Alumni spotlight on Derek Kreager: Getting in on the Ground Floor of Statistics for Network Analysis
Dr. Derek Kreager is a Professor and Associate Head of the Department of Sociology and Criminology at Pennsylvania State University. He earned his PhD in Sociology at the University of Washington (UW) and was among the first students to ever complete a CSSS track. His research focuses on social networks, health, and criminal or delinquent behavior.
Kreager spoke with CSSS about his path into research and what he’s finding exciting now. This interview has been edited for length and clarity.
Question: What led you into your current work?
Answer: Growing up, I was interested in joining the Army, mostly because my father was in the military and encouraged me to pursue a similar career. When I got into West Point, I wasn’t really focused on academics as I didn’t think I’d need an advanced degree for a military career. I chose psychology because I thought it would be easy! To graduate from West Point, you have to be an engineer. So, I picked systems engineering because I thought it sounded easier than thermodynamics or electrical engineering. It was a relatively new field at the time, and unexpectedly, it was also a precursor to my future career. What I do now is essentially systems engineering — turning systems into networks and using those networks to analyze behavior, health, and crime.
After graduating from West Point, I served six years in the military. At that point, I wasn’t interested in staying longer, and I thought I wanted to be an academic. Being curious as a profession and really immersing myself in learning felt like a fit with my personality.
So I went to Seattle — where my family is from — and worked as a bartender for a year before applying for grad school. I didn't know exactly what I wanted to do, I just knew I wanted to go into higher education. I took a couple of night classes at UW while I was bartending, and I gravitated towards criminology. I applied to two grad programs, and got accepted into one — that was at UW!
I started in the sociology program in 2000, and Ross Matsueda ended up being my advisor. One of my early instructors was Kate Stovel who was doing work with adolescent health and social networks. I was interested in adolescent health and delinquency, and ended up doing my master's thesis on student isolation and delinquency using the just-released Add Health network data.
It was also fortuitous that Mark Handcock and a lot of the UW network folks were building statnet for R when I was there, around 2000 to 2006. It meant I got to learn a lot of the statistics of network analysis very early, with the people who were developing those methods. That exposure helped me transition into my dissertation research, which looked more broadly at peer influence and adolescent delinquency.
After the PhD, I came to Penn State, and I’ve been here since. Using network theories and methods, I study how friendships and romantic relationships are associated with risk-taking and crime among adolescent students and adult prisoners. I also look at how social relationships change over time and their impacts on criminal trajectories and desistance. After tenure, my research direction shifted towards incarceration, looking at the correctional system and re-entry — basically applying a lot of network ideas to a new population.
Q: How did you initially get involved with CSSS?
A: My advisor, Ross, was involved very early as a sociology representative when CSSS was being established. He encouraged me to pursue advanced statistical training and taught some of the courses as part of the track. I took all of the classes, which I thought were tremendously difficult and rewarding. I also started going to all of the CSSS seminars.
At the time, I didn't understand how unique that experience was. I had no idea that this is not the way graduate students typically take classes. I thought any sociologist would take classes from statisticians on causality or networks. And, I had no idea that CSSS was a brand new thing. It wasn't until I was given the certificate at the end that they said, ‘congratulations, you were one of the first people to complete this!’
Q: What did you take away from your experiences with CSSS?
A: Being connected to CSSS meant I was on the ground floor of some statistical packages like exponential random graph models (ERGM), which I’ve ended up using throughout my entire career. In part because of that early exposure, those statistical methods and packages became very central to my research identity.
My CSSS training meant I started out doing work that was very much quantitative analytical — working with sophisticated methods, learning network dynamics, network structure — and I continued doing a lot of that research for the early part of my career. Over time, I’ve actually done less quantitative analytical work, and started doing more theory development and data collection. I still work closely with people who are network statisticians — I mostly run projects that ask mixed-methods kinds of questions, aiming to get at some of the underlying meanings associated with relationships and structure.
Q: What’s a project you would highlight?
A: One exciting project is focused on evaluating a peer-led re-entry housing program, where previously incarcerated men help peers negotiate the transition from prison to the community.
We're looking at the program through a network lens, and using mixed methods (combining qualitative and quantitative) including interviews to try and understand their social networks over time. We can tell detailed narratives about people's relationships and situate them within social structures, while also using longitudinal statistical models.
I especially enjoy working with the project’s interdisciplinary team, which includes researchers with backgrounds in rehabilitation, public policy, criminology, and sociology. We have an ethnographer, qualitative folks, network folks — it's a big group of people who are all working together to try to understand the context. I get to be the manager and ask, ‘how can we work together to do this project?’ It’s also exciting because the re-entry program is peer-led, so the peer mentors’ and staff stories are part of the research itself.
Q: If you could go back and give yourself advice during your PhD, what would you say?
A: I remember suffering from feelings of imposter syndrome, and I know that’s common. Wondering “are you good enough?” or “can you do this?” So, my advice would be to just keep moving through it — take advantage of all the opportunities provided, continue learning. Because over time, those kinds of insecurities can sort of work themselves out — as long as you keep moving forward.
Q: What were some of your favorite spots around the UW campus or the university district in Seattle?
A: I really loved going to Agua Verde on a nice day. It was such a good view. You could watch kayaks go by, drink margaritas and eat Mexican food — for the perfect mix of environment and food that felt relaxing and fun at the same time.
Q: What have you been reading, watching, or listening to lately that you’d recommend?
A: Alien: Earth, because I love sci-fi and it's just such a great show — good characters, and asks some very interesting questions about AI, robots and humanity. So it's intellectual and also super fun.
Learn more about Kreager’s research on Google Scholar and his Penn State profile.
Derek Kreager and UW cohort-mates in 2001 (Left to right: Heather, Lori, Laura, Derek and Cara)
