Who Needs Real Managers When You've Got Fraternity Boys?

Pretend for a moment that you're a psychology professor at Oshkosh Universityand you're looking to do some research. Perhaps you want to examine why some couples stay in love and others wind up on Sally Jessy Raphael. Or maybe you're curious how politicians' hair styles affect the preferences of Republican voters. It doesn't matter what the topic is—you want to understand some facet of human behavior, and you've designed a clever study to figure it out.

The trouble is, you're missing one crucial ingredient to actually do the study: married people, or conservative voters. Fortunately, there's a solution. Like many universities, the Oshkosh campus is swarming with, well, students. Why not just have some sophomores pretend to be old married couples, or Republicans, and report on their behavior instead?

Silly as it sounds, this scenario isn't all that rare. When University of Kentucky psychologist Michael Nietzel, Ph.D., reviewed the research done on jury behavior through 1993, he found that 90 percent of the studies had used simulated juries—composed, in most cases, of students. "I will not take the nihilistic position that [mock juries] cannot teach us anything," Nietzel wrote. "However, it does seem appropriate to suggest that investigators who study jury phenomenon...be expected, once in awhile, to study real juries."

The problem with using college students, of course, is that they are not like the rest of us. Yes, they're better educated than the average person. But they're also more liberal, more prone to risk-taking, and more likely to sneak a cow into the dean's office as a prank. These factors, among others, could affect how students behave during a study.

Take, for example, the article "Courting the Jury" that appears on page 9 of this issue. The story describes a study that found that jurors often don't understand courtroom "legalese" and thus may rely on the judge's body language to determine a defendant's guilt. Giving juries instructions in plain English, it turned out, eliminated the problem. What the story doesn't tell you is that the researchers performed the experiment twice. When a jury was composed of adults of various ages and education levels, the simplified instructions helped. But in an initial experiment performed largely on Harvard students, the plain English instructions didn't make much of a difference. If the researchers had simply relied on student "jurors," they might have missed a finding with important real-world implications.

The issue has also plagued studies of workplace behavior. In 1996, Angelo DeNisi, Ph.D., of Texas A&M University, published the latest in a long line of experiments (by him and others) looking at how people appraise the performance of others. But his was the first done outside of a research lab.

Why not use actual managers for all such studies? "I would love to," says DeNisi. "But it's difficult to convince organizations to allow you access to their employees when they may not see an immediate payoff." Indeed, the efforts required to find willing participants can be positively Herculean. "I'm doing a research project with a student, and she's contacted 45 organizations to ask for a half hour of their employees' time," says Kevin Murphy, Ph.D., a psychologist at Colorado State University. "And the answer is always `no.'"

Another problem in using student "managers)' is that they are more prone to behave idealistically. Suppose you ask study participants to select a new member for a task force, and one of the candidates is physically disabled. "Students know what they're 'supposed' to do," says DeNisi: pick the disabled candidate. "They're more willing, I think, to do what is socially desirable than real managers, who have other forces operating on them. In the real world, it's not easy to do the right thing all of the time."

Despite these pitfalls, it often does make sense to use students as guinea pigs, Murphy notes. "If you're dealing with basic psychological processes"—say, how short-term memory works--"student-based research can be extremely informative." It's also a logical first step in new areas of research: there's little point in performing an elaborate field study until you have some preliminary lab evidence.

What does all this mean for you? Pay attention to the details of research studies—the stuff magazine articles and news reports bury at the end (or don't mention at all). Just as heart-disease research performed on elderly women may not mean much for 21-year-old fret guys, make sure that psych experiments done on those fret guys applies to you.