I am an associate professor in and associate director for scientific outreach of the Center for Dissemination and Implementation Science (CDIS), an interdisciplinary research center focused on the processes that promote or inhibit the adoption of evidence-based and empirically-supported interventions in real-life settings in the US and globally. CDIS is housed in the Department of Medicine, in the College of Medicine, at the University of Illinois at Chicago. Formerly, I was an associate research professor in and associate director of The Methodology Center, an interdisciplinary research center in the College of Health and Human Development at Penn State. Take a look around to learn about my research and teaching! My work focuses on developing advanced statistical methods for research on the prevention and treatment of risky behaviors. I am particularly interested in the development of substance use and comorbid behaviors like gambling and sexual risk-taking.
Discrete latent variable methodology including latent class and latent transition analysis and their extensions, model interpretation in discrete latent variable models, and issues of model estimation and identifiability in discrete latent variable models.
Substance use and comorbid problem behaviors including alcohol, tobacco, marijuana and other drug use, risk and protective factors for substance use, comorbid problem behaviors like gambling and risky sexual behavior, comorbid mental health disorders like depression and anxiety, prevention and treatment of substance use and comorbid behaviors/conditions, and the model of addiction syndrome.
a brief history
I received my B.S. in mathematics and applied statistics from the University of Michigan – Dearborn in 2000. After attending graduate school at the University of Michigan for a year and a half, I transferred to Penn State to work with Linda Collins in The Methodology Center. During my time as a student, I received my M.S. in human development and family studies in 2005, my M.A.S. in applied statistics in 2006, and my Ph.D. in human development and family studies in 2007.
After graduating, I stayed in The Methodology Center as a research associate for two years, and then joined the faculty of the Psychology Department at Virginia Tech in 2009. After three years, I took a leave of absence in 2012 and returned to The Methodology Center to do research. After a highly productive and stimulating year, I decided to stay at The Methodology Center as research faculty in 2013 until 2019, when I moved to my current position.
statement of interests
My research focuses on developing and applying advanced statistical methods to questions about the longitudinal development of alcohol and other drug use, with a special emphasis on its relation to the development of comorbid problem behaviors like gambling and risky sexual behavior. My methodological research is motivated by challenges faced in substance abuse research, and it provides new ways to address questions that arise in the area of substance use and comorbid problem behavior.
Specifically, my research addresses the challenge of how to consider multiple dimensions of substance use, gambling and risky sexual behavior when examining developmental relations between substance use and gambling and between substance use and risky sexual behavior. Considering multiple dimensions of substance use including type of substance, quantity of use and frequency of use provides a richer picture of behavior and allows an examination of how dimensions may interact to increase the risk for negative outcomes. Similarly, considering multiple dimensions of gambling like type of activity, venue and wager size allows an examination of how dimensions may interact to increase risk for gambling disorder. Further, considering multiple dimensions of risky sexual behavior like dating, number of sexual partners and condom use allows an examination of how dimensions may interact to increase risk for sexually transmitted infections. By developing and applying methods that model the development of substance use and gambling or substance use and risky sexual behavior simultaneously, a detailed picture of the complex interactions among multiple behaviors may be examined.
My approach to dealing with these methodological challenges builds on the latent class model. The methods I develop and apply can be used to identify groups of individuals characterized by similar multidimensional patterns of behavior, describe change over time in behavior, and model longitudinal relations among multiple behaviors. These methods make it possible to identify groups of individuals who may be at high risk for advancing to stages of behavior involving problems in multiple domains, and thereby suggest how prevention programs may be targeted to individuals undergoing different etiological processes.