Data Pedagogy - Tips for New Principal Investigators in Statistics and Data Science Education Research

Background

I have been invited to give a talk on statistics and data science education research as part of the Preparing to Teach Workshop. The workshop is designed for graduate students, postdoc scholars, and early career professors who are interested in teaching-focused positions in academia. In this post, I will share and expand on some of the points I made in my talk.

Finding Collaborators

If you are taking small steps in statistics education research and do not know many researchers at a personal level personally, do not worry. With time you can build these relations and find collaborators. Solely based on my personal experience, I think the statistics education community is very welcoming.

Reflecting on whom I have collaborated with and how our paths crossed, I built different connections in different situations. I have met some of my collaborators at conferences and on social media. One of my collaborators had written to me after they heard me speak on a podcast. I reached out to someone whose paper I had read and loved. I do not like the advice of “networking” much as it has a business-like ring to it. However, I would suggest being active in making human connections. This is not only because you would collaborate on research, possibly you may not, however, human connections can teach you a lot both in research and in life.

Developing Research Questions

If you are switching from methodological research to educational research, developing research questions might be difficult at first. To familiarize yourself with the field and the questions other researchers have you need to read, A LOT. Some journals in the field include

Subscribing to notifications of these journals as well as Google scholar alerts of researchers’ profiles would help you follow the research closely even if you only have time to read the abstracts.

Build your questions and research based on your own curiosity, not because of fear of “publish or perish” or that a topic would get you more Google Scholar citations. These kinds of metrics are not motivating enough to keep a research project going.

Many of my own research questions are derived from my own teaching which is why I am often curious about these questions. You may not have a lot of time to dedicate to research while teaching, however, you absolutely need to note down your reflections as you are teaching, daily or weekly. You can clearly identify questions you may want to investigate in the near future in these notes.

Thinking about Research Methodology

Many statistics students graduate with strong training in quantitative research methods. These skills are extremely important in educational research. However, many statistics students lack training in qualitative methods which are also extremely important in educational research. Some common methods include focus groups, think-out-loud protocols, structured and semi-structured interviews, and observations. If you can familiarize yourself with these methods whether through coursework or research, they will be useful. How to Design and Evaluate Research in Education book can serve as an introduction to research methodology.

Many educational research will require you to work with human subjects (e.g., students, instructors, etc.,). Thus it is crucial for you to familiarize yourself with the Institutional Review Board (IRB) process at your institution. You will need to complete a training and complete an IRB application before you can conduct human-subjects research. Plan ahead.

Running a “Lab”

Whether you consider it a “lab” or a “research group” or some group without a specific term, running this group would literally and figuratively will be a run. First and foremost, you should try to recruit students to your group as soon as possible. I personally enjoy working with mentees at all levels: undergraduate, graduate, and postdoctoral. Working with mentees at varying degrees helps me look at research from different perspectives.

There are varying mentorship styles varying from very structured to less structured mentoring. Before you recruit mentees and as you work with them you should continuously question your mentorship style and evaluate yourself. You will probably have a set of rules and priorities that will match your scientific and career goals as well as world view. It is best if you can communicate these with your students. For instance, for me, nothing is more important than someone’s physical and mental well-being, thus I try to communicate this very clearly to those around me.

Having a group of mentees will also mean that you will manage people from hiring to ending employment, sometimes managing timesheets, their required onboarding, etc. Do ask a lot of questions to staff to figure these out. It is also YOUR responsibility to make sure that those who work with you get paid on time. You will really need support from staff and thus it is good to make these connections early on. When I was on the job market it had not occurred to me to understand the staff support provided in the department but having worked at a large public university for four years, I understand and appreciate the support of the staff much more. Navigating the university system without their help would be impossible. In fact, when you go for job interviews, thinking about Professor X who does phenomenal research in area Y will probably have less impact on your research if your areas do not overlap, however, it is almost given that the staff will have an impact on your work, and often on a day-to-day basis. Whether this impact will be negative or positive will depend on your department.

Last but not least, you will manage a lot of projects while also managing teaching. You will possibly improve your project management skills along the way. One simple suggestion is to use some sort of app(s) or tool(s) that would help you keep track of your progress and your to-do list.

“Research” output can take many forms

What “research” is will often be defined by your department, tenure/promotion committees etc. From what I have seen in statistics education community is that there are many outputs that contribute to the field. These include open-access resources, R packages, and blogs.

Writing Tips

Publishing Peer-Reviewed Work

You don’t have to start alone. Find a collaborator.
Start with a pre-print to get feedback from the community.
Get to know your target journal well by reading the journal, reading author guidelines, and possibly serving as a reviewer for the journal. Some journals have reviewer forms on their page or you can reach out to the editor.
Consider conferences such as IASE Satellite or ICOTS etc. which publish peer-reviewed proceedings. Conference papers are shorter which would be less overwhelming for beginner researchers and proceedings are peer-reviewed thus you would still receive feedback.

Blogs

Blogs can be a good way to get into the habit of writing regularly. Without high stakes, they still provide a medium for getting the word out. This blog, for instance, has led to other opportunities for me. I wrote a post comparing teaching careers at liberal arts colleges and research universities which later was published on Amstat News. Similarly, I wrote a series of blog posts on how we held ourselves accountable to accessibility and inclusion principles while writing the Bayes Rules! book. These posts set the initial ideas for our manuscript Framework for Accessible and Inclusive Teaching Materials for Statistics and Data Science Courses. This is all to say, possibly blog posts can become more than blog posts.

Some blog examples from the field include:

Other Dissemination Venues

Conferences and manuscripts are traditional venues to disseminate your work. However, these days there are many non-traditional venues where you can share your research and connect with other researchers. These include:

Grant Writing

I have given many tips on grant writing that I now created a separate blog post on grant writing.

Acknowledgments

In the past few years, I have been involved in projects that I would have never imagined I would be when I finished graduate school. All of this has been partially possible thanks to collaborations I had with members of my research group. Thanks to

Giles Carlos - former undergraduate student
Feiyi Sun - former undergraduate student
Catalina Medina - graduate student
Federica Zoe Ricci - graduate student
Dr. Wendy Rummerfield - former graduate student
Dr. Sinem Demirci - former postdoctoral scholar

Collaborating with my mentees has definitely been one of the most fulfilling parts of my career.

Categories