Self-esteem is such a commonly used term you probably already know what it is: thinking highly of yourself. You have probably heard self-esteem mentioned on talk shows, in magazine articles, and even in popular songs (the song “The Greatest Love of All” is about loving yourself, and there’s a song by the band The Offspring called “Self-Esteem.”) But social psychology research has discovered a lot of things about self-esteem that have not yet made it to popular culture, and this research might surprise you.
Academic psychologists recognize two types of self-esteem. The first is general self-esteem, often measured using the Rosenberg Self-Esteem Scale (which includes items such as “I take a positive attitude toward myself”). The second type of self-esteem is specific, often measuring self-esteem in a particular domain such as school, work, athletics, or appearance. These subdomains are then combined to form a complete self-esteem score (for example, in scales such as the Tennessee Self-Concept Scale or the Coopersmith Self-Esteem Inventory). Although nonpsychologists sometimes use the term self-esteem to refer to body and appearance concerns, a psychologist is more specific and instead calls these body image or appearance self-esteem.
People high in self-esteem seem to know more about themselves and their preferences. They can furnish longer lists of their likes and dislikes, and they are more confident about their self-ratings. They are also more self-serving; they are more likely to take credit for their successes and blame outside sources for their failure. Self-esteem is also correlated with emotional stability: People with low self-esteem experience negative moods more often and report more fluctuation in their moods.
Self-Esteem Differences and Predictors
Which groups of people are high in self-esteem, and which are low? You might have heard that teenage girls have very low self-esteem, but this is not true. Men and boys do score higher on self-esteem than women and girls, but the difference is small; gender explains only about 1% of the differences in self-esteem (this number tells you how much of the variation in self-esteem is caused by a specific variable—here, gender—rather than by other factors). The gap does widen a bit during adolescence, with gender explaining about 2.6% of the differences and boys scoring higher. But this doesn’t happen because girls’ self-esteem drops at adolescence; girls’ self-esteem rises between middle school and high school, but just not as much as boys’ does. Between high school and college, women’s self-esteem increases sharply, and the gender difference shrinks back to 1% of the variance.
Are rich and well-educated people higher in self-esteem? Yes, but not by much—socioeconomic status explains less than 1% of the variance in self-esteem. The correlation between socioeconomic status and self-esteem peaks during middle age, but even then, it accounts for only 1.5% of the differences. So social status and money are only very weak predictors of self-esteem.
What about racial and ethnic differences—are racial minorities, many of whom experience prejudice, more likely to be low in self-esteem? The answers here are complex: Overall, racial differences in self-esteem seem to be caused more by cultural differences than by racial discrimination. Black Americans, who probably experience the most prejudice and discrimination in the United States, actually score higher in self-esteem than are White Americans (though this is yet another of those 1% of the variance small findings). This might occur because they protect their self-esteem by attributing criticism to prejudice (a theory called stigma as self-protection). However, Hispanic Americans score lower than Whites do in self-esteem (though this is a very small difference accounting for only about .2% of the variance), and they experience prejudice as well. So prejudice alone cannot explain why Blacks score higher on self-esteem measures. Cultural differences provide a more consistent explanation. Black American culture champions self-respect, whereas Asian cultures emphasize humility and self-criticism. Sure enough, Asian Americans score lower on self-esteem than do Whites, a somewhat larger difference that explains 2.2% of the variance. Asians living in Asia score even lower compared with White Americans, a difference that explains about 4.5% of the variance. These differences are all consistent with the idea that cultural ideas about the self influence levels of self-esteem.
Cultural differences can happen over time and generations as well. The culture of 1950s America was very different from the culture of 1990s America, and one of the main differences is the increased emphasis on the self during recent decades. And indeed, 1990s college students scored higher on self-esteem measures than did 1960s college students, a difference that explains 9% of the variance in self-esteem scores. Overall, culture (of time and regions) is a stronger influence on self-esteem than is being a certain race, gender, or income level.
So what does self-esteem cause? In psychological language, what are the outcomes of self-esteem? You might have heard that high self-esteem leads to better academic achievement and less bad behavior like aggression and teen pregnancy. However, a large body of research suggests that this is not the case. Self-esteem does explain about 5% of the variance in school achievement, a small but statistically significant effect. However, as in any correlational study, there are three possibilities: High self-esteem could cause school achievement, school achievement could cause high self-esteem, or a third variable (such as income level) could cause both. To use a common analogy, the horse could be pulling the cart, or things could be reversed and the cart has been put before the horse. A third variable resembles the horse and the cart being towed on a flatbed truck: Neither the cart nor the horse is causing the motion in the other even though they are moving together.
Most studies have found that achievement leads to self-esteem, not vice versa. Another set of studies finds that controlling for third variables (such as family income) eliminates the correlation. This occurs because rich kids are both higher in self-esteem and do better in school. Self-esteem is also not consistently correlated with alcohol and drug abuse or teen pregnancy. Some studies have found that high self-esteem actually predicts earlier intercourse among teens. Overall, self-esteem does not seem to cause good outcomes for kids; the two are unrelated.
Despite this research, numerous school programs aim to increase children’s self-esteem. A 2006 Google search showed that more than 300,000 elementary schools mention self-esteem in their mission statements. Most of these say that they seek to encourage or develop children’s self-esteem. Some of these programs promote self-esteem without rooting it in achievement, in the belief that children should feel good about themselves no matter what they do. Although the results of these programs are continuing to be debated, it seems likely that they will not have much impact if self-esteem does not cause achievement and good behavior (which appears to be the case).
There has recently been some debate about whether low self-esteem leads to antisocial behavior. Experimental lab studies consistently find no correlation between self-esteem and aggression. Two recent correlational studies, however, found that low self-esteem was correlated with delinquent behavior in a sample of adolescents, even after controlling for academic achievement, income, and parental support. Other variables, such as associating with delinquent friends, might explain the effect, which accounts for about 4% of the variance in delinquent behavior. Overall, the evidence suggests that self-esteem is not correlated with aggression, but that low self-esteem is linked to a slightly higher incidence of delinquent behavior.
Some evidence also indicates that low self-esteem is linked to eating disorders such as anorexia and bulimia. However, low self-esteem only predicts eating disorders when women are perfectionistic and feel overweight. Low self-esteem might also follow, rather than precede, eating disorders: People might start to feel badly about themselves after they develop an eating disorder.
One thing self-esteem does strongly predict is happiness. People who are high in self-esteem report being happy, and they are also less likely to be depressed. However, these studies have not proven causation and ruled out other third variable explanations, so further research needs to be done: It is not yet known if self-esteem causes happiness, happiness causes self-esteem, or if some other variable causes both. Self-esteem also leads to greater persistence on tasks, though the causation is not known here, either, and self-control is a better predictor of persistence. Self-esteem is also correlated with greater relationship confidence. High self-esteem people who experience a threat to their self-worth are subsequently more certain of their partners’ regard for them; in contrast, low self-esteem people began to doubt their partners’ feelings, which can cause problems in the relationship.
The stability of self-esteem also plays a role. People whose self-esteem fluctuates wildly, or whose self-esteem heavily depends on a particular outcome, are more likely to be depressed and anxious. Stable self-esteem, and self-esteem that does not depend on certain things happening, is correlated with better mental health.
Origins of Self-Esteem
Where does self-esteem come from, and how does it develop in a child? One theory proposes that self-esteem is a sociometer, or a gauge of how accepted people feel by other people. Thus, self-esteem arises from feeling loved by others and belonging to groups. This theory also helps explain the main difference between self-esteem and narcissism. Narcissism is an inflated sense of self, but it goes beyond simply having very high self-esteem. Narcissists believe that they are better than others in achievement realms such as intellectual ability and sports. However, they acknowledge that they are not particularly friendly or moral. Perhaps as a result, narcissism is correlated with poor relationship outcomes: Narcissists lack empathy, are more likely to derogate their partners, and are more likely to cheat. They are also more aggressive in response to threat.
People are very motivated to preserve their self-esteem and good feelings about themselves, and this motive explains a surprising amount of human behavior. Many people tend to credit themselves when things go well, and blame others or luck when things go badly. This is called self-serving bias, and you can easily see how it preserves good self-feelings. Self-esteem boosting also explains ingroup bias, in which people believe that their own group is better than other groups. In other words, prejudice against people unlike ourselves may be rooted in our desire to feel good about ourselves. One set of researchers believes that the ultimate self-preservation—pushing away thoughts about death— explains patriotism and ingroup bias. They find that when people are reminded of death, they strongly defend their own worldviews. Another study found that when high self-esteem people are threatened, they respond by acting more boastful and rude.
Overall, self-esteem does not explain as many things as most people believe it does. Self-esteem is good for relationships, but only if it does not cross over into narcissism. People with high self-esteem are happier, but their self-esteem does not cause good things to happen in their lives. Instead, the pursuit of self-esteem can sometimes lead people to behave in ways that they might later regret.
- Baumeister, R. F., Campbell, J. D., Krueger, J. E., & Vohs, K. D. (2003). Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the Public Interest, 4, 1-44.
- Twenge, J. M., & Campbell, W. K. (2001). Age and birth cohort differences in self-esteem: A cross-temporal meta-analysis. Personality and Social Psychology Review, 5, 321-344.