Feature

Cutting-edge methods for conducting state-of-the-art research are in high demand. But courses for learning the latest methods and instructors for teaching them are in short supply, according to a new survey presented at APA's 2001 Annual Convention.

Indeed, few psychology PhD programs offer direct training in the latest data-analysis techniques, the survey found. And while demand for methods courses has increased over the past decade, the number of faculty teaching methods has remained flat, said Arizona State University psychologist Leona S. Aiken, PhD, who presented the survey findings and chaired the symposium "Methodological training for PhD psychologists."

The survey was a follow-up to a similar one Aiken and her colleagues conducted in 1987 (American Psychologist, 1990) in which they concluded that training was adequate for traditional techniques but not newer and often more useful ones. Although Aiken and her colleagues found some improvements in training over the last decade, their conclusion remains the same.

"No area of psychology is untouched by the potential for use of new methods," said Aiken. "Yet, there are only small inroads into the curriculum of these new methodologies."

In 1998­-1999, Aiken, along with her Arizona State colleagues Stephen G. West, PhD, and Roger E. Milsap, PhD, surveyed all 234 psychology departments in North America that offered a PhD; 201, or 86 percent, responded — a similar response rate to their 1987 survey.

Also similar to 1987 is the content of course offerings: Few programs require students to take a course on measurement and most programs require only a year of statistics, which is taken up in large part by a two-semester introductory PhD statistics course. At most schools surveyed, that introductory course primarily covers classic material, with a focus on "analysis of variance." The only "noteworthy" change, said Aiken, is an increase in coverage in this introductory course of multiple regression analysis. In addition, more departments are offering at least some instruction in structural equation modeling — a method that has been used regularly in many areas of psychology for at least 10 years.

Other newer methods, such as multilevel modeling, methods for conducting longitudinal research and techniques for dealing with missing data, are only taught by a minority of programs, said Aiken. "The important quantitative developments of the 1990s are weakly represented in the curriculum," she concluded.

Meanwhile, the student demand for quantitative training is on the rise, the survey found. While students are only required to take two semesters of methods — the average students took in 1987 — they're now taking three semesters.

Unfortunately, faculty available to teach methods isn't keeping pace, said Aiken. No new slots for quantitative faculty have opened up since 1987, and when programs hire faculty to teach research methods, they overwhelmingly prefer to hire someone whose focus is a substantive area of psychology rather than hiring someone who specializes in methods.

This "two-fer" model of hiring for quantitative positions poses serious problems for the discipline, said survey co-author and symposium discussant West, current editor of Psychological Methods. People hired for these posts are either overworked or, because they're rewarded more for work in their substantive area, they fail to keep up with the latest quantitative methods and eventually stop teaching methods altogether.

Adding to the shortage of quantitative methods faculty is the problem that many current instructors are older and there are few younger people in the pipeline to replace them, said West. Even when students strong in quantitative methods crop up, they are often snatched up by large research centers rather than universities.

Symposium participants Philip Kendall, PhD, Ross Parke, PhD, Harry Reis, PhD, and Mark Appelbaum, PhD — all journal editors — discussed solutions to this problem, emphasizing special training needs for their subdisciplines.

One solution might be to find new instructional models that use quantitative faculty more efficiently and effectively, said discussant Applebaum, professor of psychology at the University of Cali-fornia, San Diego, and founding editor of Psychological Methods. And that means looking beyond a traditional "course" model at techniques such as packaging training into intensive units that can be offered over weekends or via videotape, or taking advantage of faculty at other schools through distance learning.

The bottom line, said Appelbaum, is that students are "pushing us" to provide more quantitative training, and funding agencies are demanding cutting-edge methodology.

Beth Azar is a writer in Portland, Ore.