonging is protective for wellbeing, but no study has followed how Mainland Chinese postgraduates come to belong during a one-year UK degree.

QUESTION研究问题

How does belonging develop across the first year, and can a brief intervention at orientation improve it?

METHOD研究方法

A four-wave survey, follow-up interviews, then a pre-registered randomised trial of a brief belonging intervention.

CONTRIBUTION研究贡献

A timeline of when support matters most and a low-cost intervention universities can run at orientation.

The worked example used in this guide本指南使用的范例

Example applicant 范例申请人

Background 背景
BSc Psychology (Mainland China); MSc Psychology (UK), with a dissertation on loneliness among international students; research assistant on a wellbeing intervention study.

Applying to 申请
PhD in Psychology in Hong Kong.

Working title 暂定题目
Finding a Place: Belonging, Loneliness and Wellbeing among Mainland Chinese Postgraduate Students in the UK

How to use the example 如何使用范例

The same psychology proposal appears in every section so you can watch one project move from gap to timeline. Don’t copy the topic. Study the moves, the numbered notes beside each extract, then make the same moves with your own field, data and supervisor. Citations shown as (Authors, Year) are placeholders; in your own proposal every claim needs a real source you have read. The examples are in English only, because your application will be written in English.

同一份心理学计划书贯穿每个部分,方便你看到一个项目如何从研究空白一路走到时间安排。不要照搬课题,要学习写法:先看每段旁边的编号解析,再用你自己的专业、数据和导师做同样的事。(Authors, Year) 为占位符,你自己的计划书里每个论断都必须有你读过的真实文献。范例仅用英文,因为你的申请材料也用英文写。

Three other projects that run through this guide贯穿本指南的另外三个项目

Sociology 社会学 Question: Does a parent’s return from migrant work change rural children’s educational aspirations, and through which family processes? Method: household panel data plus interviews in two counties.
Politics & IR 政治学 Question: Why do some local governments comply fully with central environmental inspections while others comply only on paper? Method: four comparative case studies with process tracing.
Economics 经济学 Question: Do minimum-wage increases push low-wage workers into informal employment? Method: staggered city-level wage changes in a difference-in-differences design.
Section 0 第 0 部分
0

Title & Abstract题目与摘要

Typical length 典型篇幅 title 12-20 words · abstract 150-250 words

The title and abstract are often all a reader sees before deciding how carefully to read the rest. Social-science titles name the concept, the population and the setting, and many signal the design after a colon.

读者常常只看题目和摘要,就决定其余部分要读得多仔细。社会科学的题目要写出核心概念、研究对象与情境,许多题目还会在冒号后点明研究设计。

Anatomy of a title题目结构

Finding a Place: Hook (optional) 开头(可选)
Belonging, Loneliness and Wellbeing Concepts 核心概念
among Mainland Chinese Postgraduate Students in the UK Population + setting 对象 + 情境

Weak vs stronger titles较弱与更好的题目

✕ Weak 较弱“Mental Health of International Students”

Far too broad: no specific concept, population or setting. A panel can’t tell what you will do.

过于宽泛:没有具体概念、对象或情境,评审看不出你要做什么。

✓ Stronger 更好“Belonging Trajectories and a Brief Belonging Intervention for Mainland Chinese Postgraduate Students in the UK: A Longitudinal and Experimental Study”

Names the concept, who is studied and where, and signals the design.

点明概念、研究对象与地点,并提示研究设计。

✕ Weak 较弱“Why Are Chinese Students So Lonely in the UK?”

Assumes the answer before the research and reads like a headline.

在研究之前就预设答案,读起来像新闻标题。

✓ Stronger 更好“How Does Belonging Develop in the First Year? A Study of Mainland Chinese Taught Postgraduates”

Neutral, researchable, and clear about what is measured.

中立、可研究,并清楚说明测量内容。

Sociology 社会学 Coming Home: Parental Return Migration and Children’s Educational Aspirations in Rural China
Politics & IR 政治学 Campaigns and Compliance: Local Government Responses to Central Environmental Inspections in China
Economics 经济学 Minimum Wages and Informal Employment: Evidence from City-Level Wage Changes in China
Topic as title 以领域作题目

“Social media and youth” names a field, not a project.

“社交媒体与青年”只是一个领域,不是一个项目。

Jargon pile-up 术语堆砌

Five technical terms in a row make the title hard to read aloud.

连续五个专业术语,题目念都念不顺。

Answer in the title 题目预设结论

“How X harms Y” tells the panel you have decided the result.

“X 如何损害 Y”等于告诉评审你已经认定了结论。

Section 0 continued 第 0 部分(续)

The abstract摘要

Blueprint结构

1Move 1 · Context

Why it matters now.

为什么现在重要。

2Move 2 · Gap

What is missing, and why that matters.

缺什么、为什么重要。

3Move 3 · Question

The main research question, stated outright.

直接写出主要研究问题。

4Move 4 · Method

Design, sample, data in one or two sentences.

用一两句话写设计、样本与数据。

5Move 5 · Contribution

What changes if you succeed.

成功之后有什么改变。

Worked example · Abstract 范例 · 摘要

1Mainland Chinese students make up one of the largest groups of international postgraduates in UK universities, yet little is known about how they come to feel they belong during a one-year taught degree. 2Research shows that belonging protects wellbeing, but existing studies are mostly cross-sectional and drawn from undergraduate or non-UK samples, so we cannot say when belonging is won or lost.

3This project asks how belonging develops across the first year and whether a brief, low-cost intervention can improve it. 4Study 1 will follow 400 students at three UK universities in a four-wave survey; Study 2 will interview 24 students about turning points; Study 3 will test a brief belonging intervention in a pre-registered randomised trial (n = 120). 5The project will show universities when support matters most and give them an evidence-based intervention they can run at orientation.

Why it works 写法解析
1Context 背景

Names a large, current group, so the reader sees the stakes at once.

点出规模大且当下的人群,让读者立刻看到意义。

2Gap 空白

Says what is missing and why that limits what we know.

说明缺什么,以及这为何限制了现有认识。

3Question 问题

Stated in one sentence the reader can’t miss.

一句话直接说出,读者不会错过。

4Method 方法

Three studies, three sample sizes, one line each.

三项研究、三个样本量,各一句。

5Contribution 贡献

Ends on what changes in practice.

以实践中会发生的改变收尾。

The same move in other fields: the question sentence其他专业的同一写法:问题句

Sociology 社会学 This project asks whether a parent’s return from migrant work raises rural children’s educational aspirations, and whether it does so through parental involvement or through family resources.
Politics & IR 政治学 This project asks why local governments facing the same central inspection respond with substantive enforcement in some cities and symbolic compliance in others.
Economics 经济学 This project asks whether minimum-wage increases in Chinese cities shift low-wage workers from formal to informal employment, and for which groups of workers.
Write it last 最后再写

Write the abstract after the rest of the proposal is finished, then check that every sentence matches a section. Where a form gives an abstract box with a character limit, paste in your version and cut it there.

摘要最好在整份计划书完成后再写,并核对每句话都能对应到某一部分。如果申请表格设有限定字符数的摘要栏,把你的摘要粘贴进去后再在原处删减。

Section 1 第 1 部分
1

Background, Theory & Gap背景、理论与研究空白

Typical length 典型篇幅 500-800 words · about 26% 约占 26%

Social-science proposals fold the introduction, the literature and the theoretical framework into one section. Move from the issue to the population, name the theory or concept that organises the project, review two or three strands of evidence, and finish on the gap in one sentence.

社会科学计划书把引言、文献综述与理论框架合为一部分:从现实问题推进到研究对象,点明统领项目的理论或概念,综述两三条研究脉络,最后用一句话点明研究空白。

Blueprint结构

1Para 1 · Hook with evidence

A current fact or trend, cited.

一个有引用的当前事实或趋势。

2Para 2 · Concept & theory

The concept, defined, and the theory that explains it.

界定核心概念,并给出解释它的理论。

3Para 3 · What is known

Two or three strands, with real sources.

两三条研究脉络,附真实文献。

4Para 4 · The gap

What is missing, in one clear sentence.

用一句话写明缺什么。

Worked example · Background (paragraphs 1-4) 范例 · 背景(第 1-4 段)

1Mainland China is the largest single source of international students in UK higher education, and most arrive for a one-year taught master’s degree (Authors, Year). A year is a short time in which to build a social world: students leave established networks, adapt to a new academic culture and often live and study alongside other Chinese students rather than local peers (Authors, Year).

2I define belonging as the sense of being accepted and valued in a community (Authors, Year). The project draws on the view that belonging is a basic human need, so that its absence harms wellbeing (Authors, Year), and on acculturation research, which treats adjustment as a process that changes over time rather than a fixed state (Authors, Year). Together they predict that belonging should move during the first year, and that the moves should matter for wellbeing.

3Belonging is among the strongest known protectors of student wellbeing (Authors, Year), and loneliness among international students has been linked to depression and withdrawal from study (Authors, Year; Authors, Year). Yet most studies measure belonging once, samples are dominated by undergraduates in the US and Australia, and interventions have been tested with home students rather than with groups facing language and cultural distance (Authors, Year).

4At present we cannot say when belonging is won or lost in a one-year degree, which is exactly when a university could act.

Why it works 写法解析
1Hook with evidence 证据开头

Opens with a checkable fact and a citation, not a dictionary definition.

以可核查的事实和引用开头,而不是词典定义。

2Concept and theory 概念与理论

Defines the concept, then shows how two bodies of theory lead to a prediction about change.

先界定概念,再说明两套理论如何共同推出关于“变化”的预测。

3Strands, then limits 脉络与局限

Summarises what is known, then names three limits in one sentence.

先总结已知,再用一句话指出三处局限。

4End on the gap 以空白收尾

The last sentence is the reason the project exists.

最后一句就是这个项目存在的理由。

Section 1 continued 第 1 部分(续)

Using theory, and common mistakes运用理论与常见错误

✕ Weak 较弱“Since ancient times, people have needed friends. Nowadays, more and more students study abroad and face many difficulties.”

Generic, could open any proposal, and wastes the reader’s first 30 seconds.

空泛,可以放在任何计划书开头,白白浪费读者最初的 30 秒。

✓ Stronger 更好“Mainland China is the largest single source of international students in UK higher education (Authors, Year).”

Specific, checkable and backed by a citation. The reader knows the topic in one line.

具体、可核查且有引用。读者一句话就明白主题。

✕ Weak 较弱“This study is based on social capital theory, acculturation theory, ecological systems theory and self-determination theory.”

A list of theories with no job to do. The panel cannot see which one drives the questions.

罗列理论却不说明用途,评审看不出哪个理论在驱动研究问题。

✓ Stronger 更好“Acculturation research treats adjustment as a process, which predicts that belonging should change across the year.”

One theory, one prediction, tied to the design.

一个理论、一个预测,并与研究设计挂钩。

Theory in other fields其他专业如何写理论

Sociology 社会学 Social capital theory predicts that a returning parent adds supervision and information inside the family; resource accounts predict that lost remittances work the other way. The project tests which effect is larger.
Politics & IR 政治学 Principal-agent accounts expect compliance where central monitoring is strongest; accounts of campaign-style governance expect short bursts of enforcement that fade after the inspection team leaves.
Economics 经济学 A competitive model predicts that a binding minimum wage moves low-wage workers into the uncovered informal sector; models with employer wage-setting power predict small or no movement.

Common background mistakes背景部分常见错误

Starting too broad 开头过大

Opening with “globalisation” or the history of the discipline.

以“全球化”或学科史开篇。

Theory as decoration 理论只是装饰

A theory named in paragraph two and never used again.

第二段提到一个理论,之后再未使用。

One paper per sentence 一句一篇

A list of studies with no argument linking them to the gap.

罗列研究,却没有把它们与研究空白联系起来的论证。

Longer fields, longer theory 篇幅更长的专业,理论也更长

In politics, IR and sociology proposals of 3,000 words or more, the theoretical framework usually gets its own heading and 600-900 words: the debate, the two positions, your concepts and the expectations your cases will test.

在 3,000 词以上的政治学、国际关系与社会学计划书中,理论框架通常单设标题,约 600-900 词:写清学术争论、两种立场、你的核心概念,以及你的个案将要检验的理论预期。

Section 2 第 2 部分
2

Aims & Research Questions目标与研究问题

Typical length 典型篇幅 150-250 words · about 8% 约占 8%

One overarching aim, then two or three questions the project can answer. Each question should map to one study or one strand of data. Write hypotheses only where the design is confirmatory; exploratory and interpretive questions do not need them.

一个总体目标,加上两三个项目能够回答的研究问题。每个问题对应一项研究或一组数据。只有验证性设计才写假设;探索性与诠释性问题不需要假设。

Blueprint结构

1Aim

One sentence: what you will find out and why.

一句话:你要弄清什么、为什么。

2Research questions

Two or three, each answerable, each tied to a study.

两三个,每个都可回答,且对应一项研究。

3Hypothesis

Only where the design is confirmatory.

仅在验证性设计中使用。

Worked example · Aims and questions 范例 · 目标与研究问题

1Aim. To understand how a sense of belonging develops among Mainland Chinese postgraduate students during their first year in the UK, and to test whether a brief intervention can strengthen it.

2RQ1 (Study 1). How do belonging, loneliness and wellbeing change across the first year, and which factors predict different trajectories?
RQ2 (Study 2). Which experiences do students identify as turning points in feeling that they belong?
RQ3 (Study 3). Does a 45-minute belonging intervention at orientation raise belonging at three months, compared with an active control?

3H3. Students receiving the intervention will report higher belonging at three months than controls.

Why it works 写法解析
1Aim vs question 目标与问题

The aim says why; the questions say what will be answered.

目标说明为什么,问题说明要回答什么。

2One question, one study 一问一研究

Each question names the study that will answer it, so the plan is easy to check.

每个问题都写明由哪项研究回答,方便核对。

3A testable hypothesis 可检验假设

RQ3 is confirmatory, so it earns a hypothesis; RQ1 and RQ2 are exploratory and do not.

RQ3 是验证性的,所以有假设;RQ1 和 RQ2 是探索性的,不需要。

✕ Weak 较弱“Do Chinese students feel lonely in the UK?”

A yes/no question with an obvious answer. It cannot sustain three years.

是非题,答案显而易见,撑不起三年的研究。

✓ Stronger 更好“How do belonging and loneliness change across the first year, and which factors predict different trajectories?”

Open, measurable and built for a longitudinal design.

开放、可测量,并适合纵向设计。

Sociology 社会学 RQ1: Is parental return associated with higher aspirations, net of family background? RQ2: How do returned parents and their children describe changes in everyday involvement in schooling?
Politics & IR 政治学 RQ1: Under what conditions do local governments move from symbolic to substantive compliance? RQ2: What happens to enforcement in the year after the inspection team leaves?
Economics 经济学 H1: A 10% rise in the city minimum wage raises the share of low-wage workers in informal jobs. H2: The effect is larger for workers without local hukou.
Section 2 continued 第 2 部分(续)

Questions that fit the tradition与研究传统相符的问题

Panels in each field expect a different kind of question. A question that reads well in psychology can look naive in politics, and the reverse. Use the form your target department uses.

不同专业的评审期待不同类型的问题。在心理学中读来很好的问题,放到政治学里可能显得稚嫩,反之亦然。请使用目标院系惯用的提问方式。

FIELD 专业USUAL FORM OF QUESTION 常见提问方式EXAMPLE OPENING 示例开头
Psychology
心理学
Does X change Y, how large is the effect, and through which mechanism? Often with hypotheses.
X 是否改变 Y、效应多大、经由何种机制?常配有假设。
“Does a brief intervention raise …”
Sociology · Social policy
社会学 · 社会政策
How and why does a social pattern arise, and for whom? Hypotheses for survey strands only.
某种社会模式如何、为何产生,对谁产生?只有问卷部分配假设。
“How do families negotiate …”
Politics · IR
政治学 · 国际关系
Why do similar cases produce different outcomes? Under what conditions does an institution work?
为何相似的个案产生不同结果?制度在什么条件下起作用?
“Why do some governments …”
Economics
经济学
What is the causal effect of a policy or shock, and on which margin? Signed hypotheses.
某项政策或冲击的因果效应是什么,作用在哪个边际?配有方向明确的假设。
“What is the effect of …”
✕ Weak 较弱“What are the impacts of the minimum wage on society?”

No outcome, no group, no setting. An economics panel would read this as a topic.

没有结果变量、对象和情境。经济学评审会把它当作一个题目而不是问题。

✓ Stronger 更好“Do minimum-wage increases raise informal employment among low-wage workers in Chinese cities?”

One outcome, one group, one setting, and a direction the data can test.

一个结果、一个群体、一个情境,并有数据可以检验的方向。

Common question mistakes研究问题常见错误

Too many questions 问题太多

Six questions for a three-year project signal a lack of focus.

三年的项目写了六个问题,说明不够聚焦。

Questions that are methods 把方法当问题

“To interview 30 parents” is an activity, not a question.

“访谈 30 位家长”是一项活动,不是研究问题。

Hypotheses everywhere 处处写假设

Hypotheses for interview questions, where nothing is being tested.

给访谈类问题写假设,而这类问题并不做检验。

Section 3 第 3 部分
3

Choosing Your Methods选择研究方法

Typical length 典型篇幅 150-250 words, at the start of the methodology · about 8% 约占 8%

Start from the question, not from the method you already know. Panels check that the method can answer the question you asked, and they read quantitative, qualitative and mixed designs with different expectations. Open the methodology with a short paragraph that says which you chose and why.

从研究问题出发,而不是从你已经会的方法出发。评审会检查方法能否回答你提出的问题,并且对定量、定性与混合设计有不同的期待。请在方法部分开头用一小段说明你选了哪一种、为什么。

APPROACH 方法FITS QUESTIONS LIKE 适合的问题WHAT PANELS EXPECT TO SEE 评审希望看到
Quantitative
定量
How much, how often, for whom, and does X cause Y?
多少、多频繁、对谁、X 是否导致 Y?
A named design (survey, experiment, panel, natural experiment); validated measures or defined variables; a sample size with a reason (power analysis or the size of the available data); the analysis model; how you will handle missing data; pre-registration for confirmatory tests.
明确的设计(问卷、实验、面板、自然实验);经过验证的量表或界定清楚的变量;有依据的样本量(统计功效分析或现有数据规模);分析模型;缺失数据处理;验证性检验的预注册。
Qualitative
定性
How and why, what does it mean to the people involved, how does a process unfold?
如何、为何,对当事人意味着什么,一个过程如何展开?
A named approach (for example reflexive thematic analysis, interpretative phenomenological analysis, grounded theory, ethnography, process tracing); purposive sampling with a reason for the number; the interview guide or fieldwork plan; your position as a researcher; how quotations will be translated and anonymised.
明确的路径(如反身性主题分析、诠释现象学分析、扎根理论、民族志、过程追踪);有理由的目的性抽样及人数依据;访谈提纲或田野计划;研究者自身立场的反思;引文如何翻译与匿名化。
Mixed
混合
Questions that need both a pattern and an explanation of it.
既需要发现模式,又需要解释模式的问题。
A named design (explanatory sequential, exploratory sequential or convergent); the point at which the strands meet; what each strand does that the other cannot. “Mixing for completeness” is not a reason.
明确的设计类型(解释性序列、探索性序列或聚合式);两部分在哪一点整合;各自能做而对方不能做的事。“为了更全面”不是理由。
Worked example · Why this design 范例 · 为何选择此设计

1The three research questions need different kinds of evidence. RQ1 asks how much belonging changes and for whom, which requires repeated measurement in a large sample. RQ2 asks what the turning points mean to students, which a survey cannot capture. RQ3 asks whether an intervention works, which requires random assignment.

2The project therefore uses an explanatory sequential mixed-methods design followed by a trial. 3The strands meet at two points: Study 2 samples interviewees from the trajectory groups found in Study 1, and the turning points identified in Study 2 shape the content of the intervention tested in Study 3.

Why it works 写法解析
1Question decides method 问题决定方法

Each question is matched to the evidence it needs before any method is named.

先说明每个问题需要什么证据,再说方法。

2Named design 设计有名称

“Explanatory sequential” tells a methods reader exactly what to expect.

“解释性序列设计”让懂方法的读者立刻知道接下来是什么。

3Integration points 整合点

Says where and how the strands meet, which is what panels look for in mixed designs.

写明两部分在哪里、如何整合,这正是评审在混合设计中要找的内容。

Section 3 continued 第 3 部分(续)

Justifying the choice为方法选择提供理由

✕ Weak 较弱“I will use mixed methods because it combines the strengths of quantitative and qualitative research.”

True of every mixed study, so it says nothing about this one.

对任何混合研究都成立,所以对你的研究什么也没说。

✓ Stronger 更好“Study 2 samples interviewees from the trajectory groups found in Study 1, so the interviews explain the patterns the survey finds.”

States the integration point and what each strand adds.

写明整合点以及每部分的作用。

✕ Weak 较弱“I chose qualitative methods because the sample is small.”

Treats qualitative work as a fallback. Panels read this as a weakness.

把定性研究当作退而求其次的选择,评审会视为弱点。

✓ Stronger 更好“The question asks how parents describe changes in involvement, so in-depth interviews are the right instrument.”

The method follows from the kind of question.

方法由问题的性质决定。

The same choice in other fields其他专业的方法选择

Sociology 社会学 An explanatory sequential design: panel models estimate the association between parental return and aspirations, and 30 interviews with families sampled from the panel counties explain how involvement changes.
Politics & IR 政治学 A comparative case design: four prefectures chosen as two matched pairs that faced the same inspection round but differed in enforcement, studied through process tracing of government documents and local news.
Economics 经济学 A quantitative causal design: city minimum wages change at different times, so a staggered difference-in-differences model compares workers in cities that raised the wage with those that had not yet done so.

Common method-choice mistakes方法选择常见错误

Method first 先有方法

“I want to use machine learning” before there is a question it answers.

还没有问题,就先说“我想用机器学习”。

No integration 混而不合

Two studies side by side with no point where the results meet.

两项研究并列,结果从未交汇。

Unnamed approach 路径无名

“I will analyse the interviews” without saying how.

“我将分析访谈”,却不说用什么方法分析。

Case selection is a method too 个案选择本身就是方法

In politics, IR and sociology, the reason for choosing your cases or field sites carries as much weight as a sample-size calculation does in psychology. Say whether cases are most similar, most different, typical or deviant, and what each comparison lets you rule out.

在政治学、国际关系与社会学中,个案或田野点的选择理由,与心理学中的样本量计算同样重要。请说明你的个案属于最相似、最不同、典型还是异常个案,以及每组比较能帮你排除哪种解释。

Section 4 第 4 部分
4

Research Design & Analysis研究设计与分析

Typical length 典型篇幅 500-650 words · about 24% 约占 24%

This is the part panels use to judge whether you can do the work. Give a map of the design, then for each study say who, how many, when, what is measured and how it will be analysed. Keep ethics and recruitment for the next section unless the form asks you to combine them.

评审靠这一部分判断你能否完成研究。先给出设计总览,再逐项写明:研究对象、人数、时间、测量内容与分析方法。伦理与招募放在下一部分,除非表格要求合并。

Blueprint结构

11 · Design map

One sentence naming every study.

一句话列出所有研究。

22 · Each study

Who, how many, when, measures, analysis.

对象、人数、时间、测量、分析。

33 · Sample size

A reason for the number, including attrition.

样本量依据,含流失率。

44 · Data & openness

Data management, pre-registration, sharing.

数据管理、预注册与共享。

Worked example · Design and analysis 范例 · 设计与分析

1The project has three studies. 2Study 1 is a four-wave longitudinal survey of 400 Mainland Chinese taught-postgraduate students at three UK universities, measuring belonging, loneliness (a validated three-item scale) and wellbeing (a validated wellbeing scale) on arrival and at 3, 6 and 9 months. Growth-curve models will estimate trajectories and test predictors such as English-language confidence and the number of local friends.

Study 2 comprises semi-structured interviews with 24 students sampled from Study 1 to cover rising, falling and stable trajectories, conducted in Mandarin or English and analysed with reflexive thematic analysis. Study 3 is a pre-registered randomised trial (n = 120) comparing a 45-minute online belonging intervention at orientation with an active control, with belonging at three months as the primary outcome.

3Based on (Authors, Year), 400 participants give 80% power to detect small-to-medium effects after 30% attrition; 24 interviews allow eight per trajectory group. 4Analysis plans for Studies 1 and 3 will be registered on the Open Science Framework before data collection, and anonymised data will be shared at the end of the project.

Why it works 写法解析
1Design map 设计总览

One sentence tells the reader what is coming.

一句话让读者知道接下来是什么。

2Each study in full 逐项写全

Who, how many, when, measured how, analysed how, and in which language.

对象、人数、时间、如何测量、如何分析、用何种语言。

3Numbers with reasons 数字有依据

The survey number comes from power; the interview number from the trajectory groups.

问卷人数来自功效分析,访谈人数来自轨迹分组。

4Openness 研究透明

Psychology panels now expect registration and data sharing.

心理学评审现在一般都要求预注册与数据共享。

Section 4 continued 第 4 部分(续)

Making the design convincing让研究设计更有说服力

✕ Weak 较弱“I will use surveys and interviews to study how students feel.”

No numbers, no timing, no analysis. The panel can’t tell whether you can do it.

没有数字、时间和分析方法,评审无法判断你能否完成。

✓ Stronger 更好“Study 1: four-wave survey, N = 400, three universities, growth-curve models.”

Specific enough that a supervisor could start planning the study.

足够具体,导师可以据此开始规划研究。

✕ Weak 较弱“The data will be analysed using SPSS.”

Software is not an analysis. It tells the panel nothing about the model.

软件不是分析方法,没有告诉评审任何关于模型的信息。

✓ Stronger 更好“Growth-curve models will estimate change in belonging and test English confidence as a predictor of slope.”

Names the model and what it tests.

写明模型及其检验内容。

Design in other fields其他专业的研究设计

Sociology 社会学 Panel strand: children aged 10-15 in a national household panel, with aspirations regressed on parental return using child fixed effects. Interview strand: 30 families in two counties, 15 with a returned parent and 15 without, analysed thematically.
Politics & IR 政治学 For each prefecture, a timeline of enforcement actions is built from government work reports and inspection notices for the two years either side of the inspection; process tracing tests which account explains the sequence.
Economics 经济学 Outcome: informal employment, defined as work without a labour contract or social insurance. Model: staggered difference-in-differences with city and year fixed effects, an estimator suited to staggered timing, and an event-study test of pre-trends.

Feasibility checklist可行性清单

CHECK 检查项WHAT THE PANEL WANTS TO SEE 评审希望看到
Data 数据Where each variable or source comes from, and whether you can get it: open data, a licence you must apply for, or data you collect. 每个变量或资料的来源,以及你能否获得:公开数据、需申请许可的数据,还是自行采集的数据。
Skills 技能Methods you already use (R, Stata, NVivo, interviewing) and the training you will take for the rest. 你已掌握的方法(R、Stata、NVivo、访谈)以及为其余部分安排的培训。
Language 语言Which languages the data are in, who translates, and how translation is checked. 数据使用何种语言、谁来翻译、如何核对翻译。
Risks 风险One or two risks with a fallback, such as attrition, slow recruitment or a closed field site. 一两个风险及备选方案,如样本流失、招募缓慢或田野点无法进入。
Section 5 第 5 部分
5

Ethics, Recruitment & Access伦理、招募与进入现场

Typical length 典型篇幅 200-300 words · about 10% 约占 10%

Any project with human participants needs approval from a university research ethics committee before data collection starts. Panels read this section to judge whether you can reach your participants and protect them. Write it for your design: surveys, interviews and mixed projects raise different issues.

任何涉及人类参与者的项目,在开始收集数据前都必须获得大学研究伦理委员会的批准。评审通过这一部分判断你能否接触到研究对象并保护他们。请针对你的设计来写:问卷、访谈与混合项目面临的问题各不相同。

DESIGN 设计RECRUITMENT AND ACCESS 招募与进入ETHICS POINTS PANELS LOOK FOR 评审关注的伦理要点
Quantitative surveys and experiments
定量问卷与实验
Named channels (participant panels, mailing lists sent by gatekeepers, student societies); a target number with expected response and attrition rates; incentives; a pilot.
具体渠道(被试库、由守门人代发的邮件列表、学生社团);目标人数及预期回复率与流失率;激励措施;预试。
Online informed consent; how waves are linked without names (participant codes); data protection law; a debrief after experiments; a distress protocol and signposting when measures touch on mental health.
线上知情同意;如何在不使用姓名的情况下关联各轮数据(参与者编码);数据保护法规;实验后的情况说明;涉及心理健康测量时的危机处理方案与求助指引。
Qualitative interviews and fieldwork
定性访谈与田野调查
Purposive sampling; gatekeepers and how you will reach them; the language of the interviews; how long fieldwork takes and where.
目的性抽样;守门人及联系方式;访谈语言;田野调查的时长与地点。
Consent to recording and to being quoted; pseudonyms; your own position and any power relation with participants; translation of quotations; risks to participants in sensitive settings; secure storage of recordings.
对录音与引用的同意;化名;研究者自身立场以及与参与者之间的权力关系;引文翻译;敏感情境中参与者面临的风险;录音的安全存储。
Mixed
混合
Consent at Study 1 to be contacted for later studies; how participants move from one strand to the next.
在研究一时取得后续联系的同意;参与者如何从一部分进入下一部分。
Linking datasets safely; total burden on each participant; whether each phase needs its own approval or an amendment.
安全地关联不同数据集;每位参与者的总负担;每个阶段是否需要单独审批或提交修订。
Worked example · Ethics and recruitment 范例 · 伦理与招募

1Participants will be recruited through the international offices and Chinese student societies of three universities, which have agreed in principle to circulate the invitation; a pilot with 20 students in month 4 will test the materials. 2All studies will be approved by the university research ethics committee before data collection, and the partner universities’ procedures for external researchers will be followed.

3Because the surveys measure loneliness and wellbeing, every questionnaire will end with details of university counselling and other support services, and a distress protocol will be agreed with my supervisors. Survey waves will be linked by participant codes, not names, and data will be stored in line with UK GDPR. 4Materials will be offered in English and Chinese, translated and back-translated, and as a fellow Chinese student I will reflect on how my position may shape what interviewees choose to tell me.

Why it works 写法解析
1Named channels 具体渠道

Shows who will pass on the invitation and that a pilot tests it.

写明由谁转发邀请,并用预试检验。

2Approval route 审批路径

Covers both the home committee and the partner universities.

同时覆盖本校伦理委员会与合作院校。

3Sensitive measures 敏感测量

A distress protocol and signposting match the risk of the measures.

危机处理方案与求助指引与测量内容的风险相匹配。

4Language and position 语言与立场

Translation checks and reflexivity are what a qualitative reader expects.

翻译核对与反身性反思正是定性评审希望看到的。

Section 5 continued 第 5 部分(续)

Access in other fields其他专业如何写进入现场

✕ Weak 较弱“Participants will be recruited from universities. Ethics approval will be obtained.”

Says nothing about how, from whom, or what the ethical risks are.

没有说明如何招募、向谁招募,也没有说明伦理风险是什么。

✓ Stronger 更好“Invitations will be circulated by three international offices that have agreed in principle, with a pilot in month 4.”

Names the gatekeepers and builds in a test, which shows feasibility.

写明守门人并安排测试,体现可行性。

✕ Weak 较弱“I will interview government officials about their work.”

Officials are hard to reach and may face risk. The panel will doubt it can be done.

官员难以接触,且可能面临风险。评审会怀疑能否实现。

✓ Stronger 更好“The main evidence is published government documents; a small number of interviews with retired officials will add context where access allows.”

Builds the design on sources you can reach and treats interviews as a supplement.

把设计建立在可获得的资料上,访谈只作补充。

Sociology 社会学 The children are minors, so the project needs written consent from a parent and assent from each child, interviews held at home or school with an adult nearby, and a plan for what to do if a child discloses harm.
Politics & IR 政治学 Documentary sources carry the design. Any interviewee is anonymised by role and region, notes are stored offline, and no question asks for information the person is not permitted to share.
Economics 经济学 The household panel data are released under a user agreement; the application for access will be made in month 1, and analysis will run on the university’s secure server. No new data collection means a light ethics review.

Common ethics and access mistakes伦理与进入常见错误

Ethics as a sentence 一句带过

“Ethics approval will be obtained” with no risks named.

只写“将获得伦理审批”,不说明任何风险。

Access assumed 默认能进入

Recruiting in schools or companies with no contact or permission in place.

在学校或企业招募,却没有任何联系人或许可。

Translation ignored 忽视翻译

Interviews in Chinese with no plan for translating and checking quotations.

用中文访谈,却没有翻译与核对引文的计划。

Approval routes differ by setting 审批路径因情境而异

Every university has a human research ethics committee, but the route changes with the setting. In Hong Kong, the university’s own ethics committee reviews the study, and schools or hospitals that host your data collection may add their own permission. If you collect data in mainland China, ask your supervisor which local permissions apply.

每所大学都设有人类研究伦理委员会,但审批路径随研究情境而变。在香港,研究由所在大学的伦理委员会审查;提供数据收集场地的学校或医院也可能另有许可要求。如果你在中国大陆收集数据,请向导师确认需要哪些当地许可。

Section 6 第 6 部分
6

Contribution & Fit贡献与匹配度

Typical length 典型篇幅 200-300 words · about 10% 约占 10%

Say what the project adds to theory and to practice or policy, then why this supervisor and this department are the right place. Fit is the section applicants most often leave generic, and the one supervisors read most closely.

先写项目对理论以及实践或政策的贡献,再写为什么这位导师、这个院系是最合适的地方。匹配度是申请人最常写得空泛的部分,也是导师读得最仔细的部分。

Blueprint结构

1Contribution

To theory and to practice or policy, in two sentences.

对理论和实践或政策的贡献,各一两句。

2Why this supervisor

A named group and a specific link to your question or method.

点名课题组,并与你的问题或方法建立具体联系。

3Why this department

Data, training, centres or partners you cannot get elsewhere.

别处没有的数据、培训、研究中心或合作伙伴。

4Your readiness

Prior work that shows you can start now.

表明你可以马上开始的前期工作。

Worked example · Contribution and fit 范例 · 贡献与匹配度

1The project will contribute to belonging theory by modelling when belonging develops as well as how much of it students have, and to practice by testing an orientation intervention that universities can adopt at low cost.

2I have chosen the School of Psychology because Professor A’s group has led work on belonging and wellbeing in higher education (Authors, Year), and Dr B’s expertise in longitudinal modelling (Authors, Year) matches the growth-curve analysis in Study 1. 3The school’s participant panel and its advanced quantitative training make a multi-site study of 400 students feasible within the funded period.

4My MSc dissertation on loneliness among international students and my work as a research assistant on a wellbeing intervention gave me experience of survey design, R and trial procedures, so I can begin Study 1 in the first year.

Why it works 写法解析
1Two contributions 两类贡献

One to theory and one to practice, each in one plain sentence.

一项对理论、一项对实践,各用一句平实的话。

2Supervisor fit 导师匹配

Links one paper to the question and another to the method.

一篇论文对应问题,另一篇对应方法。

3Department fit 院系匹配

Names resources, including training, that make the project feasible here.

写明让项目在此可行的资源,包括研究方法培训。

4Readiness 准备度

Prior work shows you are ready to start.

前期工作表明你已经准备好开始。

✕ Weak 较弱“I chose this university because it is highly ranked and has a beautiful campus.”

Could be written by anyone about any university. It says nothing about the research.

任何人都可以对任何大学这样写,没有一句涉及研究本身。

✓ Stronger 更好“Dr B’s longitudinal modelling work (Authors, Year) matches the growth-curve analysis in Study 1.”

Names a person, a paper and the exact link to the project.

点名一位老师、一篇论文,以及与项目的确切联系。

Sociology 社会学 The project adds family processes to research on migration and education, and gives county education bureaus evidence on which children to support when parents return.
Politics & IR 政治学 The project explains variation within one authoritarian system rather than between systems, and links the department’s strength in comparative politics with its China research centre.
Economics 经济学 The project measures the informal-sector margin that most minimum-wage studies leave out, and uses the department’s applied microeconomics group and its access to licensed survey data.
Section 7 第 7 部分
7

Timeline & Feasibility时间安排与可行性

Typical length 典型篇幅 100-250 words or one table · about 8% 约占 8%

Match the timeline to the programme. A full-time Hong Kong PhD is usually planned over three to four years (university studentships at HKUST and CityU run for up to four years), with a qualifying examination or candidacy stage in the early years, as at CityU and CUHK. Some programmes add coursework, so check the programme page. Economics programmes often begin with coursework and a qualifying examination. Show overlaps and a fallback, not a list of year headings.

时间安排要与项目结构相符。全职香港博士通常按三到四年规划(港科大与城大的大学研究生奖学金最长资助四年),前几年设有资格考试或候选资格阶段,例如城大和港中文。部分项目另有课程要求,请以项目页面为准。经济学项目常以课程与资格考试开始。要写出重叠安排与备选方案,而不只是罗列年份标题。

Worked example · Timeline (three-year research programme) 范例 · 时间安排(三年制研究型项目)

MONTHS 月份ACTIVITY 活动OUTPUT 产出
1-4Ethics approval; pre-registration; translation and pilot of measuresApproved protocol; OSF registration
5-18Study 1: four survey waves across three universitiesDataset; paper 1 draft
9-12Upgrade review (end of Year 1; check your university)Upgrade report
12-24Study 2: 24 interviews, transcription, translation and analysisPaper 2 draft
22-32Study 3: randomised trial and follow-upTrial data
25-33Analysis and writing of thesis chaptersFull thesis draft
33-36Revisions and submissionThesis submitted

2Risk: if attrition passes 30%, missing data will be handled with multiple imputation, and a fourth university can be added in month 10.

Why it works 写法解析
1Overlap 重叠安排

Studies overlap, as real projects do, and translation time is built in.

各项研究有重叠,符合真实项目节奏,并预留了翻译时间。

2Risk and fallback 风险与备选

One risk and one fallback show you have planned.

一个风险加一个备选,说明你做过规划。

✕ Weak 较弱“Year 1: literature review. Year 2: data collection. Year 3: write-up.”

Could describe any project and hides every risk.

适用于任何项目,并回避了所有风险。

✓ Stronger 更好A month-by-month plan with ethics, overlaps, an upgrade review and a fallback.

Shows you understand how the degree runs.

表明你了解读博的真实流程。

Sociology 社会学 Fieldwork in two counties is planned for the school year in Year 2, when children and teachers are available, with the panel analysis done before it to select families.
Politics & IR 政治学 Document collection runs through Year 1; a field visit of three to four months in Year 2 adds local sources; Year 3 is for writing the four case chapters and the comparison.
Economics 经济学 Years 1-2: coursework and qualifying examinations, with the dataset built alongside. Years 3-5: three papers, the first as a job-market paper.
Section 8 第 8 部分
8

References参考文献

Usually excluded from the word count 通常不计字数 20-35 core sources

References show a panel that you know the field and the people in it. Many programmes exclude them from the word count, but some include them. HKU’s Faculty of Education, for example, excludes them from its 20-page limit, but departments and forms differ. Check before you cut.

参考文献向评审证明你了解这个领域和其中的学者。许多项目不将其计入字数,但也有项目计入。例如,港大教育学院的 20 页上限不含参考文献,但各院系与表格的规定不同。裁剪前请先核对。

Blueprint结构

Choose 20-35

Core sources, not everything you read.

核心文献,而不是读过的全部。

Include the supervisor

Cite the prospective supervisor’s work where it applies to your project.

在与项目相关时引用拟定导师的研究。

Be current

Most sources from the last five to ten years, plus the classics of your theory.

多数来自近 5-10 年,另加你所用理论的经典文献。

One style

APA 7 in psychology; Harvard or Chicago author-date in sociology and politics, unless the programme says otherwise.

心理学用 APA 7;社会学与政治学用 Harvard 或 Chicago 作者-年份格式,项目另有规定的除外。

Worked example · Reference list (excerpt) 范例 · 参考文献节选

Author, A. A., & Author, B. B. (Year). Belonging and wellbeing in higher education: a meta-analysis. Journal Name, 00(0), 000-000.

Author, C. C. (Year). Loneliness among international students: a longitudinal study. Journal Name, 00(0), 000-000.

Author, D. D., Author, E. E., & Author, F. F. (Year). A brief belonging intervention for first-year students. Journal Name, 00(0), 000-000.

Why it works 写法解析
1Real sources only 只用真实文献

In your own proposal every reference must be a work you have read.

你自己的计划书里,每条参考文献都必须是你读过的。

2Consistent format 格式统一

One style, no mixed punctuation.

一种格式,标点不要混用。

Sociology 社会学 Books and monographs carry weight alongside journal articles; cite the survey’s official documentation for the panel data.
Politics & IR 政治学 Cite the monographs that define the debate, and list primary sources (government documents, reports) separately from scholarship.
Economics 经济学 Recent working papers are normal and expected, alongside the leading journals and the papers behind your estimator.
Padding 凑数

Long lists of works you did not use.

列出一长串并没用到的文献。

Missing the main authors 漏掉核心学者

Leaving out the people the panel will assume you know.

漏掉评审默认你应当了解的学者。

Broken citations 引用不匹配

In-text citations that do not appear in the list.

正文引用在参考文献中找不到。

Putting it to work 实际应用

Adapting to each university按校调整

Keep the 2,500-word master version and cut from it. Never write a new proposal for each programme; change the fit section and trim, in the order below.

保留 2,500 词母版,在此基础上裁剪。不要为每个项目重写一份,只需按下列顺序调整匹配部分并删减。

TARGET 目标WHAT TO DO 做法
A studentship or DTP form
资助申请表或 DTP 表格
HKPFS: reference number on the RGC system by 12:00 noon HKT on 1 December 2026, then the university application HKPFS
Keep two questions, add a short paragraph on training needs and impact, and name both supervisors in the form.
保留两个问题,增加一小段培训需求与研究影响,并在表格中写明两位导师。
One page to 1,000 words
一页至 1,000 词
CityU: about 3 pages, single-spaced 城大约 3 页(单倍行距)
Cut the background to one paragraph, keep the aim and one question, describe only Study 1 in detail and one-line the rest, and fold ethics into one sentence.
背景压缩为一段,保留目标和一个问题,只详写研究一,其余一句带过,伦理并为一句。
1,500-2,500 words
1,500-2,500 词
CUHK Psychology (research proposal; the department sets the length) 港中文心理学
Use the master version; check whether references count, and add the ethics and access detail the department asks for.
使用母版;确认参考文献是否计入字数,并补充院系要求的伦理与进入细节。
3,000 words or more
3,000 词以上
HKUST Social Science 10-12 pages; HKU HKIHSS up to 8 pages plus a writing sample 港科大社会科学部、港大人文社会研究所
Expand the theoretical framework into its own section, add case-selection logic or pilot evidence, and give a fuller analysis plan.
将理论框架扩展为独立部分,补充个案选择逻辑或预试证据,并写出更完整的分析计划。
Forms, templates and character limits
表格、模板与字符限制
HKU faculty proposal form; PolyU Form GSB/1A with a personal statement; HKUST research plan in the online form (as well as the proposal) 港大学院表格、理大 GSB/1A、港科大网申研究计划
Paste each section under the template headings; for character limits, about 12,000 characters is roughly 1,800-2,000 English words.
把各部分粘贴到模板标题下;字符限制方面,12,000 字符大约相当于 1,800-2,000 个英文单词。
Supervisor agreement first
先取得导师同意
CityU (send the proposal to a staff member before applying) 城大
Send the supervisor a two-page summary with the question, design and sample, plus your CV; write the full proposal with their comments once they agree.
先把包含问题、设计与样本的两页摘要和简历发给导师,对方同意后再结合其意见写完整计划书。

The email before the proposal计划书之前的那封邮件

In Hong Kong your first email is part of the application. CityU asks for a supervisor’s consent before you apply, and because universities pre-screen and nominate HKPFS candidates, a supervisor who already knows your plan helps everywhere. Write in September or October, well before 1 December. Give the supervisor enough detail to judge the fit: one specific link to their work, your question and method in one sentence each, one easy question, and your CV plus a short summary attached. Write the first draft yourself; your academic will check the tailored paragraph.

在香港,第一封邮件就是申请的一部分。城大要求申请前先取得导师同意;由于各校会先筛选并提名 HKPFS 候选人,一位已经了解你计划的导师在任何学校都有帮助。请在 9 月或 10 月发出,远早于 12 月 1 日。写得足够详细,让导师能判断是否匹配:联系对方一项具体研究,用一句话写研究问题、一句话写方法,只问一个容易回答的问题,并附简历和一份简短摘要。第一稿由你自己撰写,导师只负责核对针对导师的那一段。

Before you submit 提交之前

Read it like a panel用评审的眼光重读

Reread your proposal as a stranger with ten minutes and twenty others to read. Score each question 1 (weak), 2 (adequate) or 3 (strong), and fix the lowest score first.

像一个只有十分钟、还有二十份要读的陌生评审那样重读。每项打 1(弱)、2(一般)、3(强)分,先修改得分最低的一项。

#CHECK 检查SELF-SCORE 自评
1Can I state the gap in one sentence from memory? 我能凭记忆用一句话说出研究空白吗?1 2 3
2Does one theory or concept do visible work in my questions and design? 是否有一个理论或概念在研究问题和设计中看得出在起作用?1 2 3
3Does every research question map to a study and a method? 每个研究问题都对应一项研究和方法吗?1 2 3
4Have I said why this method, and named the design and the analysis? 我说明了为何用此方法,并写明了设计与分析吗?1 2 3
5Have I shown how I will reach participants or data, and the main ethical risks? 我说明了如何接触研究对象或数据,以及主要伦理风险吗?1 2 3
6Is the fit section specific to this supervisor and department? 匹配部分是否只适用于这位导师和这个院系?1 2 3
7Is it the right document, within the limit, with references counted as the programme says? 文件类型正确、字数符合要求、参考文献按项目规定计算了吗?1 2 3

Three tests三个测试

The sentence test

Pick any sentence. Can you say which of the five panel questions it answers? If not, cut it.

随便挑一句,你能说出它回答了评审五问中的哪一个吗?不能就删掉。

The supervisor test

Would the named supervisor recognise their own work in your fit section? If not, rewrite it.

被点名的导师能在你的匹配部分认出自己的研究吗?不能就重写。

The ethics test

Imagine the ethics committee reads Section 5. What is their first question? Answer it in the text.

想象伦理委员会读第 5 部分,他们的第一个问题是什么?在正文中回答它。