Feature

Psychology graduate students, postdocs and faculty hunkered down to master the latest techniques for conducting longitudinal research at an APA-sponsored Advanced Training Institute (ATI) in longitudinal methods held at the University of Virginia, June 4-9.

The course, part of the association's Academic Enhancement Initiative, is the second that the association has sponsored. Last year, at the suggestion of the Board of Scientific Affairs (BSA), APA organized its first ATI in functional magnetic resonance imaging--a course that was held for a second time this summer.

The ATI on longitudinal methods offered lectures and hands-on training in the computer lab for the course's 30 participants. Four instructors--University of Virginia psychologists John McArdle, PhD, John Nesselroade, PhD, Karen Schmidt, PhD, and Ohio State University psychologist Michael Browne, PhD--discussed methods for planning, conducting and analyzing change over time.

The course included instruction on how to account for missing data, develop complex "growth models" using data collected in a time series, compare change in a study's overall sample with individual differences in patterns of change and simultaneously assess change in multiple outcome variables. Throughout the course, the instructors emphasized that the success of even the most sophisticated longitudinal techniques depends on some basic principles: using an appropriate sample, legitimate experimental manipulations and sound dependent measures.

Methodological and statistical techniques for modeling change over time have advanced so rapidly in recent years that it's been hard for psychologists to keep up, says course participant William F. Chaplin, PhD, a personality psychologist at the University of Alabama.

"Many of the techniques available to us now would have been too computationally intensive even 10 years ago," he notes. As a result, he says, "We had to ask much more simplistic questions and gather crude data, because we didn't have the ability to analyze the data." The ATI in longitudinal methods, Chaplin says, "alerts us that it's now possible to think in a more sophisticated way--particularly those of us who grew up in a less advanced age and who were almost forced to think simplistically about change and about the theories that we developed."

But even for new psychologists, Chaplin says, the course provided a valuable addition to the classic fare in graduate methodological and statistical training.

"In general," he argues, "the courses that we teach our graduate students have not fully caught up with the state of the art." He plans to incorporate much of what he learned at the ATI in longitudinal methods into his graduate-level statistics courses, Chaplin says.

The ATI did more than provide the theoretical underpinnings for understanding longitudinal techniques, adds Lauren M. Papp, a graduate student at the University of Notre Dame. Armed with the examples and instructions provided in the course, Papp says, students left the course prepared to immediately begin applying the most current longitudinal methods to their own data sets.

"We took so many resources home with us--methods and strategies that I don't think will really hit the literature full-force for a long time," she says.

"Within APA, the Advanced Training Institute is the key vehicle by which technological and methodological innovations are integrated into research training," comments BSA member Mahzarin R. Banaji, PhD, of Yale University. "New generations of psychologists will benefit immeasurably."