ArticlePDF Available

The Impact of Lifelong Learning on Happiness and Well-being

Authors:

Ricardo Sabates

University of Cambridge

Content uploaded by Ricardo Sabates

Content may be subject to copyright.

The Impact of Lifelong Learning on Happiness and Well-being

Ricardo Sabates & Cathie Hammond

January 2008

FINAL VERSION

Dr Ricardo Sabates

Senior Research Officer and NIACE Fellow

Institute of Education

London

WC1H 0AL

R.Sabates@ioe.ac.uk

Cathie Hammond

Research Officer

Institute of Education

London

C.Hammond@ioe.ac.uk

Contents

1. Introduction............................................................................................................8

1.1. Well-being: Definitions and Indicators..........................................................8

2. Review of Empirical Literature .............................................................................9

2.1. Happiness & Life Satisfaction (subjective well-being).................................9

2.1.1 Life Satisfaction...................................................................................10

2.1.2 Happiness.............................................................................................12

2.1.3 Satisfaction with economic situation...................................................13

2.2. Self-esteem and self-efficacy (psychological well-being)...........................15

2.3. Mental disorders (Depression).....................................................................18

3. Gaps in Evidence .................................................................................................21

4. Key Messages for the Inquiry..............................................................................22

References……………………………………………………………………………23

Executive Summary

Introduction

1. This review summarises and synthesises research-based evidence concerning

the impacts of lifelong learning on happiness and wellbeing. It is a

contribution to the development of an empirically based, full understanding of

the complex mechanisms through with education impacts on happiness and

well-being, which we hope will be of use to policy makers who wish to

promote these important outcomes.

Scope

2. The review focuses on ‘well-being’, which is a multi-dimensional concept.

We limit our analysis to an exploration of subjective well-being and

psychological well-being.

3. Subjective well-being encompasses self-assessed life satisfaction and

happiness, which may apply generally or more specifically to particular

aspects of individuals’ lives, for example satisfaction with one’s job

4. Psychological well-being concerns an individual’s potential and encompasses

self-esteem and self-efficacy.

5. We also investigate the impact of learning on reducing the risk of depression,

which is one of the most common mental health problems amongst adults in

the UK.

6. A key issue in this review is whether relationships between learning and well-

being reflect effects of learning. It could be that the sort of people who do well

in education are those who tend to be happy and mentally resilient in the first

place and that attaining educational qualifications per se makes little

difference. There are no randomised control trials in this area and we cannot

make assumptions about the causalities that underlie statistical associations.

However, some studies use methodologies that provide evidence about

causality, namely the positive effects of learning on well-being.

Key Messages

7. The positive impacts of education on happiness and well-being

In general, having higher qualifications is associated with greater happiness, life

satisfaction, self-esteem, self-efficacy, and reduced risk of depression. There is some

robust evidence that having higher qualifications has positive effects on these

outcomes.

8. How education affects happiness and well-being

The positive effects of education on happiness and well-being result from a variety of

intermediary processes, which probably include higher income, non-alienating work,

household composition, health behaviours, use of health services, emotional

resilience, social capabilities and, amongst older adults, better physical health.

9. Effects of adult learning on self-esteem and self-efficacy

There is robust evidence that adult learning leads to increases in self-esteem and self-

efficacy. Adult learning has its most positive impacts on self-esteem and self-efficacy

when the learning provided meets the needs of the learner and when the learner is at a

stage in their life when they are ready and receptive to benefit from it.

10. Education may have negative effects on happiness and well-being

Associations between education and happiness and well-being are not always

positive. Adults with higher qualifications report relatively low job satisfaction, and

there is some evidence that adults with high level qualifications are not much happier

or resistant to depression than adults with qualifications at an intermediate level.

Maybe education has negative as well as positive impacts, for example through

raising aspirations and expectations that are not met and by leading to occupations

that carry high levels of stress.

12. Defining well-being

For purposes of measurement, ‘well-being’ should be defined clearly and consistently.

Otherwise, (i) it is not possible to compare results across different studies when these

studies focus on different aspects of well-being and (ii) research may mislead policy.

Specific findings

Subjective well-being

7. The effects of education depend on the measurement of subjective well-being,

e.g. life satisfaction, happiness, job satisfaction. Direct comparisons can only

be made between studies that use the same indicator of subjective well-being,

with similar scales, and also assessed by groups who do not interpret the

indicator in different ways. We can make inferences from comparing findings

from studies that draw on different groups and use slightly different measures

of well-being.

8. There is robust evidence that people with more education are both happier and

have higher levels of life satisfaction than people with lower levels of

education. The size of the association is small.

9. Most studies measure level of education by years at school or highest

qualification attained. There is also some evidence for a positive association

between participation in adult learning and increases in life satisfaction.

10. Some European studies find that people with higher education qualifications

are less happy and have lower levels of life satisfaction than those with

intermediate levels of education. A suggested explanation for this finding is

that attaining higher education qualifications in Europe raises aspirations that

prove to be unrealistic.

11. Some findings suggest that effects of education on life satisfaction are

channelled through higher income, better health, perceived trust and household

composition

12. Education appears to moderate effects of relative income on happiness and life

satisfaction. There is no consistent relationship between happiness or life

satisfaction and absolute levels of income. However, how rich a person is

relative to other people does matter for happiness and life satisfaction. Those

with relatively high incomes tend to be relatively happy and report relatively

high levels of life satisfaction. This finding applies to adults with lower levels

of education but it does not apply to those with higher levels of education.

Perhaps those with higher levels of education feel more confident about their

status irrespective of relative income, and also believe that they have the

capability to improve their economic situation if they choose to.

13. Most of the evidence suggests that individuals with higher levels of education

are relatively dissatisfied with their employment.

14. More highly educated adults tend to be relatively satisfied with their pay and

more generally, their financial situation.

15. The previous finding indicates that the reasons why more educated individuals

tend to report relatively low levels of job satisfaction are not to do with the

financial rewards of the job. Other possible explanations include stress related

to jobs at higher positions and lack of control and autonomy, especially in

positions that carry high levels of responsibility and mismatches between

expectations and employment possibilities.

Psychological well-being

16. There is consistent evidence that adults with higher levels of education tend to

enjoy relatively high levels of self-esteem and self-efficacy.

17. Participation in adult education is associated with positive changes in self-

esteem and self-efficacy.

18. Qualitative studies identify a variety of processes through which participation

in adult learning can have positive effects on self-esteem and self-efficacy.

19. Education is not beneficial to the self-esteem and self-efficacy of all

individuals. Nearly all members of society participate in education but it is

only those who attain high levels of qualifications who have relatively high

levels of self-esteem and self-efficacy; the rest have relatively low levels. In

addition, qualitative studies find that participation in adult learning can

undermine the self-esteem and self-efficacy of individuals who feel that they

have not succeeded. Participation involves risk taking and when individuals

are not prepared for the learning experience or when the content, pedagogical

style or peer group do not suit the learner, then self-confidence, self-esteem

and other aspects of psychological well-being can be negatively affected.

Mental disorder (depression)

20. A few sound empirical studies find that higher educational attainment is both

associated with a lower prevalence of depression and can actually reduce the

risk of depression.

21. Having intermediate as opposed to low levels of qualifications affords

protection from the risk of depression, but having high-level qualifications

affords no protection over and above that associated with having intermediate

level qualifications.

22. The previous finding parallels findings from some studies that happiness and

life satisfaction are positively associated with education only up to an

intermediate level of education (see point 10). Although success in higher

education almost certainly has a range of outcomes that have positive impacts

on happiness and mental health, it appears also to be potentially undermining,

possibly through raising expectations and aspirations to levels that are difficult

to meet and through leading to stressful lifestyles and occupations.

23. Channels through which education protects individuals from the onset and

progression of depression include financial/economic situation, non-alienating

work, emotional resilience, health behaviours, use of health services, social

responsibility, social capabilities, and, amongst older adults, physical health.

Gaps in the Literature

24. In order to inform policies designed to improve levels of happiness and well-

being, we need to fully understand the mechanisms through which education

impacts on these valuable outcomes. Of prime importance is the need for more

robust empirical studies that provide evidence about effects of education, over

and above mere associations.

25. A variety of studies provide clues to the mechanisms through which education

impacts on happiness and well-being. These include studies which find

negative or null associations between level of education and job satisfaction,

studies which find negative or null associations between higher as opposed to

intermediate levels of education and happiness and well-being and other

studies that seek to identify the channels through which education affects

happiness and well-being. Thorough comparisons of these existing studies and

more quantitative studies of this kind, coupled with in-depth qualitative

studies, all in a range of contexts, will enhance our understanding of the roles

that education can play in promoting happiness and well-being.

26. There is little research on the impacts of highest educational qualifications and

continuing adult learning on satisfaction with specific aspects of life, for

example, marriage, social support, involvement with the community and

housing.

27. There is very little research on the impact of adult learning on happiness or

well-being. Are those who engage in learning in adulthood happier or more

satisfied with their lives than those who do not engage in learning?

28. Furthermore, little is known about the impact of learning on individuals’ well-

being over time. In other words, does learning over the life course increase

happiness or life satisfaction?

29. The evidence about relationships between education, well-being and income is

contradictory. Absolute level of income is not associated with well-being but

relative income is. Two studies provide evidence that education moderates this

association. The first suggests that the well-being of adults with higher levels

of education is relatively resilient to relative income, whilst the second

suggests that relative income matters more for the well-being of adults with

high level qualifications than it does for adults with lower level qualifications.

More research is needed in this area.

30. What do we mean by well-being? There is a need to bring together research on

well-being from different disciplines.

31. Is the impact of learning on psychological well-being short term or does it

have long-lasting effects?

32. Is lifelong learning one of the important channels for the impact of highest

educational qualifications on reducing the risk of depression?

1. Introduction

This report 1) summarises findings from a review of the evidence about the impact of

lifelong learning on happiness and well-being; 2) points out the gaps in the literature

and; 3) provides some key messages for NIACE’s Commission of Inquiry.

In order to undertake this review, we identified the main indicators of well-being. For

each of these indicators, we reviewed empirical studies where education or learning is

used at least as an explanatory factor if not the main focus of the study. We

summarise the main findings and address the following areas of interest for the

Commission:

• What does evidence tell us about the impact of lifelong learning on happiness

and well-being?

• In particular, what do we know about the impact of lifelong learning on mental

health, satisfaction with life, and self-efficacy?

• What are the gaps in evidence?

• What are the key messages for the Inquiry?

We have not conducted a systematic review of the literature; this would not have been

feasible within the time available. Rather, we relied on previous work undertaken by

researchers from the Centre of Research on the Wider Benefits of Learning (Feinstein,

et al. 2006), which we updated with more recent work on education and well-being.

We conducted searches in Scholar Google as well as scientific publishers such as

Ingenta, Elsevier Science Direct and JSTOR. As a search criterion we focused

exclusively on articles published within the last five years (from 2002 onwards).

1.1. Well-being: Definitions and Indicators

There are two main domains of research traditions on well-being (Keyes, Shmotkin,

and Ryff, 2002). One tradition, referred to as subjective well-being, deals with

happiness and investigates factors related to self-assessed life satisfaction in general

or applied to specific areas of one’s life. The other tradition, referred to as

psychological well-being, focuses on human potential, which reflects personal

growth, sense of control over one’s actions and purpose in life.

Subjective well-being emerged in the late 1950s in the search for useful indicators of

quality of life to monitor social change and improve social policy. The two main

indicators of subjective well-being that emerged from this research were life

satisfaction and happiness. Life satisfaction reflects individuals’ perceived distance

from their aspirations (Campbell, et al 1976). Happiness results from a balance

between positive and negative affect (Bradburn, 1969). Although these indicators are

different, they are used interchangeably in the empirical literature.

Conceptualisations of psychological well-being draw on formulations of human

development in the context of external life challenges. On the basis of these

formulations, Ryff (1989) suggested seven dimensions of psychological well-being,

differentiated by the different challenges that individuals encounter as they strive to

function positively. The dimensions were: self-acceptance, ability to shape the

environment, personal autonomy, self-determination, purpose in life, personal growth,

and positive relations with others.

In this review we focus on happiness and life satisfaction as indicators of subjective

well-being. For psychological well-being we focus on just two of the dimensions

proposed by Ryff (1989). The first indicator is self-esteem, defined as self-acceptance

and a basic feeling of self-worth (Rosenberg, 1965). The second is the ability to shape

one’s own environment (Ryff, 1989). This is closely related to Bandura’s (1994)

notion of self-efficacy, which he describes as an individuals’ beliefs about his or her

capabilities to produce designated levels of performance that exercise influence over

events that affect their lives.

In their broadest definitions, subjective well-being and psychological well-being

encompass many aspects of good mental health. According to the World Health

Organisation, positive mental health incorporates the state of well-being in which the

individual realises his or her own abilities, can cope with the normal stress of life, can

work productively and is able to make contributions to the community (WHO, 2007).

Mental health is not only the absence of disease or infirmity, but includes physical,

mental and social well-being.

But these definitions of well-being do not include mental disorders, such as

depression or anxiety, which affect mainly the adult population and for which

learning can have important benefits. According to Butler, et al. (2004) depression is

the third most common reason for consultation in general practice in the UK. It is

estimated that the prevalence of major depression among 16 to 65 year olds in the UK

to be 17/1000 for males and 25/1000 for females. For a broader category of mixed

depression and anxiety the prevalence is estimated to be 71/1000 for males and

124/1000 for females. Therefore, we also focus on depression as an indicator of a

mental disorder.

2. Review of Empirical Literature

We start by reviewing evidence on the impact of learning on subjective well-being

(Section 2.1), followed by psychological well-being (section 2.2), and finally on

reducing the risk of depression (Section 2.3).

2.1. Happiness & Life Satisfaction (subjective well-being)

What are the measures for subjective well-being? Studies in this area usually gather

information from answers to satisfaction questions and use them as measures of

subjective well-being. Commonly these answers are on a discrete scale, ranging from

unsatisfied to satisfied, with zero at one end and anything between 5 and 11 at the

other. One measurement of subjective well-being records individuals’ satisfaction

with life as a whole. Others focus on specific domains, like leisure, work, finances,

health, housing, family, and so on. Sometimes questions ask about feelings of

happiness whilst others ask about satisfaction. The relationships found between

education and subjective well-being depend on the measure used.

Empirical studies using data from different countries have shown a low positive

association between education and life satisfaction (Veenhoven, 1996; Caporale, et al.

2007) and between education and satisfaction with financial situation (Seghieri,

Desantis and Tanturri, 2006) for most countries. There is also a positive association

between education and happiness, which is robust to the inclusion of background

family factors (Hartog and Oosterbeek, 1998; Gerdtham and Johannesson, 2001).

Studies from Britain and the United States found a negative correlation between

education and job satisfaction, indicating dissatisfaction among individuals with

higher levels of education (Clark and Oswald, 1996; Ross and van Willigen, 1997).

This dissatisfaction may be due to the lack of jobs at higher levels, the stress related to

jobs at higher positions, and to mismatches between aspiration and expectations with

employment possibilities.

In the next sections, we review empirical evidence on the impact of education on

different indicators of subjective well-being: overall life satisfaction, happiness, and

satisfaction with economic situation (which includes job satisfaction, financial

satisfaction and pay satisfaction).

2.1.1 Life Satisfaction

There is evidence for both positive and negative correlations between education and

life satisfaction. We consider this evidence and also evidence concerning the channels

through which education might affect life satisfaction.

Helliwell (2002) estimated a multivariate regression using individual level data from

the World Values Survey for 46 countries. Results showed that when overall life

satisfaction was regressed on education (measured as the age at which individuals

completed full-time education), there was a strong, statistically significant, positive

association. However, when other individual and national variables were included in

the model, the association disappeared. Similar results were found by Ferrer-i-

Carbonell (2005) for Germany. The author used the German Socioeconomic Panel

dataset and modelled the determinants of life satisfaction. The impact of years of

education on life satisfaction was not statistically significant in a model that included

income, relative income, and other demographic and socioeconomic variables. These

results suggested that educational effects may be channelled via higher income, better

health, higher perceived trust and household composition, which were controls

included in the models.

Another positive association between education and life satisfaction was found by

Oswald and Powdthavee (2007) using the British Household Panel Survey.

Individuals with A-levels and higher degrees reported higher levels of life satisfaction

than individuals with lower levels of educational qualifications. In the United States,

Oreopoulos (2003) found that years of schooling were positively associated with life

satisfaction, and that this result held up after using features of compulsory schooling

laws as an instrumental variable for schooling to address the possibility of reverse

causation (that is, that having greater life satisfaction in the first place may cause

people to be more likely to stay longer at school).

In contrast, Caporale, et al. (2007) found a negative association between having

higher education qualifications and life satisfaction using data from the European

Social Survey. The authors suggest that having a higher education qualification could

raise aspirations that are not often fulfilled, leading to dissatisfaction and the negative

association found. When reference income (income of individuals with similar

characteristics) was included in the analysis, the association between having a higher

education qualification and life satisfaction became positive. A possible explanation is

that the higher education qualification leads to aspirations about income, which are

not often met. Once these are taken into account (i.e. included in the analysis), the

positive impacts on life satisfaction of having a higher education qualification prevail.

Frey and Stutzer (2002) analysed data from Switzerland gathered in 1992-1994 and

estimated that achieving middle and high levels of formal education increased life

satisfaction by 2.19 and 2.09 percentage points, respectively. However, when

aspirations were included in a later analysis, Stutzer (2004) found a statistically

significant difference in life satisfaction only between those with mid-level

educational qualifications and those with low-level educational qualifications. He did

not find evidence of differences in life satisfaction between those with low-level

qualifications and those with higher education, which could be explained by

aspirations raised in unrealistic ways as a result of success in higher education.

In contrast to these findings that suggest negative as well as positive impacts of

attaining higher education qualifications, Kingdon and Knight (2004) found a small

positive association between overall life satisfaction and education but only at higher

levels of education (achieving a degree). This study was conducted in South Africa.

Possibly, attaining a higher education qualification in South Africa is associated with

aspirations that are different from those which result from attaining a similar level of

qualification in Europe. Also, the economic situation for those with these

qualifications is probably different in South Africa and Europe.

Ferrante (2007) investigated the channels through which education affects life

satisfaction, hypothesising that the main channels were via (i) access to stimulating

employment and consumption activities matching preferences and skills and (ii) how

the rewards of such activities compare with expectations. Results using data from the

2004 Italian Survey of Household Income and Wealth showed that education may

exert a negative effect on life satisfaction through raising individuals’ aspirations and

expectations above real life chances. This empirical result confirmed earlier

theoretical suggestions by Ross and van Willigen (1997) about the mechanisms

through which education may negatively impact on job satisfaction.

It is also possible that learning begets learning and this strengthens the relationship

with life satisfaction. Feinstein and Hammond (2004) used the 1958 cohort to

examine the contribution of adult learning to changes in life satisfaction between the

ages of 33 and 42 years, controlling for initial level of education and many other

factors. Their results showed that participation in adult learning had positive effects

on changes in life satisfaction. Effect sizes were small in absolute terms. However,

for most cohort members, there is little change in life satisfaction during mid

adulthood, and relative to this baseline, participation in adult learning is an important

driver for change in life satisfaction.

The work of Ferrer-i-Carbonell (2005) and Helliwell (2002) described above suggests

that channels for the positive impacts of education on life satisfaction are health, trust,

household composition and income. Research on the impacts of income on happiness

indicates that relationships are complex. Layard (2003) reports that the proportions of

people who are very happy in Western countries has not changed at all even though

the level of real income in each group, especially at the top end of the distribution, has

risen considerably. This phenomenon is known as the Easterlin Paradox (Easterlin,

1995), after the seminal work of Easterlin (1974) on income and happiness. Easterlin

pointed out that while richer individuals in a country are happier than their poorer

fellows, income increases in themselves do not lead to increases in well-being of the

same magnitude. This suggests that it is relative rather than absolute levels of income

that are important for life satisfaction.

The impacts of income on happiness are moderated by level of education. Income

matters for happiness more for those with lower levels of education than it does for

those with higher levels of education. Castriota (2006) analysed the effect of absolute

income on well-being by education level using data from the World Bank’s World

Value Survey. His results showed that the higher the education level, the less relevant

the absolute income level for predicting self-reported life satisfaction. Higher income

makes everybody happier but, everything else being equal, the marginal utility of

additional income is higher for less educated people. A possible explanation is that

highly educated persons have on average more interesting jobs and more active and

stimulating cultural lives. Consequently, the quantity of material goods a person can

buy becomes less important. This result contradicts that found by Caporale, et al.

(2007) mentioned above. Maybe this is because Caporale used European data whilst

Castriota used data from the World Bank.

2.1.2 Happiness

In Sweden, Gerdtham and Johannesson (2001) found a direct positive association

between education and the probability of being happy; individuals with the highest

educational qualifications were most likely to report the highest levels of happiness.

In contrast, Caporale, et al. (2007) found a negative association between education

and happiness using data from the European Social Survey. Compared with

individuals without educational qualifications, higher levels of educational

qualifications were associated with higher levels of unhappiness, controlling for

socioeconomic and demographic factors including absolute income. However, when

relative income was included in the analysis, the authors found contrasting results.

This suggests that for individuals with more education, their level of happiness is

particularly dependent on relative income. They differ from individuals with less

education in that being relatively rich or poor makes more difference to their level of

happiness. Perhaps this is because having high-level qualifications raises expectations

about income, which are not always met.

Easterlin (2003) also found a positive association between education and happiness,

but only at one point in time and not over the life course. His results showed that at

any given age, individuals with higher levels of education are happier than those with

lower levels of education. However, over the life course, there is no significant trend

in happiness for those with higher levels of education or those with lower levels of

education and there is no evidence to support a happiness differential by educational

status. Although those fortunate enough to start out with higher income and education

remain, on average, happier throughout the life cycle than those of lower

socioeconomic status, there is no evidence for either group that happiness increases

with income (although there is no empirical evidence linking happiness and lifelong

learning over the life course).

Hartog and Oosterbeek (1998) used data from a cohort of adults born in 1940 in

Holland in the province of Noord-Brabant to investigate the relationship between

education and happiness, controlling for wealth and health and early life

circumstances. Their results indicated that the positive relationship between

education and happiness reached a maximum for intermediate level of qualifications.

The relationship between education and happiness was parabolic and it remained

statistically significant only for intermediate qualifications and for a higher vocational

qualification.

2.1.3 Satisfaction with economic situation

We examine research on the relationships between education and three measures of

economic situation; job satisfaction, pay satisfaction and financial situation. Negative

associations are found between level of education and job satisfaction, whereas the

associations between education and pay satisfaction, and between education and

satisfaction with financial situation are positive. One reason why more educated

individuals report lower levels of satisfaction with their jobs may be that education

raises job aspirations, which are not met. These aspirations are not simply about pay,

for the argument does not apply to satisfaction with pay or financial situation. More

likely, they concern autonomy and control within work and the opportunity to work

creatively.

Clark and Oswald (1996) found a strong negative association between job satisfaction

and education levels using data from the British Household Panel Survey. Similarly,

Ross and van Willigen (1997) report that education has a negative association with

job satisfaction in the US. Results using two nationally representative datasets showed

that among people in the same work situation and the same income bracket, those

with higher levels of education reported more dissatisfaction with their work than

those with lower levels of education. The authors suggested that the impact of

education on job dissatisfaction may be due to higher expectations. People are

generally satisfied with conditions that are the best they can expect. Thus, all else

being equal, those who expect more are more dissatisfied and individuals with higher

levels of education have higher expectations. Another reason why more educated

individuals tend to report lower levels of satisfaction with their jobs could be that they

are more aware of their dissatisfaction and more articulate in voicing it, as opposed to

actually being more dissatisfied with their jobs than individuals with lower levels of

education.

In contrast, Balchflower and Oswald (2007) did not find statistically significant

gradients between job satisfaction and education for Australians. No associations

were found between education and stress at work or levels of tiredness after returning

home from work. This may be because Australians, whatever their level of education,

have very low levels of job satisfaction. In the same paper, Balchflower and Oswald

report that using data on approximately 50,000 randomly sampled individuals from 35

nations, Australians had some of the lowest levels of job satisfaction in the world.

In the paper cited above, Clark and Oswald (1996) found a positive association

between pay satisfaction and education. This association disappears when income is

controlled for in the regression, suggesting that educational effects on pay satisfaction

are strongly mediated by income.

A positive association between education and satisfaction with financial situation was

found by Seghieri, Desantes and Tanturri (2006) for Spain, Greece, Portugal, Italy,

France, Ireland and Belgium using the European Household Panel Survey. They did

not find evidence for this association in Denmark and they found a negative

association in the Netherlands.

Summary

• The effects of education depend on the how subjective well-being is measured,

e.g. life satisfaction, happiness, job satisfaction. They also depend on context,

and hence the country in which the data were collected. Hence, comparisons

can only be made when using the same indicator of subjective well-being

(with similar scale and comparability between different groups who may

interpret these indicators differently).

• There is a positive association between levels of education and life

satisfaction, but this association is small and tend to disappear when controls

for background family factors are included in the analysis.

• One of the reasons why the effects of initial education are small (or

statistically insignificant) is because education can have both positive and

negative impacts on life satisfaction. It can increase employment possibilities

and promote further training opportunities but it can also raise aspirations

above unreachable or unattainable levels.

• There is robust evidence indicating that individuals with higher levels of

education are happier than those with lower levels of education.

• However, there is no evidence to show that happiness increases with income

over the life course (although there is no empirical evidence linking happiness

and lifelong learning over the life cycle).

• Most evidence shows that individuals with higher levels of education are more

dissatisfied with their employment. However, those with higher levels of

education are more satisfied with their pay and financial situation.

• The impact of education on job dissatisfaction may be due to lack of jobs at

higher levels, the stress related to jobs at higher positions, or mismatches

between aspiration and expectations with employment possibilities.

• The mixed results reflect differences in measurement and context as

mentioned in the first bullet point. They also depend on which controls are

included in models. More important theoretically, it is likely that education

generates a range of outcomes, some of which have positive and some of

which have negative effects on life satisfaction. These include employment

possibilities and further learning and positive consumption activities, which

would increase life satisfaction, and also raised aspirations which may,

unfortunately, be unrealistic and lead to disillusionment and lower levels of

satisfaction.

2.2. Self-esteem and self-efficacy (psychological well-being)

In this section we describe evidence from quantitative studies that assess the

relationships between initial education and levels of self-esteem and self-efficacy,

quantitative and qualitative studies investigating the impacts of adult learning on these

outcomes, evaluations of specific learning programmes and a survey of learning

practitioners.

There is consistent evidence that adults with higher levels of education tend to also

enjoy relatively high levels of self-efficacy and optimism. For example, correlations

have been found between years of education and self-efficacy, self-esteem, optimism

and happiness amongst residents of the US aged between 70 and 79 (Kubzansky et al.,

1998) and pregnant women living in California (Rini et al., 1999). Hammond and

Feinstein (2006) analysed data from the British cohort study of individuals born in

1958 and found that after taking social and psychological background factors into

account, cohort members who had flourished at secondary school (during the late 60s

and early 70s) had, at age 33, relatively high levels of optimism and self-efficacy.

Interestingly, both educational attainment and engagement at school were important

aspects of school flourishing for these positive outcomes in adulthood.

Feinstein and Hammond (2004) and Hammond and Feinstein (2006) analysed the

1958 cohort study data and found correlations between participation in adult learning

and positive changes in efficacy and optimism between the ages of 33 and 42, after

controlling for family, social and educational background, and current life

circumstances. They were found for both men and women, regardless of their levels

of education at the beginning of the course. Hammond and Feinstein (2005)

supplemented the findings for self-efficacy with in-depth interviews with members of

the same cohort study who had left school with few, if any, qualifications. This small-

scale qualitative study found that taking courses in adulthood can lead to

improvements in self-efficacy for people who did not attain many qualifications at

school.

Dench and Regan (1999) used a combination of quantitative and qualitative methods

to investigate the impacts of participation in learning for adults aged between 50 and

71 living in England and Wales. Respondents reported that learning had led to

increases in their self-confidence, their enjoyment and satisfaction with life, positive

changes in how they felt about themselves and their ability to cope with everyday life.

Similar findings are reported by Schuller and colleagues, who conducted a large-scale

in-depth qualitative investigation of the impacts of adult learning (Schuller et al.,

2002; Schuller et al., 2004). This involved 145 biographical interviews with adults

who were currently participating in various types of adult education, together with

group interviews with practitioners providing adult education. One of the authors’

conclusions is that:

The most fundamental and pervasive benefit from learning of every kind is a

growth in self-confidence (Schuller et al., 2002: 14)

The study found that participation in a range of adult learning programmes could give

adults the confidence to take on more active social roles, to try out new things, and to

tackle issues rather than brush them under the carpet. It empowered some learners to

take additional courses and apply for jobs and to visit places that they would not

otherwise have visited, such as art galleries, museums, libraries, and to travel abroad.

Learning led to improvements in self-esteem, self-understanding, a clearer sense of

identity, the capacity to think independently, a sense of purpose and hope, improved

competences and communication and better social integration (Hammond, 2004). All

of these are important aspects of psychological well-being. Respondents of both

genders, all ages, every ethnic background interviewed, every occupational class, all

levels of previous education and living in families or households of every kind,

mentioned that they had experienced increased self-esteem (self-acceptance or self-

worth) as an outcome of learning at some point during their life. Respondents also

mentioned as outcomes of learning self-understanding, doing something for oneself,

purpose and hope (Hammond, 2004).

Numerous evaluations of educational initiatives provide evidence that outcomes

include self-esteem, self-efficacy and self-understanding for different types of courses

and for different groups of people. These include evaluations of courses in Higher

Education in England taken by mature women (Cox and Pascall, 1994), adults

participating in Higher Education and Access courses in England (West, 1995), adults

in England who were returning to education (Hull, 1998), older adults receiving

mentoring support on a psychosocial support programme in the US (Koberg et al.,

1998), and courses offered at various levels to adults with chronic health problems

and/or employment difficulties living in England (McGivney, 1997).

Wertheimer (1997) reviewed evaluative studies of community-based adult education

courses that were attended by mental health service users. She also conducted a

survey of over 30 such courses using questionnaires to investigate the experience of

participants. She concludes that for mental health service users, participation in these

community-based courses led to improvements in confidence, self-esteem, self-

efficacy and mental health. More recent qualitative studies indicate that participation

in education has positive outcomes for mental health service users (e.g., Westwood,

2003).

Theodorakou and Zervas (2003) found that physical education had impacts on

children’s self-esteem. They examined the influences of two physical education

teaching methods on children’s self-esteem in a publicly funded school in Athens.

Both methods were associated with increases in self-esteem but the more child-

centred approach, which used group discussion and creative techniques, was

associated with increases in all aspects of self-esteem – cognitive, physical and social,

as well as global – whereas the more traditional teacher-directed methods of teaching

physical education were associated with increased global self-esteem, but more

particularly with the physical as opposed to the social aspects.

Preston and Hammond (2002) investigated the views of lecturers and managers in

Further Education colleges in England and Wales about the impacts of participation

on their students (n>2,700). The most commonly reported outcomes of participation

were self-esteem and self-efficacy. Analysis of answers to open-ended questions

indicated that practitioners believed that amongst their students, increased self-

efficacy led to increased propensity to participate in the community and

improvements in psychological health. These results are important in understanding

the possibility that education can lead to improvements in self-efficacy, but would be

given more strength if responses were also obtained from learners as well as from

learning providers.

Bandura (1997) reviews the evidence to examine the sources of self-efficacy. We

suggest that these sources are likely to be affected by education. One source of self-

efficacy is vicarious experience, which refers both to learning from the competence of

others (e.g. teachers and peers) and social comparison. Social comparison is difficult

to avoid in educational settings because students are aware of each others’ attainment

and ability. Students who are more successful at school may engage more with, and

gain more self-efficacy from, role models provided by other successful students and

teachers, and they may also gain self-efficacy because social comparisons are

positive, at least in relation to attainment.

Another source of self-efficacy is verbal persuasion, when significant others express

faith in one’s abilities rather than convey doubts. This source of self-efficacy is also

relevant to education because students receive from their teachers and peers explicit

as well as implicit feedback on their performance and abilities. Bandura suggests that

verbal persuasion has more impact when it is within realistic bounds and on people

who already believe that they can produce effects through their actions (Chambliss &

Murray, 1979a, 1979b), so we predict that teachers’ feedback to pupils is particularly

likely to impact on their self-efficacy.

The arguments and studies described above present a glowing impression of the

potential impacts of the educational experience on self-esteem and self-efficacy. It is

important to remember, however, that education also has the potential to undermine

self-esteem and self-efficacy and create confusion. In a study described earlier in this

section, Hammond and Feinstein (2006) found that adults who had flourished at

secondary school had relatively high levels of efficacy, life satisfaction and optimism.

However, every member of this cohort should have attended secondary school and

according to the definition used in the research, about half of this group did not

flourish there, and later in life experienced relatively low levels of efficacy, life

satisfaction and optimism.

Similarly, not all evaluations of adult education programmes report positive impacts

of participation. For example, Randle (2003) found that amongst students on a

diploma course in nursing, self-esteem decreased dramatically between the start and

the end of the course. The qualitative part of the study indicated that during the 3

years, students felt increasingly powerless to be the sort of nurse they wished to be.

Participation in particular educational streams also appears to have consequences for

self-esteem and they are positive for some streams and negative for others. Houtte

(2005) examined the consequences for self-esteem of being in a technical / vocational

secondary school as opposed to a general secondary school in Belgium. Houtte found

that boys in technical / vocational schools have lower self-esteem than boys in general

schools, but for girls there was no difference with school type.

So which aspects of education are important if we wish to increase levels of self-

efficacy and self-esteem? In the large-scale qualitative study described above,

Schuller et al. (2002, 2004) found also that participation in adult learning programmes

has the potential both to promote and to undermine psychological well-being. This is

because participation involves risk taking; in order to learn, the individual must be

prepared to admit a degree of ignorance and adopt new aspects of knowledge and

perspective or try out new skills. If the learner feels successful in their endeavours or

if they feel that they have benefited from the experience, this will make them feel

more confident in themselves and more confident about taking risks. Consequently, it

will build their sense of self-esteem and self-efficacy. It may also broaden their

horizons so that they understand themselves in a different context, invest more in their

future and change their hopes and aspirations. Lack of success, on the other hand, can

undermine self-esteem and aspirations and lead to alienation.

The content of what is learnt, the pedagogical style and who one learns with are

aspects of learning that influence psychological well-being (Schuller et al., 2004). For

example, self-esteem and self-efficacy increase as learners are praised or receive

formal feedback or accreditation for succeeding in tasks which they perceive as

challenging (Schuller et al., 2004). Courses in the social sciences taught through

discussion with students from diverse backgrounds promote self-understanding and

independent thinking and can lead to changed hopes and aspirations (Preston and

Hammond, 2002). These can in turn increase overall life satisfaction.

Summary

• There is consistent evidence (at the level of associations) that adults with

higher levels of education tend to also enjoy relatively high levels of self-

efficacy and optimism.

• In particular, taking courses in adulthood can lead to improvements in self-

efficacy for people who did not attain high or indeed any qualifications at

school.

• Participation in learning can also increase self-confidence and self-esteem.

This has been documented by several qualitative evaluations of adult learning

programmes.

• However, learning has also the potential to undermine psychological well-

being. This is because participation involves risk taking and when individuals

are not prepared for the learning experience or when the content of what is

learnt, the pedagogical style and who one learns with are inadequate, then self-

confidence, self-esteem and other aspects of psychological well-being can be

affected.

2.3. Mental disorders (Depression)

Within the scope of this paper, we cannot examine all mental health disorders but

rather focus on depression. We choose depression for three reasons. Firstly,

depression is the most common psychiatric disorder (Miroswki and Ross, 2002). It is

the third most common reason for consultation in general practice in the UK (Butler,

et al. 2004) and its estimated prevalence in the UK amongst 16 to 65 year olds is

17/1000 for males and 25/1000 for females. Secondly, various studies have shown

that societal factors contribute to the onset and progression of depression (e.g. Brown

REF) and so it is possible that educational experiences are important. Related to this

is the third reason, which is that there is more research concerning the societal causes

of depression than there is for most other mental health disorders.

It is well established that the prevalence of depression is greatly influenced by a

number of socioeconomic factors, which interact in a complex way. Episodes of

depression are strongly associated with adverse social and economic circumstances,

such as unemployment, divorce or separation, inadequate housing, lack of educational

qualifications and poverty. There are complex interactions between these factors and

an important role for lifelong learning in protecting individuals from the risk of

depression. In this review, we first establish whether education can help to protect

individuals from depression and then examine the main mechanisms through which

education affects depression.

Is education causal?

Various studies have found positive correlations between more education and lower

rates of depression. For example, Mirowsky and Ross (2002) investigated the role of

education as a protective factor against depression in the context of entry age of

parenthood, controlling for a large set of background characteristics. Using the U.S.

1995 Survey of Aging, Status and the Sense of Control, they found that years of

schooling were associated with a 6% decrease in the symptoms of depression. When

other socio-economic variables and physical health were introduced as controls in the

analysis, the estimated coefficient was reduced to 2.3%, but remained statistically

significant.

Feinstein (2002), using data from the 1958 and 1979 British cohorts and matching

methods, showed that taking childhood abilities, health and family background factors

into account, women from the 1958 cohort with qualifications at U.K. National

Vocational Qualification at Level 1 – which is roughly equivalent to lower secondary

education – had 6 percentage points lower likelihood of depression than women with

no qualifications. For women in the 1970’s cohort the equivalent estimated effect was

10 percentage points. For men these effects were smaller. In general, results showed

that differences between those with different levels of qualification but with

qualifications above Level 1 were substantially eroded when background factors and

previous childhood controls were introduced in the analysis.

Chevalier and Feinstein (2006) used on a rich longitudinal dataset to control for

childhood determinants and measure mental health over the individuals life span to

estimate the causal effect of education on depression. They found that achieving

qualifications significantly reduced the risks of adult depression. The effect was non-

linear and was greater for differences between lower educational levels. Estimates

using two-stage least squares were much larger but in most cases, it was not possible

to reject the exogeneity of education. Using propensity score matching, they

estimated that having level 2 or higher qualifications reduced the risk of adult

depression by 6 percentage points. This effect was similar for men and women.

What are the mechanisms for educational effects?

Having established above that education plays a part in reducing the risk of

depression, we now investigate the mechanisms through which this might take place.

Hammond (2002) found clear evidence that highest qualifications achieved can

protect against that onset and progression of depression and suggested that the impact

of education on depression was mainly mediated by economic factors, the adoption of

health behaviours, the development and maintenance of resilience, access to and take

up of health services, and the promotion of social responsibility, social values and

social skills.

Miech and Shanahan (2000) looked at the relationship between education and

depression over the life course. Using data from the 1990 Work, Family and Well-

Being Study in the U.S., they found that the association between education and

depression increases with age, and that in older age, this association is (to some

extent) mediated by physical health. Amongst older individuals, higher levels of

education are associated with having better physical and mental health.

Other mechanisms for the effect of education on mental health were investigated by

Ross and Van Willigen (1997). They hypothesised that education is valuable for

individuals’ mental health because it provides access to two primary sources of well-

being: non-alienated paid work and supportive relationships. Compared to individuals

with low levels of education, those with high levels of education have access to non-

alienated paid work that increases the sense of personal control. Also, those with high

levels of education have access to stable social relationships, especially marriage, that

increase social support. Results using two nationally representative datasets from the

United States showed that individuals with high levels of education have lower levels

of emotional distress (including depression, anxiety, and anger) and physical distress

(including aches and pains and malaise). In terms of their hypotheses, they found that

education reduces distress largely by way of paid work, non-alienated work, and

economic resources, which are associated with high personal control. However, the

extent to which education reduces distress by way of marriage and social support is

much more modest.

Summary

• Sound empirical studies find robust evidence that higher educational

attainment can reduce the risk of depression. The impact of education is more

important amongst those with low to intermediate levels of education.

• For higher levels of education, the effect of education on depression may be

both positive and negative.

• For example, a higher occupational grade is associated with more control over

working lives, more varied and challenging work and thus has a positive effect

on mental health (Marmot et al., 1991). However, higher occupational

attainment also leads to higher levels of stress (Rose, 2001).

• Therefore, there may be important trade-offs between stress and satisfaction

that may lead to a complex and non-linear relationship between educational

success and risk of depression.

3. Gaps in Evidence

33. In order to inform policies designed to improve levels of happiness and well-

being, we need to fully understand the mechanisms through which education

impacts on these valuable outcomes. Of prime importance is the need for more

robust empirical studies that provide evidence about effects of education, over

and above mere associations.

34. A variety of studies provide clues to the mechanisms through which education