Notes for Prospective PhD Mentees

If you are considering a PhD, I am glad you are looking into it carefully. People write to me about applying to our program often enough that it seemed easier to answer here than one message at a time, so everything I would say in a first email is below.

Should I do a PhD at all?

This is a bigger question than whether you fit with me, and it is worth answering first.

My version of it is whether you have a question, a specific one, that keeps returning to you uninvited. A topic is not the same thing, and most people who say they want to do research mean that they like a topic. “I am interested in immigrant mental health” is a topic. “Do immigrant parents use school-based mental health services less than other parents, and if they do, is that about language, about cost, or about how referrals are made” is a question, because you can say what evidence would answer it and what evidence would not.

The difference is whether evidence could settle it, and the quality of your question will matter more than almost anything else you bring. Much of a doctorate is the work of turning something you care about into something answerable and then answering it. When the question stays vague, no amount of method repairs it, and an analysis can be executed perfectly and still settle nothing.

You do not need the finished version now, and mine changed more than once. Before you apply, find out whether you enjoy the narrowing itself: taking something you care about to the edge where the existing evidence stops, then stating something precise enough that evidence could contradict it.

Are you taking PhD students?

Yes, though only a small number, and I want to be straightforward about why. My bandwidth is limited at the moment, and taking on a student I cannot properly supervise would be worse for them than not taking them at all. So I am looking for a close match rather than filling a quota, and the next answer is the one to read before deciding whether to name me.

The way the assignment happens is worth knowing, since most applicants assume otherwise. You apply to the program, not to a person. The order runs like this.

1. You submit an application and name faculty whose work is close to yours.
2. A committee reads the applications and decides who is admitted.
3. Admitted students are matched with advisors, and my view enters at this step.

So naming me is not something I can act on, and my interest in your work cannot admit you. I say that plainly because it means a warm reply from any faculty member, or the absence of one, tells you very little about your chances.

Michigan has two routes: the PhD in Social Work and Social Welfare, and the Joint PhD in Social Work and Social Science with anthropology, psychology, or sociology. I can advise in either, and of the three social science options, sociology is closest to how I work.

Whose work fits with mine?

I think about my own work as three lines, and it is worth checking yours against all three.

• Topic. Language access, service use and delivery systems, and culturally and linguistically appropriate care.
• Population. Immigrants, refugees, and racial and ethnic minority communities.
• Method. Multilevel modeling, geospatial analysis, natural language processing, and machine learning.

Two of the three is usually enough. All three is rare, and one is generally not enough, because a project that touches only my topic or only my method leaves the two of us without much to talk about week to week. My research page has the current projects in detail, and the part of the agenda I am most invested in now is where technology and migration meet: how immigrant communities use digital tools, what happens to service access when it moves online, and what large language models change about both.

I am also more experienced supervising quantitative work than qualitative work. If your dissertation is going to be interview-based or ethnographic, a colleague whose own training is there would serve you better, and I would say so at the matching stage anyway.

Do I need to code already?

No. I did not when I started.

Solid quantitative methods training matters more, along with wanting to build the rest, and the building is something we can do during the program. When you are not sure where your quantitative preparation stands, my beginner’s guide to quantitative methods in social work research covers what I want someone to have met before starting. I have also written a guide to computational methods for social work research that describes how that actually goes, including which courses to take, which summer institutes are worth applying to, and what a first project looks like. If that post reads as something you would enjoy spending five years on, that is a better signal of fit than any coding you have done already.

What is your advising style like?

I should say first that I am new at this. I started at Michigan in 2025, so I am still working out what kind of advisor I am, and I would rather tell you that than describe a settled philosophy I do not yet have. The paragraphs below describe how things work at the moment.

The lab meets every other Friday, and members bring their own research to present and discuss, not only work of mine. My publications include a number of student co-authors, several of them as first authors.

Michigan also expects advisors and students to talk explicitly about how they will work together, and Rackham publishes a mentoring handbook for that conversation. The National Academies’ review of the mentorship evidence recommends the same thing, setting expectations out at the start instead of leaving them implicit. If we do work together I want to have that conversation early, instead of assuming we mean the same thing by supervision.

How can I make my application stronger?

• Research experience matters more than anything else you can still change. It does not have to be prestigious or paid. An undergraduate thesis, an RA line, a community evaluation, or a project you did in a master’s program all count, because a committee is looking for evidence that you have done research and want to do more of it.
• Name a question, not a topic. A statement that says you are interested in immigrant mental health tells us very little, since a great many people are. One that names a question, says why it is not answered yet, and says what you would need in order to answer it tells us how you think. The question does not have to be your eventual dissertation, and it will probably change.
• Place that question twice. The first placement is in the scholarly conversation, which for this work usually means immigration studies, social work, or the wider social sciences, and there you should say what answering your question would change about what the field currently takes for granted. The second placement is in practice, meaning who is helped by the answer and what they would do differently once they had it. A statement that does only the first is an academic exercise, and a statement that does only the second is a program evaluation. Doing both is, to my mind, a necessary condition for a strong social work statement.
• Ask for letters from people who have watched you work, rather than the most senior person who knows your name, since a specific letter from a lecturer beats a vague one from a dean.
• Have somebody read the whole application before you submit it, ideally somebody outside your field, who will notice where you have assumed knowledge the reader does not have.

Should I put your name on my application?

Yes, if the fit above holds, and name more than one person. Several colleagues at the School of Social Work work on immigration and migration from different angles, and one of them may be closer to your question than I am. Listing two or three faculty makes your application easier to place, and it does not read as indecision.

Is the program funded?

Both are fully funded. The School’s own description is that each “provides all incoming doctoral students with a multi-year funding package that includes tuition, stipends, and health care during the school year and summers.” Check the program pages for the current terms, since the details change from year to year and I would rather you read them at the source than take my summary of them.

For comparing offers across schools, PhD Stipends collects self-reported figures by university and department, which is more useful than the numbers programs advertise.

What else should I read?

Most of what I tell applicants is written down somewhere on this site.

• Resources for graduate school application, which is the practical version of the question above about strengthening an application. I also wrote a book chapter on applying to graduate school with BIPOC and international applicants in mind.
• CV tips for social work PhD students, which is about the document itself and applies as much to an application CV as to a later one.
• Computational methods for social work research and the beginner’s guide to quantitative methods, for the methods side.
• External funding and grant writing, which matter later but are worth knowing exist before you arrive.

Can we meet before I apply?

I do not reply to individual emails asking for a meeting, for feedback on application materials, or for an assessment of whether your work would fit with mine. This is a standing policy rather than a judgment about any particular message, and it holds however carefully the message is written. If your question is answered on this page, this page is my answer.

Volume is part of the reason. The larger part is that I have come to think of asking for help as a psychological resource, and one that is distributed very unevenly.

Writing to a professor you have never met takes a particular kind of confidence, and that confidence is not evenly available. It comes more easily to someone who grew up around academics and therefore experiences a professor as an ordinary person who can be approached, to someone who writes email in English without effort, and to someone whose circumstances leave enough slack to prepare months ahead.

Calarco found that middle-class children are coached from elementary school to ask adults in authority for help, while working-class children learn to manage difficulty on their own, and that the two habits persist. Stephens and colleagues found the same pattern at university, where institutions that reward independent self-advocacy disadvantage first-generation students without intending to. Jack followed low-income undergraduates at an elite college and found the divide running inside that group, since students who had attended well-resourced high schools approached professors readily while students from under-resourced ones often avoided office hours altogether. For students on the wrong side of that, the same email can feel presumptuous, or simply impossible, and there are capable applicants who never send one for reasons that have nothing to do with how good a researcher they would be.

If I answer the people who write and not the people who cannot, I am rewarding that resource rather than the qualities I actually care about. Writing in public is the one way I have found to serve both groups at once. So I put those hours into a page like this instead of into meetings and individual feedback, and I hope that seems like a fair trade.

None of this applies once you are here. If you become my mentee, or a student at Michigan, I will gladly make time for you.

Whatever you decide about applying here, I hope the question you are carrying finds a good place to be worked on.