在完成专业与地区的选择后，由于申请费用与精力分配的限制，申请者需要在众多的学校/项目中选择出最合适自己的几个。这个数字从个位数到几十个不等，但核心目的是要找到最 fit 的那一些，进而与潜在导师的选择也密不可分。从我自己的经验来看，学校/项目的选择不仅需要对 stipend/placement 此类硬性指标有所了解，也要明确氛围/方向这类的软性指标。相对来说，潜在导师的选择可以更模糊一些，但筛选导师列表本身也是帮助学校选择的过程。由于两者联系较为紧密，这条帖子将给出整体选校的建议步骤，而并不拆分成两个分开的部分。
潜在导师搜寻
在确定具体的一个或几个专业以及地区的选择之后，我认为第一步应当是搜寻目标导师而非直接去看学校/项目。这是因为学校/项目排名无法代表你所感兴趣的方向上这些项目的优势。举个简单的例子，某些学校的社会学系排名高是因为他们在人口与健康领域有很好的老师，但在文化方面基本不涉及。因此，通过潜在导师对学校/项目进行初筛是一个比较合理的行为。
潜在导师搜寻的过程分为两步。第一步，借助已知知识搜寻领域内的老师及其学生与合作者。这一点往往需要申请者对于自己感兴趣的领域已具备一定量的阅读与了解，知道该领域的大牛/rising star 大概是谁。同样地，以这些学者为核心，我们可以找到他们的学生与合作者，并将他们也列为潜在导师。这一步可以迭代 1-2 次。这一步也可以通过询问本科/硕士导师或是在读 PhD 等熟悉这个领域的人来完成。
第二步，浏览这些潜在导师所在系所的 core faculty 列表。这一步的目的有二：首先，我们可以通过这一步找到一些可能遗漏的潜在导师；其次也是最重要的，通过浏览所有老师的研究方向，我们可以确定一个系所的大致擅长的研究领域和氛围。例如，尽管有一个做文化的老师在某一个系，但如果其他老师大部分研究移民与种族议题，那意味着这个系在文化上并不具备强项，甚至可能老师本人也不在权力中心。
一旦确定一个学校/项目有 3+ 的老师符合你对于研究话题（研究话题可以是很具体的内容、略微宽泛的理论取向，或者是某种特定的方法）的兴趣，那么就可以将这个系所纳入你的第一阶段选校列表。
如果你的时间并不充裕，可以尽量专注于导师搜寻过程并多次迭代，直到有一个数量较为符合预期的选校列表。我推荐将导师记录下来，以供文书写作使用。
学校/项目状况确认
当一个系所进入选校列表，即意味着它的软性条件基本与我们是 fit 的。一个较为容易被忽视的点在于，尽管并不必要，但对其某些非学术条件进行确认可以帮助我们更好地判断。这一过程包括：地理位置、stipend、placement、整体氛围以及往年及当年的招生情况。
地理位置与 stipend 基本属于左右未来很多年生活的条件，但在选校阶段可以暂时忽视。每个人有不同的生活癖好与用钱习惯，并没有一个固定的判断标准可以适用于所有人，所以一个最好的方式就是搜寻相关信息来了解当地生活状态以及 PhD 生活状态。
placement 直接联系到这个系所本身的发展状态。在选校阶段，这个可以作为学校学术水平的代理变量。一般的系所都会在官网放出自己往年的 placement 情况。我曾听说过：如果一个系不放出自己的 placement，那一定是有点问题。
整体氛围不仅包括学术风气，也包括诸如多样性、师生关系、甚至系内斗争（有无）等情况。多样性对于亚裔来说可能很重要，因为一个当前 faculty 与学生中没有什么亚裔（或中国人）的系所往往意味着它在招生中也不会选择亚裔。师生关系等问题需要去跟在读 PhD 进行了解，也可以拖到最终选 offer 环节来进行。
最后，近年的招生政策（包括规模、多样性等信息）也很重要。尤其在 funding cut 情况下，招生政策的变化很快，例如尽管以前亚裔招收数量较多，但去年锐减，这可能说明学校在减少风险；甚至是否招生都会发生很大的变化。
套磁
在决定申请某一个系之后，一个抉择可能在于是否要向最有意向的导师发送套磁信来表达自己的兴趣。这个选择本身首先取决于系里的招生政策。基于套磁信的作用，招生可以分为严格 committee、强 committee 与强导师制度。严格 committee 往往要求申请人不得向潜在导师发送邮件，老师也会拒绝看套磁。这些项目往往严格落实 committee 制度，不过导师依旧会根据自己对于申请人研究方向的兴趣来确定录取。强 committee 尽管说明套磁信没有太大用处，但实际上老师还是会因为套磁而对申请人有一定的事先了解。从身边统计学来说，套磁对于申请这些项目有一定的帮助。强导师制则更不必多说：导师往往对于申请有绝对的主导权，套磁一般是必不可少的一部分。
尽管社会学系是一个极其强 committee 甚至严格 committee 的专业，但我依旧支持大家多发套磁，原因有三。首先，套磁信尽管可能无用，但一般（认真写的）信不会产生负面效果。这就意味着套磁本身作为额外的努力，有可能产生意料外的效果。其次，套磁本身是一个获得信息的过程。正如上文所说，在 funding cut 情况下，每个学校的招生政策的变化很快，faculty 是对于这些政策最为了解的人。比如某几个潜在导师就在给我套磁的回复中明确提及今年会减招甚至不招。第三，套磁的回复（或者不回复）本身也是一种 signal，可以提前明确自己的定位。当收到模板回复甚至收不到回复的时候，我们就大致知道自己并不在这些老师的考虑范围内（但也有很小概率是这些老师过于忙碌），进而可以降低自己的预期以及修改自己申请的范围。
当然，对于套磁的选择见仁见智。在时间有限的情况下，完善自己的选校与文书可能更为重要。至此，我们应当有了一个较为固定的选校列表。
学校划分与保底校/彩票校的（不）分类
现在获得的这一张选校列表往往并不存在某一种排序，或者说也只是潜在导师影响力/能力/fit/意愿的某种排序。然而人的分类本能会让我们想要给这些学校一个阶梯划分，并明确什么是保底。我的观点是，我们应当给这些学校几种不同的阶梯，同时保底校与彩票校的分类本身可能是不必要的。
我们要明确，申请 PhD 本身是一个多维度的过程：你能否进入某个学校取决于各种因素而非学校本身的好坏。所以，在给学校排序的过程中，我们也最好要明确多维的性质。有几个不同的排序方式是较为合理的：1. 按照 fit。如果对于潜在导师有过了合理的筛选，你应当能够测量你和一个系的匹配程度，这也是申请中最重要的几个因素之一。2. 按照意愿强度。根据你对于这个系总体的认知，你会有一个直觉上想去或不想去一个系的意愿，这可能可以帮你区分出一些并不适合的学校。3. 按照系的排名（QS/USNews 等）。尽管并不全面与客观，一个系的排名是最为顺理成章的分类法。对世俗标准的看重也可以让我们从这些排名中筛选掉一些并不符合预期的学校。
正因为我们对所有学校有很多种排序方式，保底校和彩票校本身并不应该存在。我们并没有把握进入在任意维度上排名最低 or 最高的系，相反所有的系我们都有概率进。这种划分本身可能会带来很多负担。
至此，你应当拥有了一张具有清晰顺序（可以通过不同维度进行 sort）的选校列表。
选校后记
我自己的选校可能并不完全顺利，但现在想来却有那么多的"要是当初这样就好了"。于是，我把很多我未能实践但看到他人实践的内容也囊括在了这里，希望能帮到大家。
总之，恭喜你完成了 PhD 申请的第一步！
选校只是一个开始，但也是最为重要的基础。
After finalizing your choice of major and region, and given the constraints of application fees and limited energy, applicants need to select the most suitable programs from the many available schools. This number can range from a handful to several dozen, but the core goal is to find the best-fit options—which is also closely tied to the selection of potential advisors. From my own experience, choosing schools/programs requires understanding not just hard metrics like stipend and placement records, but also softer indicators like departmental culture and research orientation. The selection of potential advisors can be somewhat more exploratory, but the process of building your advisor list itself helps refine your school selection. Since the two are closely connected, this post offers holistic advice on school selection as an integrated process rather than splitting it into two separate parts.
Searching for Potential Advisors
After identifying one or more specific disciplines and regions, I believe the first step should be searching for potential advisors rather than looking directly at schools/programs. This is because school/program rankings cannot capture the strengths of a given program in your particular area of interest. For example, some schools have highly ranked Sociology departments because they have excellent faculty in demography and health, but barely engage with cultural sociology. Therefore, using potential advisors to pre-screen schools/programs is a more reasonable approach.
The process of searching for potential advisors involves two steps. The first is using your existing knowledge to identify scholars in the field along with their students and collaborators. This often requires applicants to have done considerable reading and have some familiarity with the field—knowing roughly who the major figures and rising stars are. Using these scholars as a starting point, you can find their students and collaborators and add them to your list of potential advisors. This step can be iterated 1–2 times. It can also be completed by asking your undergraduate or master's advisor, or current PhD students familiar with the field.
The second step is browsing the core faculty lists of the departments where these potential advisors are located. This serves two purposes: first, you may discover potential advisors you had missed; second and more importantly, by reviewing all faculty members' research interests, you can get a sense of a department's overall strengths and academic culture. For example, even if there is one cultural sociologist in a department, if most other faculty study immigration and race, this suggests the department is not particularly strong in cultural sociology—and that faculty member may not be in a position of central influence.
Once you determine that a school/program has 3+ faculty members whose work aligns with your research interests (whether that means a specific topic, a broader theoretical orientation, or a particular method), you can add that department to your initial school list.
If time is limited, focus on the advisor search and iterate multiple times until your list reaches a manageable size. I recommend keeping track of potential advisors for later use when writing application materials.
Confirming School/Program Details
Once a department enters your school list, it means the soft conditions are generally a good fit. One point that's easy to overlook: while not strictly necessary, confirming some non-academic conditions can help you make more informed assessments. This includes: geographic location, stipend, placement record, overall departmental culture, and recent admissions trends.
Geographic location and stipend are conditions that will shape many years of your life, but can be temporarily set aside at the school selection stage. Everyone has different lifestyle preferences and spending habits, so there is no fixed standard applicable to all. The best approach is to seek out information about local living conditions and PhD student life.
Placement records directly reflect the health of a department. At the school selection stage, this can serve as a proxy for the department's academic standing. Most departments post their placement records on their official websites. As the saying goes: if a department doesn't publish its placement record, there's probably a reason.
Overall departmental culture includes not only the academic environment but also matters such as diversity, faculty-student relationships, and even internal politics (or lack thereof). Diversity may be particularly important for Asian applicants, as a department with few Asian (or Chinese) faculty and students often signals that it also tends not to admit Asian applicants. Faculty-student relationship questions are best explored through conversations with current PhD students and can also be addressed later when making your final offer decision.
Finally, recent admissions policies (including scale, diversity, etc.) are also important. Especially in the context of funding cuts, admissions policies can change rapidly—for example, a sharp drop in the number of Asian students admitted last year may signal the school is managing risk; even whether a program is admitting students at all can change significantly.
Cold Emailing
After deciding to apply to a particular department, one decision you may face is whether to send a cold email to your most preferred potential advisors to express your interest. This depends first on the department's admissions structure. Based on the role cold emails play, admissions can be categorized as: strictly committee-based, strong committee with some advisor influence, and strong advisor-based. Strictly committee-based programs often ask applicants not to contact potential advisors, and faculty will decline to read cold emails. These programs implement committee decisions strictly, though advisors still factor in their own interest in an applicant's research direction. Strong committee programs, while indicating that cold emails are not particularly impactful, in practice still lead some faculty to develop prior familiarity with applicants. From those around me, cold emailing does seem to have some benefit for these programs. Strong advisor-based systems are even more straightforward: advisors typically have near-complete authority over admissions, and cold emailing is generally an essential component.
Even though Sociology departments tend to be extremely committee-driven or even strictly committee-based, I still encourage everyone to send more cold emails, for three reasons. First, while a cold email may not help, a genuinely written one generally won't hurt either. This means that as an extra effort, cold emailing has the potential for unexpected positive outcomes. Second, cold emailing itself is an information-gathering process. As mentioned above, in the context of funding cuts, each school's admissions policies can change quickly, and faculty are the most informed people about these changes. For example, several potential advisors explicitly mentioned in their replies to my cold emails that they would be reducing admissions or not taking students at all that year. Third, responses (or non-responses) to cold emails are themselves a signal that can help you calibrate your standing in advance. When you receive a template reply or no reply, you can reasonably infer that you're not in these professors' consideration range (though there's a small chance they're simply very busy), allowing you to adjust your expectations and refine the scope of your applications accordingly.
Of course, the decision to cold email is a personal judgment call. When time is limited, refining your school list and application materials may be more important. At this point, you should have a fairly finalized school list.
School Categorization and the (Un)Necessity of Safety & Reach Schools
The school list you have at this point generally doesn't follow any particular ranking—or at best, it reflects some combination of potential advisors' influence, ability, fit, and willingness. Yet the human instinct to categorize will lead you to want to sort these schools into tiers and identify what counts as a "safety." My view is that you should assign these schools multiple different rankings, while the categories of "safety school" and "reach school" are themselves probably unnecessary.
We need to recognize that applying to a PhD program is a multidimensional process: whether you get into any given school depends on various factors beyond just the school's prestige. So in ranking schools, it's worth being explicit about this multidimensionality. Several different ranking approaches are reasonable: 1. By fit. If you've done a careful screening of potential advisors, you should be able to gauge your compatibility with a department, which is one of the most important factors in admissions. 2. By preference intensity. Based on your overall impression of a department, you'll have an intuitive sense of whether you want to go there or not, which may help you eliminate programs that don't genuinely appeal to you. 3. By department ranking (QS/USNews, etc.). While not comprehensive or objective, a department's ranking is the most natural classification method. Attending to conventional benchmarks can also help you filter out programs that fall short of your expectations.
Precisely because there are so many ways to rank the same set of schools, the categories of "safety school" and "reach school" shouldn't really exist. You have no guarantee of getting into the program ranked lowest on any given dimension, and conversely, you have some real probability of getting into any of them. This categorization may only add unnecessary burden.
At this point, you should have a school list with a clear structure—one that can be sorted by different dimensions.
Postscript
My own school selection process wasn't entirely smooth, but looking back, there are so many "I wish I had done this differently" moments. So I've included here many things I didn't manage to do myself but have seen others implement successfully, hoping they will be helpful.
In any case, congratulations on completing the first step of the PhD application process!
School selection is just the beginning, but it is the most important foundation.