The Multiphase Optimization Strategy (MOST)
September 8-9, 2011
Linda M. Collins, Ph.D.
Sponsored by The Pennsylvania State University’s Methodology Center and the National Institute on Drug Abuse, the 16th Summer Institute provided researchers with varying levels of methodological training the opportunity to familiarize themselves with this new analytic technique. In addition, the Institute was designed to facilitate the exchange of ideas among substantive and methodological researchers.
Attendees of this two-day workshop learned about the Multiphase Optimization Strategy (MOST), which is a framework for optimizing and evaluating multicomponent behavioral interventions for prevention or treatment in any area of public health. Using MOST, it is possible to build optimized interventions and to increase the efficacy, effectiveness, and efficiency of existing interventions. The following topics were covered:
- The difference between optimization and evaluation, and why both are important
- Overview of MOST and some examples of applications of MOST
- The starting point: Your theoretical model
- What “optimized” means
- Choosing a meaningful optimization criterion
- The resource management perspective: Obtaining the most—and the most relevant—scientific information given available resources
- Efficiency of various kinds of experimental designs
- Factorial, incomplete factorial, and fractional factorial experiments
- Experimentation when there is a multilevel structure
- Building an optimized intervention based on the results of your experiment
On the afternoon of the second day, there was a computer lab on using software to choose a fractional factorial design.
Linda M. Collins, Ph.D., is Professor of Human Development and Family Studies, Professor of Statistics, and Director of the Methodology Center at the Pennsylvania State University. She holds a Ph.D. in Quantitative Psychology from the University of Southern California. Dr. Collins has over 25 years of experience teaching applied statistical methods at the graduate and undergraduate levels. Her research interests lie primarily in two areas: experimental design and analysis of longitudinal data. Dr. Collins’ research has been funded by grants from the National Institutes of Health continuously since 1984. She has published more than 100 articles and chapters in a variety of scholarly outlets, ranging from quantitative journals like Statistics in Medicine, Psychometrika, IEEE Transactions on Control Systems Technology, and Psychological Methods, to subject-matter journals such as Prevention Science, Annals of Behavioral Medicine, and Developmental Psychology. She has co-edited six books or special issues of journals, and is co-author of a recent book on latent class and latent transition analysis. Dr. Collins has given more than 100 invited talks at universities and conferences in the United States and abroad. Dr. Collins has received numerous academic honors, including the R. B. Cattell Award for Distinguished Multivariate Behavioral Research, the USC Psychology Department Teacher of the Year, and the Penn State Faculty Scholar Medal for the Social and Behavioral Sciences. She is a Fellow of the American Psychological Association and the Association for Psychological Science, and has served as President of the Society of Multivariate Experimental Psychology and the Society for Prevention Research.