The evidence on self-esteem

In this section we shall look at how self-esteem is measured and what research says about the importance of self-esteem to a variety of outcomes. As we shall see high self-esteem can have important beneficial outcomes ? happiness, for example - yet in other respects it is unimportant or may even be undesirable.
Measuring self-esteem

There are many different ways to define self-esteem so there are a number of ways to measure it. One of the most widely used measurement tools is Rosenberg?s Self-Esteem Scale which has been in use since the 1960s. This self-report questionnaire includes questions such as ?I take a positive attitude towards myself? and ?I wish I could have more respect for myself?. (The full questionnaire can be viewed in tools, tips and techniques in this section.) Rosenberg?s scale is widely used in research and has been shown to be one of the most reliable measures. Professor Roy Baumeister accepts that as a global rating of self-esteem Rosenberg?s scale is acceptable. Indeed he believes that a person?s response to one single item ??I have high self-esteem? - may be sufficient to classify individuals into high and low self-esteem.

While Professor Baumeister accepts the validity and reliability of such measures, in his research into the importance of self-esteem, he thought it vital to make a distinction between self-reported self-esteem, which he thinks may be inflated because of social pressures, and self-esteem which is measured, or corroborated, more objectively. By deciding to focus on analysing self-esteem studies which had more robust, objective measures the number of studies for him to examine dropped dramatically. Baumeister was also intent on being rigorous about causality. Being able to point to a correlation meant nothing since it was impossible to tell what caused what. For example, if drug users had low self-esteem was this a cause of the drug use or how they felt about themselves because they used drugs? Or was there another factor which caused both the drug taking, and the low self-esteem, such as poverty?

Professor Nicholas Emler is also relaxed about the measurement of self-esteem but concerned about the issue of causality. His research mainly involved looking at longitudinal studies which tracked children over time.

Level of self-esteem

Baumeister et al have shown that few people report low self-esteem. The norm, for the American population at least, is on the high side. This leads Professor Emler In his book on self-esteem, to point out that when researchers refer to people who have high self-esteem these people have a very positive view of themselves and those with low self-esteem a 'slightly' positive view of themselves.

Self-esteem research

The following are some of the findings which have emanated from research undertaken mainly by Baumeister and colleagues and set out in their 2003 paper and Professor Emler?s researched which he published in 2001.

Academic performance

Baumeister?s early research indicated a very weak link between academic performance and self-esteem. This conclusion is amply demonstrated in unpublished research into the 1970 British Cohort Study reported by Nicholas Emler. This showed that self-esteem levels at age 10 were only 'trivially related to later educational attainments'. It also has to be borne in mind that any link between self-esteem and performance may be due to the fact that good results enhance self-esteem rather than vice versa.

More recently the argument on self-esteem and academic performance is that there is little relationship. For example, the most high achieving students in American schools ? white girls ? often get the lowest self-esteem scores. This is also true of Asian students. Asian culture does not encourage self-esteem yet young Asians often excel academically. In the U.S. the group who consistently score high on self-esteem are black boys and they often achieve the least academically at school.

The factors which have consistently been shown in research to have an influence on academic achievement are: ability, IQ, and family background (socio-economic status).

Baumeister also argues that any attempt to artificially boost self-esteem for young people may backfire. He explains why in his 2003 paper:

? students may ordinarily work hard in order to be permitted to feel good about themselves, and an intervention that encourages them to feel good about themselves regardless of work may remove the reason to work hard ? resulting in poor performance.

As psychologists Professor Baumeister and his colleagues are interested in psychological studies and do not make reference to educational trends on academic achievement. We shall examine these trends more fully in the section devoted to educational themes but it is worth pointing out here that recent commentators, like Professor Jean Twenge in Generation Me, point out the worrying drop in  educational attainment figures in the USA over the past few decades. Twenge is not arguing that the decline in academic standards in US schools is attributable exclusively to the emphasis on building self-esteem in schools but, like a number of critics, she does argue it is a contributing factor.


It is often argued that you have to be able to ?love yourself? before you can love others or have good relationships but self-esteem is not as important for relationships as was previously thought. People who record high self-esteem often say they have better social and inter-personal skills life than people with low self-esteem but this is not actually the case. Studies have shown that people who have high self-esteem often have an inflated sense of their social skills since this view is often contradicted by others. One study did find, however, that people with high self-esteem were more likely to initiate new relationships and interactions and this was supported by others. It is not difficult to see why confidence would be helpful in this respect. On the other hand, having an inflated sense of yourself is not a characteristic which other people find endearing.

Low self-esteem may affect a person?s love life in that they do not think they are worthy of love but their relationships may be more stable and last. People with high self-esteem report that they are more likely to terminate relationships if they are dissatisfied. This could lead to more divorces and broken homes for children.

Sexual behaviour

Low self-esteem is linked to teenage pregnancy. Professor Emler accepts that there is perhaps a 50% increase of teenage pregnancy among teenage girl with lower self-esteem than their peers. However, he also points out that it is difficult to ascertain causality and the low self-esteem and the liklihood of becoming pregnant may be due to a third factor such as physical abuse or neglect.

Young people who record high self-esteem are more likely to indulge in sexual behaviour as they may be more confident about being independent and taking risks.  


Ordinary people very much believe that there is a close link between attractiveness and self-esteem. Certainly the evidence shows that people who report high self-esteem are also more likely to rate themselves as attractive. However, when a person?s attractiveness is rated objectively by others then the link between attractiveness and self-esteem vanishes.

Drugs, alcohol and smoking

There is no great correlation between self-esteem and abusing alcohol or taking drugs. There is a weak correlation between smoking and low self-esteem in girls but it is possible that smoking reduces self-esteem rather than vice versa.

Eating disorders

According to Professor Emler low self-esteem is a risk factor for eating disorders such as anorexia and bulimia in young girls. But again it is difficult to establish causality and there may be a third factor involved such as neglect or child abuse.

Unemployment/low income

Professor Emler reported research which shows that low self-esteem is a risk factor for young men as far as low income/ unemployment is concerned.


Various people have argued that low self-esteem is likely to manifest itself in racist attitudes but the evidence does not support this conclusion. Indeed people with high self-esteem are more likely to express unfavourable attitudes to groups to which they don?t belong and this can fuel racist prejudices. Those with low self-esteem are often negative about themselves and others.


Low self-esteem does not lead people into violence. In fact, research indicates that those indulging in violent behaviour are more likely to have high, not low, self-esteem. This may be because of narcissism which we shall investigate in a later section.  Professor Emler reports that in their research project they put sociopaths through every possible test and could not come up with evidence that these men did not like or value themselves enough.


On bullying the evidence seems to suggest that bullies can either report low self-esteem or high self-esteem. (Again the link here might be narcissism.) It is not difficult to see why people who think well of themselves may be inclined to pick on others. However, there is also evidence that those who are most likely to stand up to bullies and defend victims are also most likely to report high self-esteem. There is evidence that those who have low self-esteem are more likely to be picked on and victimised.

Anti-social behaviour

In this category we can list anti-social behaviour such as delinquency  (eg. stealing and vandalism) as well as cheating. As far as delinquency is concerned studies are contradictory but there is, on balance, some evidence of a weak link between low self-esteem and this type of behaviour. Little correlation has been found between disciplinary problems in school and self-esteem. When it comes to cheating, children who report high self-esteem are over-represented in groups that are most, or least, likely to cheat.  Baumeister et al explain this finding in the following way:

The results were consistent with the view that self-esteem is a heterogeneous construct, in that children who cheated the most and who cheated the least both had high self-esteem. More precisely, those who were high in both self-esteem and need for approval cheated the most, whereas those who were high in self-esteem but had a low need for approval cheated least. The authors proposed that a distinction between true and defensive high self-esteem could account for their results.

Behaviour in groups

There haven?t been many studies examining the effect of self-esteem in group settings but one fairly in-depth research study indicated that people with high self-esteem are marginally more likely to speak out. The difference between high and low self-esteem group members in speaking out wasn?t very significant until the group became large.

Performance at work or on specific tasks

The Work Foundation publication ?Me, Myself, Work?, mentioned in the previous section, asserts that self-esteem is very important for performance at work. In fact much of the pamphlet is predicated on this assumption. But the 2003 paper by Baumeister and colleagues argues that while this is a common assumption it is not necessarily true. They refer to one 2001 study of the relationship between work performance and self-esteem where the researchers surveyed the results of 5,000 participants across 40 studies and found that ?most of them showed weak positive relationships?. However, unlike school where across the board it was difficult to see any strong relationship between self-esteem and academic achievement, ?the reported link between self-esteem and job performance is highly variable?. They make the point that this may be because jobs vary considerably and some may require more self-esteem for good performance than others. However, they also cautioned not to assume causality. Self-esteem might in some cases improve job performance. On the other hand being good at your job may lead to an increase in self-esteem.

A number of laboratory studies have tried to ascertain a link between self-esteem and performance but again Baumeister reports ?self-esteem seems to have little or no direct relationship to task performance?. The tasks studied were arithmetic problems and video games requiring non-verbal intelligence.

One area relevant for work where self-esteem does seem significant is ?persisting longer in the face of failure? and also knowing when to quit. Another 2002 study also shows that students with high self-esteem were more likely to achieve their goals and were more likely to get satisfaction from achieving them than those with low self-esteem. These types of studies lead Baumeister et al to assert that ?in performance contexts, high self-esteem people appear to use better self-regulation strategies than low self-esteem people?.

While the Centre is generally persuaded by the research presented by Baumeister and colleagues on the relevance of self-esteem to various aspects of people?s lives, we do not find the evidence presented on ?job and task performance? entirely convincing. As well as the ?self-regulation? strategies initiating conversation, and speaking out in large groups mentioned above, we shall see below that those with high self-esteem have better coping skills and, since they are more likely to be happy, may well have a slightly more positive approach to life.  Certainly for some jobs these skills/attitudes are unimportant but if you examine the type of employees/skills employers say they want in the UK, high self-esteem may be a more important factor in the modern workplace than Baumeister and colleagues acknowledge . (For a fuller discussion of this see a later section.)


Baumeister et al conclude from examining a large variety of studies of various age groups that there is a close correlation between self-esteem and reported happiness levels. This is particularly true in individualistic cultures like the USA (and much of western Europe) and less true in collectivist cultures which urge people to pay more attention to the group and relationships than to internal feelings or ?attributes?.

As a result of these wide-ranging studies, Baumeister et al write:

Taken together these findings uniformly indicate that self-esteem and happiness are strongly interrelated. They suggest that high self-esteem may pay off handsomely for the individual in terms of subjective happiness.

They go on, however, to point out that since happiness is a subjective state of mind it cannot easily be corroborated objectively. Given that people with high self-esteem tend to say positive things about themselves (such as having good social skills) even when this is not always the case, it is possible that they overrate how happy they feel about their lives as well. The authors caution us to be aware that it is still not possible to say that high self-esteem causes happiness. It may be that having success in life causes happiness to rise. Or it could be some kind of dispositional tendency to feel good tends to lead to high self-esteem. Despite these reservations the authors still conclude: 'People with high self-esteem are significantly, substantially happier than other people.'

Coping and depression

A number of writers on self-esteem have advanced the idea that high self-esteem may act as a ?buffer? which protects the individual when times are tough. In other words, when life is going well self-esteem may not matter very much but it helps individuals weather the storm when things go wrong.

Various studies have tried to substantiate, or undermine, the buffer hypothesis and the results are very mixed. Some confirmed the idea that high self-esteem has a protective effect when life was hard or stressful. Others indicated that the advantage of self-esteem was when life was good, not bad. Some studies also concluded that low self-esteem ?poisoned the good times? and that everyone suffered when times were hard. However, Baumeister and colleagues were still prepared to conclude that while more research is needed ?the findings consistently suggest that low self-esteem leads to poorer outcomes, including depression and possibly physical illness, under some circumstances.? Low self-esteem, ?self-blame? and stressful or hard times seemed to increase a person?s vulnerability to depression.

As we shall see however in a later section, Professor Martin Seligman, argues that parents' and teachers' attempts to bolster their children's self-esteem is contributing to the escalating numbers of young people suffering from depression. This analysis is supported by the empirical research conducted by Professor Jean Twenge.

A number of studies internationally suggest that those with low self-esteem are more likely to commit suicide or have suicidal thoughts.


Given that those with self-esteem generally cope better with stress this may have beneficial effects on their physical health. Some studies have shown that, when stressed, people with low self-esteem have higher cortisol levels (a hormone linked to stress) than those with high self-esteem. A study of policemen in Helsinki showed that those with lower self-esteem were more likely to die of a heart attack over the next 10 years. Baumeister and colleagues write that much more research is necessary but that 'it is conceivable that the benefits of high self-esteem, including feeling good, lead directly to better health.' However, they say this may be due to temperament and so it may not mean that artificially trying to increase people?s self-esteem would have a beneficial effect on health. In fact, empirical research by Professor Jennifer Crocker, summarised in another section, indicates that if people actively pursue self-esteem goals it can damage their well-being.

The Centre?s conclusions

  1. While there are many positives associated with high self-esteem the evidence suggests that raising self-esteem is certainly not a panacea or magic bullet to cure all social ills.
  2. Deliberately trying to raise the self-esteem of young people may be counterproductive.
  3. There may be some significant social costs to a society focussed on boosting self-esteem.
  4. Some of the positives for self-esteem may be related to optimism and optimism training may yield better results than working on self-esteem and be more benefical for society.
Copyright: Carol Craig, Centre for Confidence and Well-Being, 2006