Unfortunately, the benefits to society from the increase in self-esteem are unclear and may actually have no correlation at all to a better overall lifestyle, according to two psychology professors-one from the University of Georgia-who conducted the new analysis.
"There are many potential benefits of self-esteem, but we wanted to see, among other things, whether those benefits are clear when viewed across time," said Dr. W. Keith Campbell, an assistant professor of psychology at UGA. "Has this increase among college students resulted in societal boons? Unfortunately, during that time, few positive changes have occurred in children's and young adults' behavior. Indeed, most of the relevant behavioral indicators have worsened."
Just what this means right now remains unclear, but Campbell believes the results may mean that the self-esteem movement that spread across the U.S. over the past few decades may actually be doing more harm than good.
The research, which was conducted with Dr. Jean Twenge of the department of psychology at San Diego State University, will be published this month in the journal Personality and Social Psychology Review.
Researchers have been studying self-esteem in general for decades, but this is the first analysis using a new technique-developed by Twenge-called cross-temporal meta-analysis. This method allows a simultaneous examination of age and group effects and allows researchers to understand generational differences in self-esteem.
"College students' self-esteem has increased substantially, but this happened at the same time that SAT scores declined and anxiety increased," said Twenge. "Thus college students' high self-esteem seems to be built on a foundation of sand."
Just how self-esteem can be measured in groups has been the subject of intense discussion for some time. For college students, the method most often used is called the Rosenberg Self-Esteem Scale (RSE). For children, psychologists use the Coopersmith Self-Esteem Inventory (SEI).
While the RSE showed a steady increase in self-esteem for college students between 1968 and 1994 (in male, female and mixed-sex groups), the SEI showed that self-esteem for children decreased between 1965 and 1979 and then increased from1980 to 1993.
The psychologists considered three main models of self-esteem. The social acceptance model proposes that self-esteem arises from others' acceptance. The competencies model suggests that self-esteem is based on the perception of competence in certain areas of life. And the culture of self-worth model predicts that a culture promoting a focus on the self will lead to higher self-esteem.
"The emphasis of the educational system on self-esteem really got kicked off around 1980," said Campbell, "and it elevated child self-esteem somewhat. It seemed like an easy fix-that if people felt good about themselves they would perform better. Unfortunately, the self-esteem movement might have increased self-esteem, but it didn't increase competency in anything. If anything, it decreased such things."
The analysis examined information from about 65,000 adults in the RSE studies and 40,000 children in the SEI tests.
Campbell said such outgrowths of the self-esteem movement as games in which no one loses and classes that are structured to avoid "hurting the feelings" of participants and thereby increase self-esteem are at best misguided and at worst harmful.
"Personally, I think the self-esteem movement is questionable," said Campbell. "If we forgot about self-esteem as a society, we'd probably be just fine."
The study, however, was published with a number of caveats that Campbell said are important to consider.
First, the SEI shows a drop in self-esteem at adolescence while the RSE does not. This may occur because the SEI considers more specific areas of self-concept, such as the family, peers and school. Second, college students may be answering questions about self-esteem in the affirmative because they have been taught that self-esteem is a socially desirable trait.
Finally, the differences in scores on the SEI and the RSE make it clear that no one measure is ideally suited to all age groups.
"When we do find correlations between social indicators and self-esteem, they occur for children and not for adults," said Campbell. "In short, although self-esteem is on the rise, it is difficult to find many clear, positive benefits of this increase when viewing broader cultural-level changes. We as individuals may think more highly of ourselves, but we as a society apparently have little to show for it."