Are you interested in joining the CARES lab and working with me (Luke Miratrix)? This document is intended to provide prospective applicants with general – and equitable – information about what I personally look for in a PhD applicant. In particular, below I provide responses to several questions that I often get from prospective doctoral students.[1] This information may also be useful for thinking about applying to (and getting in to) PhD programs more generally. Please note I’m speaking only as an individual faculty member, not on behalf of any program or the University.

I offer these thoughts on the “hidden curriculum” of this program in the spirit of leveling the playing field for all potential students. I do not take individual meetings with prospective students; this way, everyone gets the same information. If you have a question that I don’t answer, send me an email (lmiratrix@g.harvard.edu) that mentions this page and I’ll add the answer here.

First, you should know that individual professors are not in control of the admission process (at Harvard/HGSE). Both the education school and the statistics department have admissions committees that review applications as a pool. The way it works is there is an initial committee that reviews applications, and then the finalists are sent out to individual professors to gauge interest. When that step happens, I would read your application as a faculty member and express interest. 

Q: What background do I need? (for Education programs)

• Background Most of my education students have at least a couple years of post-undergrad work experience. These experiences help students identify problems that need solving and get a sense of what is important in the world of education. Some successful students do come straight from undergrad or via a master’s program, but having had some meaningful exposure to the research process (e.g., through a part-time position, undergraduate/master’s thesis, monitoring and evaluation experience) prior to enrolling is really important.
• Quantitative skills You’ll need to show that you can handle a good amount of quantitative rigor (this will help with the program, and with working with me in particular). If you haven’t taken (and gotten good grades in) advanced math classes like calculus or statistics, and are planning on doing quantitative research, you’ll need to explain this history to the admissions committee, and identify how you will make up for this in your written statement or in the explanations of grades section of your application. 
• Experience with data Hands-on experience analyzing real data to answer real-world questions is common for our PhD admits. Working on “toy data” for problem sets in classes usually isn’t enough. (By “toy data”, I mean easy-to-use data created for students or to demonstrate a method. These data don’t have complications like missing values or hierarchical structure). Ideally, you would have some experience using data to answer a real, relevant, and challenging research question.

Q: What background do I need? (Statistics)

• Background Most of my stat students have at least some research experience as an undergrad with some faculty. These experiences help students identify problems that need solving and also help the admissions department know that you know what you are getting yourself in to.
• Quantitative skills You’ll need to show that you can handle the quantitative rigor of the program. Ideally this means you have taken advanced stat courses, and are getting good grades in some advanced math. I don’t need or look for pure math powerhouses, however.  But to compete in the overall pool, you would need strong research experience or experience with data to offset less math experience.
• Experience with data Hands-on experience analyzing real data to answer real-world questions is really good to have, especially if you are interested in working with me. Alternatively, having worked on a complex simulation or some sort of methods development can help make for a strong application.

Q: How does applying work?

This is too big a question to answer here, but in brief: You will need to prepare an application and solicit some letters of recommendation.  This all happens in the fall.  Applications are read in January and February.  You start hearing back in the spring.  For a lot more about this process, see below.

Q: Where can I get more information about your work and/or the doctoral admissions process?

This website (cares.gse.harvard.edu/) is a good place to learn about my research, my students, and the culture we try to create at the Miratrix CARES lab.

For general information about this process, see the Health Policy Data Science Lab’s page for prospective students at https://healthpolicydatascience.org/prospective-student-info (although much of that material I have stolen for this page and a page overview of being a student as well).

For information on HGSE’s admissions process, I recommend the resources provided by the Admissions Office: https://www.gse.harvard.edu/admissions/apply

For broader information on the “hidden curriculum” of graduate school, I would recommend the book A Field Guide to Grad School by Jessica Calarco.

Q: Can you give me an overview of the stat or education PhD Programs at Harvard, assuming I was working with you?

Yes!  Please see here.

Q: What do your students tend to do after graduating?

The best way to tell is to scan our page of alumni.  The short version is: lots of different things!  A good number end up with various professorial posts ranging from more teaching to more research.  Many work in industry or join policy evaluation firms such as Abt. Most continue using their skills gained in their PhD program to do things that interest them. As far as I know, most are quite fulfilled by their post-graduate work.

Q: Are you available to discuss my interests/application prior to my acceptance?

For both practical and principled reasons, I don’t meet (in person or on the phone) with prospective students prior to admission, nor do I read draft personal statements or other materials outside the normal admissions process.  On the practical front, I tend to get numerous inquiries a year, and I just can’t make time to respond substantively to prospective applicants while also giving my current students the time and attention they deserve.  On the principled front, I believe strongly in trying to level the playing field for prospective applicants and am concerned that individualized meetings may lead to further inequities. 

With this being said, if you are admitted to the PhD program, you can rest assured that I will shower you with attention and answer every possible question you may have.  It is in our collective interest to ensure that the relationship is a good fit, and that you can achieve your goals if you come to Harvard!

Q: Does this mean that I shouldn’t contact other HGSE/non-HGSE faculty, either?

No, it definitely does not!  Norms and protocols regarding admissions vary widely both within and across institutions.  At some schools, individual faculty play an important role in selecting students for admission and, as a result, contacting faculty ahead of submitting your application can be an important means of demonstrating interest and ensuring your full application will be read by someone who could advocate for your selection.  It is therefore probably a good idea to reach out by email to other faculty members with whom you are considering working ahead of submitting your application.  These emails do not have to be complicated.  They can include several sentences that clearly, concisely, and politely communicate your background, key research interests, and the alignment of these interests with the prospective faculty member’s research agenda.  It can also be helpful to include your CV and any additional follow-up questions you might have for that person (e.g., whether they are accepting new students, are available to talk further, etc.).  If you don’t receive a response immediately, you can follow up with a polite reminder.   With all this being said, you should not take it personally or as a negative sign if you do not hear back from a prospective faculty mentor.

Q: Are you accepting students to work with you this coming year?

Yes, I am always enthusiastic about taking on new doctoral students!  It is important to understand, however, that unlike in some institutions, individual HGSE faculty do not have the authority to admit individual students.  Rather, we accept candidates as an entire faculty (represented by the admissions committee).  You should therefore think of the audience for your application as being the whole faculty, not just me (and other prospective advisors).  You do need to excite at least one prospective advisor, as we accept students only if there is one (or more) faculty member who is enthusiastic about taking them on.  But at the same time, you also need to excite the faculty as a whole.  Advisors retire, get ill, move institutions, get pulled into new roles, etc.  When we admit you, we commit ourselves to you as a school.

I personally think this last point deserves extra attention: the more a school engenders a community where a student can potentially find mentorship from multiple faculty, the greater the chance that the relationships can be genuine and good fits. If there is only a single advisor that works, and there is a relationship mismatch, then a student can get trapped in a bad (and imbalanced) relationship dynamic. I believe it is better for students to have freedom to change advisors.

In both Harvard Stats and HGSE, I have found that students do have this mobility which means students can obtain the support and training they need to do great work. But this means you should be looking at the school and the faculty as a whole, not just me, when evaluating fit.

Q: Do I have to do statistical methodology work to have you as my advisor?

If you are an education student, you definitely do not need to do methodology work with me. That being said, what I hope will happen is in the course of tackling interesting and engaging applied problems, you and I will find some area where the want to tweak methodology, or borrow methodology from somewhere unusual to answer the applied question you are working on. I particularly love this kind of paper, where a real methodological strength can open doors to inquiry!

Q: Do I have to work only with you as my advisor?  Can I have multiple advisors?

I have a set of specialized skills that put me between education and statistics. This means that in general, I believe my students are better served if they are working with me and other faculty as well. This gives them a broader perspective on their chosen field, and make sure that whatever gaps I have in my own mentoring are filled by mentoring from elsewhere. It's also much more exciting and fun to work with multiple people, I believe. So I strongly encourage my students to build real relationships with other faculty throughout the University as suits their research interests.

Q: What do you look for in a PhD application?

I am happy to provide some details of things that I look for when I read a PhD application, with the caveat that these are my own personal opinions.  Other readers of your application (in particular, the members of the admissions committee, who vary across any given year) may look for other implicit or explicit criteria.  As such, you should take these ideas with a grain of salt.

In general, I look across the materials of an application and try to assess whether we would have a good and productive research relationship.  For me, some signs of a good fit are

• Curiosity: I love those who are trying to understand the world around them. I also highly value those who want to understand how the methods of knowing (research methodology) itself work.  Even if you are planning on doing applied work (trying to answer questions about how society or education functions, I mean), I look for evidence that the means of learning these things is itself of interest to you.
• Excitement: Some people are mission driven (they have found a problem they want to work on and solve), others love the work (the excitement of discovery), and others see it as a path to service (the joy of building tools that can help people achieve their goals). One way or the other, I get excited by people who are taking this on because they are excited by something.
• Capability: I look for evidence that you have the potential to do and excel at quantitative research ranging from thoughtful application of quantitative methods to problems of practice to designing new quantitative methods to help others take on problems of practice. Even if you are a math powerhouse, I want to see evidence of being able to think about the application of methods to practice, and how that can succeed or fail (this is a very different kind of thinking than mathematical theory, and one can be able to do one and not the other in either direction).
• Mathematical comfort and curiosity: I tend to be more successful with students who are wanting to continue to develop their mathematical ability in addition to whatever core research program they are on. This interest can be demonstrated through prior coursework, applied quantitative work (e.g., as a research assistant), and/or explicit statements of interest in your personal essay.
• Independence: Given my role as a statistical methodologist in an education school, I am simultaneously very well positioned and not so well positioned to advise both stat and ed students. The best mentoring I do is with students who also have other mentors, and also have some independence.  For example, in education I am not good at connecting students to specific education research problems, but I am good at helping them work on a problem they have. So independence, here, would be identifying the problem.
• Explanation of how your prior experiences have led you to the point that you are ready for a PhD program. Your personal statement should not rehash your CV in narrative form (e.g., “first I did X, then I did Y, now I want to do Z”), but rather paint a more detailed picture of how your professional and academic experiences have prepared you for this moment in your career.  What did you learn in each of your positions, and how did these experiences lead you to want to address your stated research questions within a doctoral program?
• Good writing. Is your statement organized, clear, and engaging? Does it make efficient use of the short amount of space we give you?  Does it demonstrate that you will be able to write good papers?  Does your writing demonstrate that you can think, and that you have something interesting to say that you can convey cogently to others? 
• Demonstrated research experience (in general).  A PhD is, first and foremost, a research degree.  Again, the important thing here is not that you are already a fully trained researcher (this is what getting a PhD is for!), but rather that you know what you are getting into, at least a little. For ed students, having several years of full-time research assistant experience prior to enrolling in a doctoral program can give you the perspective to really take advantage of a PhD program.  For stats, this is less important as it is easier to get a sense of “I want to do more of this,” given an undergraduate experience, but I still want to know that you know what this might look like for you.

Overall, I want my education students to be leaning a bit towards being a statistics student, and for my statistics students to be leaning a bit towards being a student in some social science.

I should note that most other professors also look for a few other things (even though I do not tend to).  For a comparison, please read Dana McCoy’s version of this document (this document, in fact, is an edited version of hers): https://seed.gse.harvard.edu/info-prospective-students.  Some things to consider:

• A clear account of a topic, question, or phenomenon that you want to research.
• Compelling motivation for why you believe this problem matters.
• Evidence that HGSE/I could support you in addressing your identified topic of interest and future goals. 

In your personal statement, you do not need to:

• include statements about how wonderful HGSE is, or about the eminence of a particular faculty member, or what a privilege it would be to study here.  (In fact, please save yourself space and delete all such commentary.)  
• cite a bunch of literature, although you are welcome to if it adds to your argument.
• share personal anecdotes or vignettes that are not directly relevant to the work you want to do.  

Q: How and when will you read my application?

The PhD in Education Doctoral Admissions Committee will forward to me applications that are relevant to my areas of expertise.  You therefore do not need to send me your application separately or do anything special to draw my attention to your work. You should write my name down on the application as a person you are interested in working with, however.  This will ensure that I will review your application in the normal course of the admissions process.

Q: Can you provide me with feedback if my application is unsuccessful?

Unfortunately, providing this sort of feedback is typically not possible.  It is important to understand that the admissions committee and each individual faculty member at HGSE are always enthusiastic about more candidates than we are able to admit. Every year, for both stat and ed students, I end up identifying several students I am extremely excited about and that I think would be wonderful to admit. And usually only a few, if any, of these recommendations are let in. Often final decisions have little to do with personal aptitude, and more to do with whether the prospective candidate is a good match with the overall needs of the school in any given year.