News Archive | The Methodology Center

News Archive

Aaron Wagner and Sara Vasilenko recording podcastFebruary 26, 2015

In this podcast, we discuss the Prevention and Methodology Training (PAMT) program, which is a collaboration between The Methodology Center and The Bennett Pierce Prevention Research Center. Host Aaron Wagner talks with Melissa Boone, Michael Russell, and Sara Vasilenko, all current or former PAMT postdoctoral fellows. The opportunities and unique features of the program are discussed in under 20 minutes. PAMT also trains Penn State graduate students as predoctoral fellows, but that aspect of the program is not discussed in the podcast. For more information about PAMT predoctoral and postdoctoral fellowships, contact Bethany Bray.

February 3, 2015

Stephanie Lanza and Sara VasilenkoWe are pleased to announce that Stephanie Lanza and Sara Vasilenko will present this year’s Summer Institute on Innovative Methods, “An Introduction to Time-Varying Effect Modeling.” Sponsored by The Pennsylvania State University’s Methodology Center and the National Institute on Drug Abuse, the 20th Summer Institute will provide the theoretical background and applied skills necessary to identify and address interesting, new research questions using time-varying effect modeling (TVEM). Stephanie and Sara will introduce applications of these models to examine time-varying effects using intensive longitudinal data such as ecological momentary assessments, as well as age-varying effects using cross-sectional and panel data. By the end of the workshop, participants will have fit several time-varying effect models in SAS and will have had the opportunity to fit preliminary models to their own data.


The Summer Institute will be held on June 11 – 12, 2015 at the Bethesda North Marriott Hotel & Conference Center in Bethesda, Maryland.


Read more and register 

January 30, 2015

distal outcomes example diagramWe are pleased to announce the release of the latest version of the %LCA_Distal SAS macro. Latent class analysis allows researchers to divide subjects into underlying subgroups that cannot be directly observed. The %LCA_Distal SAS macro was created to allow researchers to estimate the impact of membership in a latent class on an outcome at a later time. The newly released version 3.0 of the macro allows users to generate standard errors for the binary case and provides more information in the onscreen results for all cases. PROC LCA version 1.3.2 or higher must be used in order to generate standard errors.  


Read more or download the macro

January 27, 2015

child using a tablet

Adaptive interventions help guide clinicians in their decisions concerning when and how treatments should be altered, but developing empirically based adaptations requires gathering the right kind of data. The sequential, multiple assignment, randomized trial is a recent innovation that can provide high-quality, experimental data for developing adaptive interventions. Recently, a group of autism researchers published the results of their SMART study in the article “Communication interventions for minimally verbal children with autism: A sequential, multiple assignment, randomized trial,” which appears in the Journal of the American Academy of Child and Adolescent Psychiatry, a top journal in child and adolescent mental health. The authors, led by Connie Kasari of UCLA, designed a project to improve spoken communication for children with autism who are minimally verbal. The study’s results show the benefit of integrating speech-generating devices (SGD) as a part of language development interventions and the potential of SMART designs for developing adaptive interventions.


Bethany Bray, Ph.D.

January 26, 2015

As part of our annual series of one-credit courses in research methodology for Penn State graduate students, in fall 2015 we will offer, “Advanced topics in latent class analysis (LCA).” LCA is an analytic method used to identify hidden subgroups within a population based on individuals’ responses to multiple observed variables. This short course, taught by Methodology Center Investigator Bethany Bray, will build on the knowledge and skills presented in the short course, “An introduction to latent class and latent profile analysis.” Credit for that course is not a prerequisite for taking this course, but familiarity with LCA and baseline category multinomial logistic regression are prerequisites.


January 21, 2015

Michael ClevelandWe are pleased to announce this semester’s Taste of Methodology workshop on multilevel modeling (MLM) for analyzing clustered or longitudinal data. The May 5 workshop will be presented by Michael Cleveland, research assistant professor of health and human development and Methodology Center faculty affiliate.


MLMs are critical tools for analyzing clustered data, repeated measures data, and other forms of hierarchical data. These data could come, for example, from students who are nested within classrooms or participants who were followed for three or more observations.


A Taste of Methodology is a workshop series that offers Penn State faculty an overview of innovative methods over lunch. This semester’s workshop will present the concepts and applications of MLM in order to give faculty an efficient way to assess its potential for their research. 


January 16, 2015

If you work at Penn State, you should know that we hold monthly special interest group meetings open to all researchers and graduate students working with innovative research methods. The groups provide a forum for individuals to discuss their own research and to learn from others. This semester’s topics include analyzing complex data, mixture modeling, and causal analysis.


For dates, please see The Methodology Center’s calendar.


Linda Collins, Ph.D.January 12, 2015

Are you interested in the new approaches that are emerging for optimization of behavioral interventions? Linda Collins invites you to join a new special interest group on optimization of behavioral interventions that is being formed within the Society of Behavioral Medicine (SBM).  


This group will explore approaches to optimization of behavioral interventions, including, but not limited to, the multiphase optimization strategy (MOST); factorial and fractional factorial experiments; the sequential, multiple assignment, randomized trial (SMART); and system-identification procedures drawn from control engineering.

December 18, 2014

Runze LiCongratulations to Runze Li who was recently named the Verne M. Willaman Professor of Statistics at Penn State! Runze is an influential and productive researcher, a great mentor, a supportive and helpful colleague, and a delightful person. We are proud and honored to have him as part of our Center. His recent work has focused on time-varying effect models (TVEMs) and the analysis of high-dimensional data. His work on TVEM is helping answer new questions about relationships over time. His work on high-dimensional data is laying the groundwork for incorporating genetic data into behavioral research.

Read more about TVEM

Read more about high-dimensional data analysis

December 3, 2014

Rebecca Evans-PolceIn the United States, rates of substance use peak during adolescence and young adulthood. Previous literature has demonstrated that rates differ by race, ethnicity, and gender. Despite knowledge of these disparities, until now researchers have been unable to understand the extent to which these disparities change across adolescence and young adulthood. In the forthcoming article, “Changes in gender and racial/ethnic disparities in rates of cigarette use, regular heavy episodic drinking, and marijuana use: Ages 14 to 32,” to appear in Addictive Behaviors, Methodology Center researchers Rebecca Evans-Polce, Sara Vasilenko, and Stephanie Lanza use the time-varying effect model (TVEM) to examine the dynamic nature of substance use rates among different groups of adolescents and young adults.


Like Us On Facebook or Tweet This Page