Timothy Johnson, Horace T. Morse Distinguished Professor of Political Science and Law, University of Minnesota

As the oft-overused adage goes, “You learn best by doing.” That is how my career as an academic began as I wrote with my undergraduate advisor, with my graduate advisor, and then with mentors who took me under their wings. So it is not surprising that writing with graduate students has always been an intuitive part of the training and development I provide. Indeed, it has been a highly productive way for me to professionalize students into the research process.

Writing with my students is also mildly selfish. That is, I do so because I enjoy the process. In a profession and discipline that still value single-authored publications, working alone is anathema to me, especially when there are amazing scholars with whom to work. And graduate students offer something that seasoned colleagues do not always offer–a fresh, unique, and usually not-yet-jaded view of how to solve a research problem and how the solution fits, theoretically, into existing literature.

The question, however, is how do projects emerge, and how does the process work? The answer is that it varies but is remarkably similar to how I write alone–with the exception that I ensure my student coauthors do most of the work. This point is not meant to suggest my students just act as research assistants on projects. In fact, it is just the opposite. The work they put in is explicitly meant to train them in how to create and complete theoretically driven, systematic, research that they have the first cut at producing. As such, I usually give them a general outline of what we need and then allow them to quickly take charge and to learn how to complete the project from start to finish. But the focus of my mentorship varies depending on the nature of a given student.

In an effort to provide advice for others working with graduate (or undergraduate) students at various stages of academic training, I consider several student types you may encounter. This is not an all-inclusive list and I am the first to admit that I do not know everything. But it comes from my own experience of success and is meant as a conversation starter. Specifically, I offer advice (in no particular order) on three types of students: those new to the literature, those looking to use their methodological prowess to the greatest extent possible, and those who are (knowingly or unknowingly) perfectionists.

The Student New to the Literature

This first student type emanates (for me) from a student I began working with more than 10 years ago. She was, for sure, one of the top three undergraduate students I have ever taught and I was so impressed that I asked if she was interested in working as a research assistant for me. That didn’t last long because she was so good that she, almost immediately, became a coauthor on several projects. The problem was that she had no experience with judicial politics and, more problematic, little social science research experience. So, we began from scratch as I handed her my graduate level Supreme Court seminar syllabus (including scores of recommended readings) and told her to immerse herself in it. She did, in spades! We also had long discussions (which is typical when I write with students) about the literature and how reading social science differs from reading textbooks, court cases, or history. So she read, we talked (a lot), and she became one of the most well-read young scholars with whom I have had the pleasure to work.

During my tenure, I have also had several graduate students switch from a different subfield to judicial politics (they surely did not switch to work with me but rather because they wanted to change their focus of study). These students presented a different issue than the student who was brand new to judicial politics and to social science research. They already had the skills to read critically in a way that social scientists read. They simply needed to learn the content and background of the Court, its decision-making process, and the judiciary more generally. So, I had them read in the same way as the first student I described. However, they were more focused on what they read because they had paper or dissertation topics already in mind. They could, initially, focus their reading lenses rather than read as widely. The latter would come later as they expanded their own research agenda.

The lesson from both these types of students? When someone is new to the field (as either a graduate or undergraduate student) give them the grounding they need by having them read, read, read, and read some more. But do not stop there. Engage with them. Have long formal and informal discussion about specific articles/books, broader trends in the literature, and how the readings may help them with their own research.

The Student Immersed in Methods Training

Every researcher knows the story of finding a new methodological hammer that leads them to look for any theoretical or substantive nail on which to use it. Mine was rare events logit modeling because so much of what happens on the Supreme Court falls into this category. But for students who have econometric methods as their first or second field of study, it is easy for them to want to employ a particular modeling technique as much as they can. Others simply want to focus more on the data and analysis section of projects because, let’s be honest, they are more fun! You get to play with data, get your hands dirty, and see results pop up quickly from your R code or Stata output.

The trick is to make clear to students that research is not driven by data or by modeling choices. Rather, and importantly, it is driven by theory. If theory suggests a phenomenon of interest should be measured in a particular way, then then that is how it should be measured. Further, theory should dictate modeling choices. In other words, if the phenomenon of interest is dichotomous, discrete, or continuous, that indicates what model to run. The point, however, is similar to what I argue about students new to the subfield–reading and thinking about how to solve a research problem is the lynchpin to success. That is done through good communication and extensive discussions with one another.

The Perfectionist

Perhaps the most difficult type of students with whom to work are the perfectionists. In many respects this is a good problem to have because perfectionists want to cite all the appropriate literature, want data that is 100 percent clean, and want to run every iteration of models as a paper takes shape “just to make sure.” The problem is that, even as we all strive for perfection, no paper, no model, no theoretical setup will ever reach that level. All we can do as researchers is to get as close as possible.

So how do you deal with a perfectionist? It is about convincing them that it is okay to let go and that it is okay to allow others (reviewers included) to see what you have written. You must possess a lot of patience and, in some sense, you must by a psychologist. That is, you must be able to convince your coauthor that it is okay a paper is not perfect, that you may have forgotten to cite some literature, and that a model may not ultimately be fully specified. To accomplish this goal, I stick with my theme of communication. Talk, explain why something is ready to go, and ensure your coauthors that all papers must ultimately be sent out for review, or they will die on your hard drive or in the cloud. That is the worst outcome. So get the paper out there, get comments, get reviews, and get it as close as possible to perfection.

A Parting (Very General) Thought

I cannot tell you how to treat your graduate students. But I can tell you how I treat mine and that my model has been very successful. Graduate students are not, in my estimation, students. Rather, they are already political scientists and judicial scholars. They just happen to be at an earlier stage in their career. So I treat them with the respect they deserve as colleagues (even when they T.A. or R.A. for me) because they truly ARE colleagues. Crafting relationships in this manner works and I am certain my treating them in this manner has made every one of them successful in their own way–and oftentimes repeat coauthors, which is always good for me!

We all encounter many types of graduate students, and there are certainly more types than I delineate in this short post. However, the theme that runs throughout each of them comes down to good, open lines of communication and mutual respect. If you can accomplish this, then you can deal with any type of student whom you may encounter.