14 The Feeling Self: Self-Esteem

As we have noted in our discussions of the self-concept, our sense of self is partly determined by our cognition. However, our view of ourselves is also the product of our affect, in other words how we feel about ourselves. Just as we explored in Chapter 2, cognition and affect are inextricably linked. For example, self-discrepancy theory highlights how we feel distress when we perceive a gap between our actual and ideal selves. We will now examine this feeling self, starting with perhaps its most heavily researched aspect, self-esteem.

Self-Esteem

Self-esteem refers to the positive (high self-esteem) or negative (low self-esteem) feelings that we have about ourselves. We experience the positive feelings of high self-esteem when we believe that we are good and worthy and that others view us positively. We experience the negative feelings of low self-esteem when we believe that we are inadequate and less worthy than others.

Our self-esteem is determined by many factors, including how well we view our own performance and appearance, and how satisfied we are with our relationships with other people (Tafarodi & Swann, 1995). Self-esteem is in part a trait that is stable over time, with some people having relatively high self-esteem and others having lower self-esteem. But self-esteem is also a state that varies day to day and even hour to hour. When we have succeeded at an important task, when we have done something that we think is useful or important, or when we feel that we are accepted and valued by others, our self-concept will contain many positive thoughts and we will therefore have high self-esteem. When we have failed, done something harmful, or feel that we have been ignored or criticized, the negative aspects of the self-concept are more accessible and we experience low self-esteem.

Self-esteem can be measured using both explicit and implicit measures, and both approaches find that most people tend to view themselves positively. One common explicit self-report measure of self-esteem is the Rosenberg Self-Esteem Scale (Figure 3.8). Higher scores on the scale indicate higher self-esteem.

Numerous studies have used the Rosenberg scale to assess people’s self-esteem in many areas of the world. An interesting finding in many samples from the Western world, particularly in North America, is that the average score is often significantly higher than the mid-point. Heine and Lehman (1999), for example, reported meta-analytic data indicating that less than 7% of participants scored below the mid-point! One interesting implication of this is that participants in such samples classified as having low self-esteem on the basis of a median split will typically actually have at least moderate self-esteem.

If so many people, particularly in individualistic cultures, report having relatively high self-esteem, an interesting question is why this might be. Perhaps some cultures place more importance on developing high self-esteem than others, and people correspondingly feel more pressure to report feeling good about themselves (Held, 2002). A problem with measures such as the Rosenberg scale is that they can be influenced by the desire to portray the self positively. The observed scores on the Rosenberg scale may be somewhat inflated because people naturally try to make themselves look as if they have very high self-esteem—maybe they lie a bit to the experimenters to make themselves look better than they really are and perhaps to make themselves feel better. If this the case, then we might expect to find average levels of reported self-esteem to be lower in cultures where having high self-worth is less of a priority. This is indeed what has generally been found. Heine and Lehman (1999) reported that Japanese participants living in Japan showed, on average, moderate levels of self-esteem, normally distributed around the scale mid-point. Many other studies have shown that people in Eastern, collectivistic cultures report significantly lower self-esteem than those from more Western, individualistic ones (Campbell et al., 1996). Do, then, such differences reflect these different cultural priorities and pressures, or could it be that they reflect genuine differences in actual self-esteem levels? There are no easy answers here, of course, but there are some findings from studies, using different methods of measuring self-esteem, that may shed some light on this issue.

Indirect measures of self-esteem have been created—measures that may provide a more accurate picture of the self-concept because they are less influenced by the desire to make a positive impression. Anthony Greenwald and Shelly Farnham (2000) used the Implicit Association Test to study the self-concept indirectly. Participants worked at a computer and were presented with a series of words, each of which they were to categorize in one of two ways. One categorization decision involved whether the words were related to the self (e.g., me, myself, mine) or to another person (e.g., other, them, their). A second categorization decision involved determining whether words were pleasant (e.g., joy, smile, pleasant) or unpleasant (e.g., pain, death, tragedy). On some trials, the self words were paired with the pleasant items, and the other words with the unpleasant items. On other trials, the self words were paired with the unpleasant items, and the other words with the pleasant items. Greenwald and Farnham found that on average, participants were significantly faster at categorizing positive words that were presented with self words than they were at categorizing negative words that were presented with self words, suggesting, again, that people did have positive self-esteem. Furthermore, there were also meaningful differences among people in the speed of responding, suggesting that the measure captured some individual variation in implicit self-esteem.

A number of studies have since explored cross-cultural differences in implicit self-esteem and have not found the same differences observed on explicit measures like the Rosenberg scale (Yamaguchi et al., 2007). Does this mean that we can conclude that the lower scores on self-report measures observed in members of collectivistic cultures are more apparent than real? Maybe not just yet, especially given that the correlations between explicit and implicit measures of self-esteem are often quite small (Heine, Lehman, Markus, & Kitayama, 1999). Nevertheless, values such as modesty may be less prioritized in individualistic cultures than in collectivistic ones, which may in turn reflect differences in reported self-esteem levels. Indeed, Cai and colleagues (2007) found that differences in explicit self-esteem between Chinese and American participants were explained by cultural differences in modesty.

Another interesting aspect of diversity and self-esteem is the average difference observed between men and women. Across many countries, women have been found to report lower self-esteem than men (Sprecher, Brooks, & Avogo, 2013). However, these differences have generally been found to be small, particularly in nations where gender equality in law and opportunity is higher (Kling, Hyde, Showers, & Buswell, 1999). These findings are consistent with Mead’s (1934) suggestion that self-esteem in part relates to the view that others have of our importance in the wider world. As women’s opportunities to participate in careers outside of the home have increased in many nations, so the differences between their self-esteem and that of men have decreased.

There are also some interesting age differences in self-esteem that have been uncovered. In a large Internet survey, Robins, Trzesniewski, Tracy, Gosling, & Potter (2002) found that self-esteem tends to decrease from childhood to early adolescence, and then rises steadily from adolescence into adulthood, usually until people are well into their sixties, after which point it begins to decline. One interesting implication of this is that we often will have higher self-esteem later in life than in our early adulthood years, which would appear to run against ageist stereotypes that older adults have lower self-worth. What factors might help to explain these age-related increases in self-esteem? One possibility relates back to our discussion of self-discrepancy theory in the previous section on the cognitive self. Recall that this theory states that when our perceived self-discrepancy between our current and ideal selves is small, we tend to feel more positive about ourselves than when we see the gap as being large. Could it be that older adults have a current view of self that is closer to their ideal than younger adults, and that this is why their self-esteem is often higher? Evidence from Ryff (1991) suggests that this may well be the case. In this study, elderly adults rated their current and ideal selves as more similar than either  middle-aged or young adults. In part, older adults are able to more closely align these two selves because they are better able to realistically adjust their ideal standards as they age (Rothermund & Brandstadter, 2003) and because they engage in more favorable and age-appropriate social comparisons than do younger adults (Helgeson & Mickelson, 2000).

Maintaining and Enhancing Self-Esteem

As we saw in our earlier discussion of cultural differences in self-esteem, in at least some cultures, individuals appear motivated to report high self-esteem. As we shall now see, they also often actively seek out higher self-worth. The extent to which this is a universal cultural pursuit continues to be debated, with some researchers arguing that it is found everywhere (Brown, 2010), while others question whether the need for positive self-regard is equally valued in all cultures (Heine, Lehman, Markus, & Kitayama, 1999).

For those of us who are actively seeking higher self-esteem, one way is to be successful at what we do. When we get a good grade on a test, perform well in a sports match, or get a date with someone we really like, our self-esteem naturally rises. One reason that many of us have positive self-esteem is because we are generally successful at creating positive lives. When we fail in one domain, we tend to move on until we find something that we are good at. We don’t always expect to get the best grade on every test or to be the best player on the team. Therefore, we are often not surprised or hurt when those things don’t happen. In short, we feel good about ourselves because we do a pretty good job at creating decent lives.

Another way we can boost our self-esteem is through building connections with others. Forming and maintaining satisfying relationships helps us to feel good about ourselves. A common way of doing this for many people around the world is through social networking sites. There are a growing number of studies exploring how we do this online and the effects that it has on our self-worth. One common way on Facebook is to share status updates, which we hope that our friends will then “like” or comment on. When our friends do not respond to our updates, however, this can negatively impact how we feel about ourselves. One study found that when regular Facebook users were assigned to an experimental condition where they were banned from sharing information on Facebook for 48 hours, they reported significantly lower levels of belonging and meaningful existence. In a second experiment, participants were allowed to post material to Facebook, but half of the participants’ profiles were set up by the researchers not to receive any responses, whether “likes” or comments, to their status updates. In line with predictions, that group reported lower self-esteem, level of belonging, level of control, and meaningful existence than the control group who did receive feedback (Tobin, Vanman, Verreynne, & Saeri, 2014). Whether online or offline, then, feeling ignored by our friends can dent our self-worth. We will explore other social influences on our self-esteem later in this chapter.

Research Focus

Processing Information to Enhance the Self

Although we can all be quite good at creating positive self-esteem by doing positive things, it turns out that we often do not stop there. The desire to see ourselves positively is sometimes strong enough that it leads us to seek out, process, and remember information in a way that allows us to see ourselves even more positively.

Sanitioso, Kunda, and Fong (1990) had students read about a study that they were told had been conducted by psychologists at Stanford University (the study was actually fictitious). The students were randomly assigned to two groups: one group read that the results of the research had showed that extroverts did better than introverts in academic or professional settings after graduating from college; the other group read that introverts did better than extroverts on the same dimensions. The students then wrote explanations for why this might be true.

The experimenter then thanked the participants and led them to another room, where a second study was to be conducted (you will have guessed already that although the participants did not think so, the two experiments were really part of the same experiment). In the second experiment, participants were given a questionnaire that supposedly was investigating what different personality dimensions meant to people in terms of their own experience and behavior. The students were asked to list behaviors that they had performed in the past that related to the dimension of “shy” versus “outgoing”—a dimension that is very close in meaning to the introversion-extroversion dimension that they had read about in the first experiment.

Figure 3.9, “Enhancing the Self,” shows the number of students in each condition who listed an extroverted behavior first, and the number who listed an introverted behavior first. You can see that the first memory listed by participants in both conditions tended to reflect the dimension that they had read was related to success according to the research presented in the first experiment. In fact, 62% of the students who had just learned that extroversion was related to success listed a memory about an extroverted behavior first, whereas only 38% of the students who had just learned that introversion was related to success listed an extroverted behavior first.

 

Sanitioso, Kunda, and Fong (1990) found that students who had learned that extroverts did better than introverts after graduating from college tended to list extroverted memories about themselves, whereas those who learned that introverts did better than extroverts tended to list introverted memories.

It appears that the participants drew from their memories those instances of their own behavior that reflected the trait that had the most positive implications for their self-esteem—either introversion or extroversion, depending on experimental condition. The desire for positive self-esteem made events that were consistent with a positive self-perception more accessible, and thus they were listed first on the questionnaire.

Other research has confirmed this general principle—people often attempt to create positive self-esteem whenever possible, even it if involves distorting reality. We tend to take credit for our successes, and to blame our failures on others. We remember more of our positive experiences and fewer of our negative ones. As we saw in the discussion of the optimistic bias in the previous chapter about social cognition, we judge our likelihood of success and happiness as greater than our likelihood of failure and unhappiness. We think that our sense of humor and our honesty are above average, and that we are better drivers and less prejudiced than others. We also distort (in a positive way, of course) our memories of our grades, our performances on exams, and our romantic experiences. And we believe that we can control the events that we will experience to a greater extent than we really can (Crocker & Park, 2004).

Once again, though, there are some important cultural differences to note with people in individualistic cultures pursuing these self-enhancing strategies more vigorously and more often than those from more collectivistic backgrounds. Indeed, in a large-scale review of studies on self-enhancement, Heine (2004) concluded that these tactics are not typically used in cultures that value interdependence over dependence. In cultures where high self-esteem is not as socially valued, people presumably do not feel the same need to distort their social realities to serve their self-worth.

There is also considerable personal diversity in the tendency to use self-enhancement. Stable differences between individuals have been uncovered in many studies across a range of self-enhancing strategies (Hepper, Gramzow, & Sedikides, 2010; John & Robins, 1994; Kwan, John, Kenny, Bond, & Robins, 2004).

Narcissism and the Limits of Self-Enhancement

Our discussion to this point suggests that many people will generally try to view themselves in a positive light. We emphasize our positive characteristics, and we may even in some cases distort information—all to help us maintain positive self-esteem. There can be negative aspects to having too much self-esteem, however, particularly if that esteem is unrealistic and undeserved. Narcissism is a personality trait characterized by overly high self-esteem, self-admiration, and self-centeredness. Narcissists tend to agree with statements such as the following:

• “I know that I am good because everybody keeps telling me so.”
• “I can usually talk my way out of anything.”
• “I like to be the center of attention.”
• “I have a natural talent for influencing people.”

Narcissists can be perceived as charming at first, but often alienate others in the long run (Baumeister, Campbell, Krueger, & Vohs, 2003). They can also make bad romantic partners as they often behave selfishly and are always ready to look for someone else who they think will be a better mate, and they are more likely to be unfaithful than non-narcissists (Campbell & Foster, 2002; Campbell, Rudich, & Sedikides, 2002). Narcissists are also more likely to bully others, and they may respond very negatively to criticism (Baumeister et al., 2003). People who have narcissistic tendencies more often pursue self-serving behaviors, to the detriment of the people and communities surrounding them (Campbell, Bush, Brunell, & Shelton, 2005). Perhaps surprisingly, narcissists seem to understand these things about themselves, although they engage in the behaviors anyway (Carlson, Vazire, & Oltmanns, 2011).

Interestingly, scores on measures of narcissistic personality traits have been creeping steadily upward in recent decades in some cultures (Twenge, Konrath,  Foster, Campbell, & Bushman, 2008). Given the social costs of these traits, this is troubling news. What reasons might there be for these trends? Twenge and Campbell (2009) argue that several interlocking factors are at work here, namely increasingly child-centered parenting styles, the cult of celebrity, the role of social media in promoting self-enhancement, and the wider availability of easy credit, which, they argue, has lead to more people being able to acquire status-related goods, in turn further fueling a sense of entitlement. As narcissism is partly about having an excess of self-esteem, it should by now come as no surprise that narcissistic traits are higher, on average, in people from individualistic versus collectivistic cultures (Twenge et al., 2008).

The negative outcomes of narcissism raise the interesting possibility that high self-esteem in general may not always be advantageous to us or to the people around us. One complication to the issue is that explicit self-report measures of self-esteem, like the Rosenberg scale, are not able to distinguish between people whose high self-esteem is realistic and appropriate and those whose self-esteem may be more inflated, even narcissistic (Baumeister et al., 2003). Implicit measures also do not provide a clear picture, but indications are that more narcissistic people score higher on implicit self-esteem in relation to some traits, including those relating to social status, and lower on others relating to relationships (Campbell, Bosson, Goheen, Lakey, & Kernis, 2007). A key point is that it can be difficult to disentangle what the effects of realistic versus unrealistic high self-esteem may be. Nevertheless, it is to this thorny issue that we will now turn.

 Social Psychology in the Public Interest

Does High Self-Esteem Cause Happiness or Other Positive Outcomes?

Teachers, parents, school counselors, and people in many cultures frequently assume that high self-esteem causes many positive outcomes for people who have it and therefore that we should try to increase it in ourselves and others. Perhaps you agree with the idea that if you could increase your self-esteem, you would feel better about yourself and therefore be able to work at a higher level, or attract a more desirable mate. If you do believe that, you would not be alone. Baumeister and colleagues (2003) describe the origins and momentum of what they call the self-esteem movement, which has grown in influence in various countries since the 1970s. For example, in 1986, the state of California funded a task force under the premise that raising self-esteem would help solve many of the state’s problems, including crime, teen pregnancy, drug abuse, school underachievement, and pollution.

Baumeister and colleagues (2003) conducted an extensive review of the research literature to determine whether having high self-esteem was as helpful as many people seem to think it is. They began by assessing which variables were correlated with high self-esteem and then considered the extent to which high self-esteem caused these outcomes. They found that high self-esteem does correlate with many positive outcomes. People with high self-esteem get better grades, are less depressed, feel less stress, and may even live longer than those who view themselves more negatively. The researchers also found that high self-esteem is correlated with greater initiative and activity; people with high self-esteem just do more things. They are also more more likely to defend victims against bullies compared with people with low self-esteem, and they are more likely to initiate relationships and to speak up in groups. High self-esteem people also work harder in response to initial failure and are more willing to switch to a new line of endeavor if the present one seems unpromising. Thus, having high self-esteem seems to be a valuable resource—people with high self-esteem are happier, more active, and in many ways better able to deal with their environment.

On the other hand, Baumeister and his colleagues also found that people with high self-esteem sometimes delude themselves. They tend to believe that they are more likable and attractive, have better relationships, and make better impressions on others than people with low self-esteem. But objective measures show that these beliefs are often distortions rather than facts. Furthermore, people with overly high self-esteem, particularly when it is accompanied by narcissism, defensiveness, conceit, and the unwillingness to critically assess one’s potential negative qualities, have been found to engage in a variety of negative behaviors (Baumeister, Smart, & Boden, 1996). For example, people with high self-esteem are more likely to be bullies (despite also being more likely to defend victims)  and to experiment with alcohol, drugs, and sex.

Todd Heatherton and Kathleen Vohs (2000) found that when people with extremely high self-esteem were forced to fail on a difficult task in front of a partner, they responded by acting more unfriendly, rudely, and arrogantly than did those with lower self-esteem. And research has found that children who inflate their social self-worth—those who think that they are more popular than they really are and who thus have unrealistically high self-esteem—are also more aggressive than children who do not show such narcissistic tendencies (Sandstrom & Herlan, 2007; Thomaes, Bushman, Stegge, & Olthof, 2008). Such findings raise the interesting possibility that programs that increase the self-esteem of children who bully and are aggressive, based on the notion that these behaviors stem from low self-esteem, may do more harm than good (Emler, 2001). If you are thinking like a social psychologist, these findings may not surprise you—narcissists tend to focus on their self-concerns, with little concern for others, and we have seen many times that other-concern is a necessity for satisfactory social relations.

Furthermore, despite the many positive variables that relate to high self-esteem, when Baumeister and his colleagues looked at the causal role of self-esteem they found little evidence that high self-esteem caused these positive outcomes. For instance, although high self-esteem is correlated with academic achievement, it is more the result than the cause of this achievement. Programs designed to boost the self-esteem of pupils have not been shown to improve academic performance, and laboratory studies have generally failed to find that manipulations of self-esteem cause better task performance.

Baumeister and his colleagues concluded that programs designed to boost self-esteem should be used only in a limited way and should not be the only approach taken. Raising self-esteem will not make young people do better in school, obey the law, stay out of trouble, get along better with other people, or respect the rights of others. And these programs may even backfire if the increased self-esteem creates narcissism or conceit. Baumeister and his colleagues suggested that attempts to boost self-esteem should only be carried out as a reward for good behavior and worthy achievements, and not simply to try to make children feel better about themselves.

Although we naturally desire to have social status and high self-esteem, we cannot always promote ourselves without any regard to the accuracy of our self-characterizations. If we consistently distort our capabilities, and particularly if we do this over a long period of time, we will just end up fooling ourselves and perhaps engaging in behaviors that are not actually beneficial to us. Most of us probably know someone who is convinced that he or she has a particular talent at a professional level, but we, and others, can see that this person is deluded (but perhaps we are too kind to say this). Some individuals who audition on television talent shows spring to mind. Such self-delusion can become problematic because although this high self-esteem might propel people to work harder, and although they may enjoy thinking positively about themselves, they may be setting themselves up for long-term disappointment and failure. Their pursuit of unrealistic goals may also take valuable time away from finding areas they have more chance to succeed in.

When we self-enhance too much, although we may feel good about it in the short term, in the longer term the outcomes for the self may not be positive. The goal of creating and maintaining positive self-esteem (an affective goal) must be tempered by the cognitive goal of having an accurate self-view (Kirkpatrick & Ellis, 2001; Swann, Chang-Schneider, & Angulo, 2007). In some cases, the cognitive goal of obtaining an accurate picture of ourselves and our social world and the affective goal of gaining positive self-esteem work hand in hand. Getting the best grade in an important exam produces accurate knowledge about our skills in the domain as well as giving us some positive self-esteem. In other cases, the two goals are incompatible. Doing more poorly on an exam than we had hoped produces conflicting, contradictory outcomes. The poor score provides accurate information about the self—namely, that we have not mastered the subject—but at the same time makes us feel bad. Self-verification theory states that people often seek confirmation of their self-concept, whether it is positive or negative (Swann, 1983). This sets up a fascinating clash between our need to self-enhance against our need to be realistic in our views of ourselves. Delusion versus truth: which one wins out? The answer, of course, as with pretty much everything to do with human social behavior, is that it depends. But on what does it depend?

One factor is who the source is of the feedback about us: when we are seeking out close relationships, we more often form them with others who verify our self-views. We also tend to feel more satisfied with interactions with self-verifying partners than those who are always positive toward us (Swann, De La Ronde, & Hixon, 1994; Swann & Pelham, 2002). Self-verification seems to be less important to us in more distant relationships, as in those cases we often tend to prefer self-enhancing feedback.

Another related factor is the part of our self-concept we are seeking feedback about, coupled with who is providing this evaluation. Let’s say you are in a romantic relationship and you ask your partner and your close friend about how physically attractive they think you are. Who would you want to give you self-enhancing feedback? Who would you want more honesty from? The evidence suggests that most of us would prefer self-enhancing feedback from our partner, and accuracy from our friend (Swann, Bosson, & Pelham, 2002), as perceived physical attractiveness is more central to romance than friendship.

Under certain conditions, verification prevails over enhancement. However, we should not underestimate the power of self-enhancement to often cloud our ability to be more realistic about ourselves. For example, self-verification of negative aspects of our self-concept is more likely in situations where we are pretty sure of our faults (Swann & Pelham, 1988). If there is room for doubt, then enhancement tends to rule. Also, if we are confident that the consequences of getting innaccurate, self-enhancing feedback about negative aspects ourselves are minimal, then we tend to welcome self-enhancement with open arms (Aronson, 1992).

Therefore, in those situations where the needs to enhance and to verify are in conflict, we must learn to reconcile our self-concept with our self-esteem. We must be able to accept our negative aspects and to work to overcome them. The ability to balance the cognitive and the affective features of the self helps us create realistic views of ourselves and to translate these into more efficient and effective behaviors.

There is one final cautionary note about focusing too much on self-enhancement, to the detriment of self-verification, and other-concern. Jennifer Crocker and Lora Park (2004) have identified another cost of our attempts to inflate our self-esteem: we may spend so much time trying to enhance our self-esteem in the eyes of others—by focusing on the clothes we are wearing, impressing others, and so forth—that we have little time left to really improve ourselves in more meaningful ways. In some extreme cases, people experience such strong needs to improve their self-esteem and social status that they act in assertive or dominant ways in order to gain it. As in many other domains, then, having positive self-esteem is a good thing, but we must be careful to temper it with a healthy realism and a concern for others. The real irony here is that those people who do show more other- than self-concern, those who engage in more prosocial behavior at personal costs to themselves, for example, often tend to have higher self-esteem anyway (Leak & Leak, 2003).