Applying to Social-Science PhD Programs in North America如何申请北美社科类博士项目

Should you do a PhD?

Two things should be true before you apply. First, a genuine interest in the questions — you actually care about the economics, or the politics, or the social problem, enough to sit with it for years. Second, a clear-eyed career decision. A PhD takes five to six years at minimum, on a modest stipend; treat those years as an investment in your own human capital, and as a job. The application is only the first step — even a great offer is where the real test begins. Read something honest about what doctoral life is actually like before you commit.

How competitive it is

Admissions are very selective, and it helps to see the scale. A strong political science program might admit ten to fifteen students a year out of several hundred applications; economics cohorts are larger but draw even more applicants. Either way, the odds are long.

Two beliefs get in the way. The first is that a PhD from a well-known program guarantees a job in academia. It does not — academic job markets are hard for everyone. The second is that perfect numbers guarantee admission: a 4.0, a near-perfect TOEFL, a high-percentile GRE, and you are in. Also untrue. Grades and language scores correlate with research potential, so they matter — but they are not the thing itself. Committees are trying to find people who will do good research, and a transcript alone does not prove that. And with grade inflation, most applicants now have very high grades anyway. The useful move is to stop optimizing the numbers and start asking where you stand as a researcher, and how to show it.

What you'll prepare

Four kinds of material, plus letters:

• Transcripts — undergraduate, and graduate if you have them.
• Standardized tests — the GRE for most programs; TOEFL for those whose first language is not English or who are applying from a non-English-speaking country.
• A statement of purpose — what you want to research, and how well you understand the field.
• A writing sample — many programs call it optional, but for showing research potential it really matters.
• Recommendation letters — among the most important parts of the file.

The rest of this guide is mostly about the last three, because that is where applications are won or lost.

The personal statement

The most common mistake — I made it myself — is to pour effort into GPA and standardized tests and treat research as an afterthought. The reason is understandable: progress on grades and test scores is visible and gratifying. But from the committee's side, the central question is whether you can show, with evidence, that you might become an independent scholar. A statement that only recites achievements does not answer it.

A good statement shows that you understand the area you want to work in. You know roughly where the literature stands, where it is stuck, what you would push on — perhaps a question your past work already started to ask. That is the signal: not your feelings about the subject, but your grasp of it. If you want to work in an area field — say China studies, or European politics — the thing that lands is knowing that field's scholarship, not a personal connection to the place. (This matters more in fields like political science and sociology than in economics, where training is more standardized and committees expect less polished research from undergraduates, though that is changing.)

It follows that sending identical materials to every program is a mistake. Departments differ — one is strong in this subfield, another in that. If you know enough to align your statement with the people whose work fits yours, the application reads very differently. A committee is weighing two things: (1) whether you are likely to grow into a strong researcher; and (2) whether the department, in the area you care about, has the resources to help you succeed. Make it easy for them to see both: that you have the potential to do good research, and that you fit the department.

Recommendation letters

Letters carry heavy weight. Applying from outside the established North-American academic network carries a real information disadvantage — not because referees elsewhere are less able to judge talent, but because academia is partly a circle of information, where the quality of a letter depends on trust between people who know each other. A committee weighs a letter partly by how well it knows and trusts the writer.

This is also why chasing big-name scholars is not necessarily the best strategy. If a big name is too busy to write an informative letter, the committee can tell at a glance. A detailed, concrete letter from someone who genuinely knows your work is far stronger: it names what you did, how you think, the problem you cracked, the tools you have learned. That specificity is itself strong evidence. So choose referees who can speak to your real strengths, not the most decorated professor whose course you happened to take.

Committee members pay attention to the specific details that reflect your research potential.

The writing sample

If the statement signals that you understand a field, the writing sample is the strongest single piece of evidence that you can actually do research. Graduate admissions are not like undergraduate teaching: a department is choosing future colleagues in research, not meeting an obligation to teach. It looks for the people most likely to thrive in research, and a sample that shows you know how research is done — how a question turns into a research output — tilts the decision your way.

A few things people get wrong:

• It does not need to be published. Committee members read and referee papers constantly; they can judge quality at a glance. Publication in a venue they don't recognize tells them little, because they can't translate it into a level. A strong unpublished paper that looks like good research is a strong signal.
• Ideally it is not co-authored — the committee wants to confirm the work is yours. Co-authoring with a mentor is fine on a larger project, but it is best for the mentor to spell out in the letter which part was yours.
• Keep it tight. Around 20-30 pages, with a clear, sharp abstract — readers skim fast, so the first page has to land. Use plain, precise language. The academic norm is clarity, not ornament; flowery prose signals the opposite of what you want. Length does not demonstrate ability; what counts is quality.

The application timeline

The timeline most people use is the one I used, and it is the wrong one: take the standardized tests by September of the application year, then spend two rushed months that fall assembling everything and shipping it off. The better approach front-loads the tests so you can spend six months to a year on the thing that actually matters — developing a strong writing sample and choosing programs that fit you.

How admissions committees actually work

It is less mysterious than it looks, and the general shape is no secret, though the specifics vary by department. A rotating committee reads the files in two or three passes. An early pass sets aside applications that clearly fall short: a glaring weakness — a very low score, or a poor grade in a course that matters — is risky, simply because there are many comparable applicants without it. Later passes weigh the rest more closely against the number of places available, and some departments draw on subfield colleagues or interviews. Throughout, the committee is looking for the one thing this guide keeps returning to: evidence of research potential. In my own experience the process is conscientious and evidence-based; the real difficulty is that the signal of research potential is noisy.

Interviews. Many programs don't do them. If you get one, that is good news — you are already on a short list. An interview checks what is hard to read off paper: how deeply your interests really run, and how you think on your feet. Strong writing with weak speaking is a common gap, and the interview closes it.

For international applicants

Applying from outside the North-American system carries a specific challenge: information asymmetry. Across a distance, with less context the committee can verify, the burden is on you to supply credible, verifiable evidence. A few implications:

• Letter writers outside academia carry less recognized weight, partly because the qualities they prize may not speak to research potential in the field you are applying to, but a detailed, substantive letter still transmits a great deal, so choose for specificity over fame.
• A research master's or pre-doc can help — usually not because the credential or position itself raises your odds, but because it gives you time to take courses, polish your materials, learn how research is done, and produce a stronger writing sample.
• Reaching out to faculty does no harm, but don't count on it: admission decisions are mostly made by the committee.
• Language and interviews matter obviously — being able to communicate clearly about your research is itself an important skill for doctoral work.

A few common questions

Can I apply straight from undergrad? Yes. Strong undergraduate applicants are very welcome — with more room to grow. A master's helps mainly by giving you time, faculty access, and a better writing sample, not through the credential itself. If your evidence already shows research potential, apply.

Can I switch fields? Yes, with two things in place: training in the basic tools of quantitative (or qualitative) research, and a genuine, informed interest in the literature of the field you are applying to. The real work is convincing a committee that your interest in the new field is real, not a flight to something that sounds prestigious.

What if my work is qualitative rather than quantitative? Qualitative research is valued in political science and sociology — a great deal of political science is field and interpretive work, and committee members get excited by a strong qualitative applicant, because they are rarer. The challenge is the same as everywhere: stand out by producing a genuinely strong piece of qualitative research that you can put on the table.

A note on this version

This guide is adapted from a talk I gave in 2017 for CNPolitics (政见), edited from a transcript into written form and lightly updated. The framing — that admissions are an evidence-based search for research potential — has held up well; a few specific figures are illustrative of that period rather than current numbers. It reflects my own experience in economics and political science, and is offered as general guidance, not professional advice.

↑ Back to top

要不要读博

在申请之前，有两点最好是成立的。第一，你对研究的问题本身有真正的兴趣——是真的关心这些经济学、政治学或者社会问题，关心到愿意为它坐上好几年冷板凳。第二，要有一个清醒的职业判断。读博至少要五六年，收入也比较微薄；这几年既是对你人力资本的一笔投资，也是一份工作。申请只是第一步，即使拿到很好的offer，真正的考验才刚刚开始。在做决定之前，最好找一些诚实讲述读博生活的文字读一读。

申请有多激烈

这几年的竞争非常激烈，先看清楚它的规模会有帮助。一个排名靠前的政治学项目，每年大概从数百份申请里录取十到十五人；经济学的班级更大，但申请的人也更多。不管哪种情况，胜算都不高。

有两个误区会成为障碍。第一个是，从知名项目拿到博士学位就等于有了学术界的工作。并不是这样——学术就业市场对所有人都不容易。第二个是，以为把各项数字都做到极致就一定被录取：4.0的GPA、接近满分的托福、高百分位的GRE，录取就稳了。这也不对。成绩好、英语好，和研究潜力当然是正相关的，所以它们看起来重要——但它们本身并不就是研究潜力。老师想找的是未来能做出好研究的人，而单凭一张成绩单证明不了这一点，而且由于grade inflation，现在多数申请人的成绩都非常好。

与其继续优化那些数字，不如换个角度想一想：在“研究者”这一类人里，我大概处在什么位置？我又怎样把它显示出来？

需要准备哪些材料

四类材料，再加上推荐信：

• 成绩单——包括本科成绩单和研究生成绩单（如果读过研究生的话）。
• 标准化考试——大部分项目都需要GRE；大部分母语不是英语的同学、直接从非英语国家申请的同学都需要托福。
• 个人陈述——告诉委员会你想研究什么，展示你对这个领域的理解。
• 写作样本——很多项目说它是可选项，但对显示研究潜力来说，它其实很重要。
• 推荐信——材料中最重要的部分之一。

这篇指南接下来主要讲后面三项，因为申请的成败往往就在这里分出。

个人陈述

我见过的最大的一个误区，也是我自己当年犯过的——就是把精力都放在GPA和英语上，把研究看得很轻。这可以理解：成绩和考分的进步看得见，也让人有成就感。但在老师那一边，最要紧的问题是：你能不能拿出证据，说明自己有可能成为一个独立的学者？一份只罗列成就的陈述，是回答不了这个问题的。

一份好的个人陈述，能让人看出你对想研究的领域确实有所了解。你大致知道这个领域的文献走到了哪里、卡在哪里、你想往哪个方向推进——也许就是你过去的研究已经开始触及的某个问题。这才是真正的信号：打动人的不是你对这个主题有多少感情，而是你对它有多少把握。如果你想做某个区域研究——比如中国研究、欧洲政治——真正管用的，是你了解这个领域的学术脉络，而不是你跟那个地方有什么个人渊源。（这一点在政治学、社会学这些学科里比在经济学里更重要：经济学训练比较标准化，从前委员会原本也不太指望本科生能写出非常成熟的研究——这一点正在发生变化。）

也正因为这样，给所有项目寄一份完全一样的材料是个误区。各个系是有差别的——这个系在这个方向强，那个系在那个方向强。如果你对它们了解得够多，能把陈述和那些跟你方向契合的老师对上号，整份申请给人的感觉就很不一样。录取委员会的老师在掂量两件事：（一）你是否有可能成长为一名优秀的研究者；（二）这个系在你感兴趣的领域，是否有足够的资源帮助你成功？你要做的，就是让他们一眼就看出来你有做出好研究的潜力，以及适合这个系的特点。

推荐信

推荐信分量很重。从北美既有的学术圈之外申请，有不小的信息劣势——倒不是说别处的老师鉴别人才的能力差，而是因为学术圈某种程度上也是一个信息圈，信息的质量往往依赖于彼此认识的人之间的信任。委员会读一封推荐信，有几分看重，一部分取决于它对写信的人有多了解、多信任。

也正因为这样，一味去追大牌学者的推荐信，未必是最好的策略。如果一位大牌教授因为工作太忙，不能给你写一封有信息量的推荐信，委员会一眼就能看出来。一封出自真正了解你工作的人、写得很具体的信，会有用得多：它会写清楚你做了什么、你怎么想问题、你啃下了哪个难题、你会用哪些工具。这种具体本身就是很强的证据。所以要找那些真能讲清你长处的人写信，而不是因为他是你上过课的、最有名的那位老师。

委员会的老师们会更关注那些能够反映出你研究潜力的细节。

写作样本

如果说个人陈述显示的是你了解一个领域，那么写作样本就是你能不能做研究最有力的一份证据。研究生录取和本科教学不一样：本科教学里，老师有把你教好的责任；而博士录取更像是系里在挑选未来一起做研究的人。各个系挑的是最有可能在研究上走得远的人，而一份能看出你懂得研究是怎么做的样本——一个问题怎样一步步变成一个项目、再变成一个结果——会把天平向着你这边压。

有几点是人们常常弄错的：

• 它不一定要是已经发表的文章。委员会的老师经常读论文、审稿，质量好不好，他们扫一眼就有数。在他们不熟悉的刊物上发表，其实说明不了什么，因为他们没办法把它换算成一个水平。反过来，一篇没发表、但看上去就“像那么回事儿”的好文章，本身就是很强的信号。
• 最好不是合作的作品。委员会想确认这是你自己的工作。如果是比较大的项目，和导师合作署名也没问题，但最好由导师在信里说明，哪一部分是你做的。
• 篇幅要紧凑。长度在20-30页左右即可，配一个清晰、利落的摘要——老师们看得很快，所以第一页一定要立得住。用平实、准确的语言。学术界要的是清晰的逻辑，不是华丽的辞藻；堆砌空洞的文字，反而是反信号。清晰的图表往往会加分。长度并不能展现你的能力，更重要的是质量。

申请的时间线

大多数人用的时间线，也是我当年用的那一种，其实是错的：九月之前把英语考完，然后在那个秋天用两个月时间手忙脚乱地把材料凑齐寄出去。更好的做法是早早把英语解决掉，这样你能腾出半年到一年，去做真正要紧的事——打磨一份扎实的写作样本，再挑选和你契合的项目。

录取委员会如何运作

录取委员会其实没有看上去那么神秘，大致的流程也算不上什么秘密——只是各个系的具体做法会有所不同。一般来说，一个轮值的委员会会分两到三轮来看材料。靠前的一轮，会先把明显达不到门槛的申请放到一边——一个明显的短板（比如某项成绩特别低，或某门要紧课程的分数较差）是有风险的，原因很简单：有很多条件相近、却没有这个缺陷的申请人。之后的几轮，会把剩下的材料对照当年的名额更仔细地比较，有些系还会请相关方向的同事帮着看，或者安排面试。但自始至终，委员会找的都是这篇文章反复强调的那件事：研究潜力的证据。就我的个人经历而言，这个过程相当审慎、以证据为基础；真正的困难在于，关于研究潜力的信号往往非常嘈杂。

面试。多数学校没有面试环节。但如果你拿到了面试，那是好消息——说明你已经在一份很短的名单上了。面试看的是材料里不容易读出来的东西：你能不能把自己感兴趣的研究讲清楚、你说的那些兴趣到底有多深、你临场怎么想问题。“写”得好、“说”不行，是一个很常见的落差，面试正好把它补上。

给国际申请人的几点建议

从北美体系之外申请，会遇到一个特别的难处：信息不对称。隔着距离，委员会能核实的背景更少，于是举证的担子就落在你身上——你得拿出可信、能核查的证据。这里有几点：

• 学术圈之外的推荐人，分量会被打一些折扣，因为他们看重的品质，并不一定能够指向你申请的方向的研究潜力——但一封具体、有内容的信仍然能说明很多问题，所以选人要看具体，不要光看名气。
• 研究型硕士和predoc可能有帮助，通常不是因为这个学位或经历本身提高了你的胜算，而是它给了你时间听课、打磨材料、学习研究方法，写出一份更优秀的写作样本。
• 主动联系老师，不至于有什么害处，但不要寄希望于此，因为录取决定主要是由委员会决定的。
• 语言和面试，对国际申请人来说往往更加重要，能清晰地进行学术交流，是博士阶段研究的一项重要技能。

几个常见问题

本科可以直接申请吗？ 当然可以。一般而言，委员会欢迎优秀的本科申请人——成长空间更大。硕士的好处主要是给你时间、给你接触老师的机会、帮你做出一份更好的写作样本，而不在学位本身。如果你的材料已经能看出研究潜力，那就申请。

申请的方向可以和本科学的方向不一样吗？ 可以，前提是两点到位：一是基本的量化研究（或定性研究）工具的训练；二是对申请领域的文献有真实、有根据的兴趣。真正要下的功夫，是让委员会相信你对新领域是真的有兴趣，而不是冲着一个听上去比较光鲜的方向去的。

如果我做的是定性研究，而不是定量研究呢？ 定性研究在政治学、社会学很受重视——政治学里有很大一部分就是田野和阐释性的工作，委员会的老师看到一位优秀的做定性研究的申请人，也会很兴奋，因为这样的申请人就不多。难处和别处一样：要靠一份拿得出手、真正扎实的定性研究脱颖而出。

关于本版本

本文整理自我2017年为政见（CNPolitics）做的一次讲座，由讲座文字稿编辑而成，并做了少量更新。里面的核心框架——录取是一个以证据为基础、寻找研究潜力的过程——这些年基本没有过时；个别具体数字说的是当时的情况，不是现在的数据。本文是我个人在经济学和政治学方面的经验之谈，只作一般性参考，并非专业建议。

↑ 回到顶部