|
|
|
|
Linda M. Collins, Ph.D. Website |
|
Home  People Linda M. Collins, Ph.D. Website
 Director, The Methodology Center
Professor, Human Development and Family Studies
Professor, Statistics
Welcome to my web site! I am Director of The Methodology Center, which is devoted to the advancement and dissemination of statistical methodology related to research on the prevention and treatment of problem behaviors. I am also a Professor in Human Development & Family Studies and Statistics.
My research program has two primary foci: design for building, optimizing and evaluating behavioral interventions, and methods for longitudinal research. Most of my funding over the years has come from the National Institute on Drug Abuse.
More About Me: click here for a full CV
If you would like a copy of an article in press or in preparation, or the slides from a talk, please
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
BUILDING, OPTIMIZING AND EVALUATING BEHAVIORAL INTERVENTIONS
Intervention scientists typically are primarily interested in establishing whether an intervention is effective. I am interested in a different and, I would argue, equally important concern: building an intervention that is as potent as it can be, in other words, an optimized behavioral intervention.
Optimizing a behavioral intervention requires methodological strategies that are different from those required to establish intervention efficacy or effectiveness. For the past several years my collaborators and I have been identifying strategies that are practical for behavioral scientists to use to optimize their interventions. To do this, we have been borrowing ideas from engineering. I believe that optimizing behavioral interventions will ultimately produce more effective and more cost-effective interventions. I also believe that optimization can be accomplished without requiring more research resources in the long term. However, a realignment of research resources will be required.
For example, our work suggests that a phased experimental approach can be used to optimize a behavioral intervention before it is evaluated in a standard randomized controlled trial (RCT). This has been called the Multiphase Optimization Strategy (MOST). To learn more about MOST, click here. In my methodological work on MOST, I have been collaborating recently with Inbal Shani, John Dziak, Susan Murphy, and Runze Li.
As part of this research program on optimization I have also been collaborating with Daniel Rivera of the Control Systems Engineering Laboratory at Arizona State University. Daniel and I have been leading a project that is laying the groundwork for eventually using principles derived from engineering control theory to optimize behavioral interventions. Currently we are working on expressing some selected behavioral interventions as dynamical systems. This work is funded by the NIH Roadmap Initiative.
I am eager to collaborate with intervention scientists who are interested in using innovative methods to optimize their behavioral interventions. I am very excited about a new collaboration that will be starting in late 2009. This collaboration will enable me to work with some outstanding intervention scientists, including Michael Fiore, Timothy Baker, and Megan Piper at the University of Wisconsin Center for Tobacco Research and Intervention (CTRI), and Robin Mermelstein at the Institute for Health Research and Policy at the University of Illinois. Together we will be using a phased experimental approach to optimize a clinic-based smoking cessation intervention.
If you are an intervention scientist who is potentially interested in collaborating on research to optimize a behavioral intervention, I encourage you to contact me to discuss this possibility.
Back to Top
METHODS FOR LONGITUDINAL RESEARCH
With Stephanie Lanza, I have worked extensively in the area of latent class models for longitudinal data, particularly Latent Transition Analysis (LTA). LTA can be used to fit discrete models of change, such as stage-sequential models, in longitudinal data. Stephanie and I have written a book on latent class and latent transition analysis that will be available in late 2009 or early 2010 (Collins, L.M. & Lanza, S.T., Latent Class and Latent Transition Analysis for Applications in the Social, Behavioral, and Health Sciences, Wiley). This book is intended to be a comprehensive introduction to latent class and latent transition models in which the indicators are categorical. It includes chapters on multiple-groups LCA and LTA and covers how to introduce covariates to predict latent class membership. We included a lot of empirical examples so that the material in the book would be relevant to real data analysis problems.
To view material on LTA in the Getting Started section of our web site, including instructions on how to download software for LTA, click here.
I am also interested in all aspects of methods for longitudinal data. I have co-edited two books on methods for longitudinal research (Collins & Horn, 1991; Collins & Sayer, 2001). More recently I wrote a chapter for the Annual Review of Psychology on methods for longitudinal research (Collins, 2006).
Back to Top |
|
|
|
