RESEARCH: Self-esteem And The Pursuit of Happiness

Note: For all diagrams and tables please see full research article.

Abstract

A large body of cross sectional and longitudinal research links strengths use to areas of sustained well-being and happiness. Related theory suggests that an individual’s use of strengths is an important predictor of self-esteem, whilst self-esteem in turn predicts unique variance in subjective wellbeing. This manuscript explores support and central reasons for the strongly supported concept that people who use their strengths find greater levels of happiness, or subjective wellbeing. The present study accesses a broad socio-economic cross section of participants in providing new information as to a possible moderating or mediating role of self-esteem within the relationship between these constructs. This analysis expands upon the question of whether or not strengths use relates to subjective well-being and looks in detail at an important aspect of ‘how’ and ‘why’ this relationship might exist. Results indicate that the predictive nature of strengths use upon happiness can be explained to a meaningful degree by increased levels in self-esteem owed to a robust mediating role. The discussion localises findings in the context of both clinical and counselling psychology, alongside consideration of political and social recognition awarded of late to the prominence of happiness studies. This research highlights the importance of self-esteem within the dynamic of positive constructs and seeks to enable health practitioners and psychologists from various fields to better devise valuable interventions in promotion of human wellbeing.

 1.   Introduction

 The present analysis begins by considering why much research today contemplates subjective well-being and how the examination of happiness has evolved within psychology over time. This manuscript then analyses how it is most frequently measured and studied today, in the midst of growing social interest and importance in this paradigm. Leading on from subjective well-being, the origins of the concept of strengths use is discussed in reference to an established body of research which highlights the predictive ability of strengths use upon subjective well-being. It then examines how strengths use has been studied and measured with emphasis upon the difference between strengths knowledge versus their daily use. The position of self-esteem within this relationship is then addressed, alongside background to the often cited ambiguity over the general situation and role of this construct. The main paradoxes of self-esteem theory are considered in relation to examination of positive concepts such as those herein. The introduction subsequently extends to highlight important findings which suggest the location of self-esteem as having influence between these two dynamics either within a moderating or mediating role, before explanation of methodology and results.

The importance of subjective well-being

The original focus of psychology was not only upon the curing of dysfunction but also attended to the promotion of optimum human functioning (Wood & Tarrier, 2010). In more recent history however, the main concern of psychology has been the diagnosis and treatment of psychological illness, with a strong focus on the measurement and treatment of psychopathology. The post World War II years saw scant funding for research other than that relating to treatment of injury and illness, a trend which precipitated a more symptoms lead direction in both research and treatment for the proceeding 50 years. Indeed the current model dominating clinical practice as characterized in the DSM-V (American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders,) defines mental health as an absence of psychopathology. This approach, which largely overlooks the active promotion of mental wellbeing is often cited in literature as the disease approach to human functioning (Keyes, 2005). In reference to this, in the late 90s positive psychology, in developing onwards from the Humanistic tradition began to address 3 main concerns. Firstly, that there was little focus or research carried out regarding positive experiences such as subjective well-being (subjective well-being). Secondly, that there was a dearth of interest in examining personality traits and in particular the character strengths of those individuals who appeared to thrive. Thirdly, on a social level the study, identification and enhancement of the qualities of institutions which enable positive subjective experiences and resilience within individuals (Seligman & Czikszentmihalyi, 2000). In answer to these concerns, the study of subjective wellbeing in various spheres of psychology has made impressive and enduring empirical strides over the last 20 years. This recent drive towards the study and promotion of happiness by psychologists, sociologists and economists, represents more than a simple extension of the ancient Aristotelian search for eudaimonia, or ‘the good life’. Since the advent of positive psychology’s focus upon enhancement of wellbeing, the concept that well-being should be measured and studied alongside psychopathology has been repeatedly promoted (Greenspoon & Sklofske, 2001; Keyes, 2007). Far from a mere reformation of ancient philosophising applied to modern day life, those who analyse subjective well-being today often argue that full assessment of mental health should include assessment of subjective well-being. These endeavours are fuelled by the well documented concept that subjective well-being influences both physical and mental health (Deiner & Chan, 2011; Proctor, Maltby, Linley, 2010). Combined with this is the worthy assertion that as mental illness becomes more widespread within society, so the need for preventative measures via the development of mental wellbeing becomes of greater importance to psychology and to society (Seligman, Steen, Park & Peterson (2005). Consequently, the pursuit of happiness, or of subjective wellbeing and the study of its causes and contributors, is of wider impact than the search for ‘the good life’ or ‘Eudaimonia’.

Naturally, the development and study of such a construct requires definition and valid measurement tools. Happiness (subjective well-being) is a broad social paradigm regarded by researchers as a concept that includes both analysis of satisfaction with life alongside positive and negative affect (Deiner & Oshi, 2005). Current measures of subjective well-being are derived from the ambiguity that when answering the question of whether or not a person is happy, they may reply either in reference to their feeling in the moment, or their feelings about their life as a whole. It is the latter and broader form of questioning which psychologists have used predominantly in measurement of subjective well-being/ happiness. One of the first questionnaires to measure happiness was the Delighted-Terrible scale (D-T scale: Andrews & Withey, 1976), a single item 7 point Likert type scale which asks: “how do you feel about life as a whole?: Terrible 1 – 7 Delighted”. Whilst this scale is certainly simple to use and interpret, it is however impossible to examine its internal consistency. However, it does have a good convergent consistency in that it correlates well with other more diverse proceeding happiness scales. Another option has been to use the Subjective Happiness Scale (http://www.ppc.sas.openn.edu/ppquestionnaires.htm). This scale is structured around four statements on a 7 point scale. The first is a broad life satisfaction type statement, whilst the following three are comparison based questions such as “compared to most of my peers I consider myself: Less Happy1 – 7 More Happy”. This scale is consequently heavily dependent upon social comparison. One of the most popular happiness measures in the UK however is the Oxford Happiness Questionnaire (Hills & Argyle, 2002; http://happiness-survey.com/survey/). Participants are asked to rate by strongly agreeing or disagreeing 29 items on a 6-point scale. Scales within this questionnaire however have no neutral midpoint.  Content validity has also been severely criticised by Kashdan (2004) in that this scale contains items which measure various other specific antecedents or consequences of subjective well-being. Questions within the scale include statements relating overtly to the constructs of ‘agreeableness, ‘autonomy’, ‘extraversion’ and interestingly ‘self-esteem’ – without proposing a theoretical concept as to the reasons for their aggregations. The various shortcomings of the above happiness scales led to the consideration and popular use of a subjective well-being measurement which encompasses a more targeted rationale.

Research continues to support the concept that in order for people to experience happiness or subjective wellbeing (subjective well-being) they must not only like their lives, but also experience a greater level of positive emotions than negative ones. Happiness, or subjective well-being in broadest terms should thus involve a multidimensional evaluation of life, including cognitive judgments of life satisfaction and affective evaluations of emotions and moods (McGillivray and Clarke 2006). One of the most widely used and validated methods of measuring happiness today is the approach used herein. This involves collating data from both the satisfaction with life scale and the positive and negative affect scale (PANAS) (as provided in Appendix C and as detailed within ‘Measurements’ herein). This is done by subtracting the negative affect subscale within PANAS, from the amalgamation of satisfaction with life and the positive affect subscale of PANAS. Studies in happiness which utilise this approach acknowledge that both emotional affect and life satisfaction are vital components. Within subjective well-being therefore, life satisfaction reflects individuals’ perceived distance from their aspirations whilst emotional affect (PANAS) results from a balance between positive and negative emotion. subjective well-being is thus regarded as a state of wellbeing, of long duration which includes both the affective and cognitive component (Bruni and Porter 2007, p. xvii). This method of measurement avoids falling foul of dimensions which may be too simple, too comparative or too dependent upon other clouding constructs, whilst at the same time has demonstrated to have not only excellent convergent, but also internal consistency.

Subjective well-being is clearly important within mental well-being concepts, is significant in everyday lives and has undergone much empirically validation as a reliable construct. Happiness, or subjective well-being is a far-reaching and inclusive paradigm which is differentiated from and studied in isolation of other positive constructs.  Previous research investigating strength use has considered its bearing upon other positive constructs such as that of satisfaction with life (Douglas & Duffy, 2014) which encompasses only a small element of happiness and the SWB measurement method. Whilst satisfaction with life examines global life satisfaction only, happiness contemplates the pivotal issue of an individual’s own assessment of his or her life whilst relating fully to the concept of both temporal or experiential happiness. Satisfaction with life is a component scale of subjective well-being and a separate paradigm. The present method of combining measures of life satisfaction with measures of happiness to create an aggregate subjective well-being score has in the past been justified empirically through consistently fair to good correlations. This highlights that whilst satisfaction with life and happiness are related constructs, they should however be considered as distinct and separate paradigms.

In addition to the benefits of using a composite scale to measure the broad construct of happiness, another key aspect of analysing subjective well-being is its ‘subjective’ nature. The reason that happiness is termed as having a ‘subjective’ element is crucially due to the view that it is the individual who can best assess their own level of happiness. To illustrate this, individuals for example who have recurring depression may have differing levels of happiness or subjective well-being, despite a chronic mental health condition (Macaskill, 2012). The subjective or introspective element is thus important in gaining a full emotional picture, where opinion based upon outwards behaviours only may otherwise have been misleading.

The increasing importance of accurately and relevantly measuring happiness as discussed above is apparent in the expanding are of emotion and happiness research today. In the wider field of general psychology, the study of emotions has been a rapidly growing area. In 2011 for example, Izard (2011) carried out qualitative research into the structure and purpose of emotions. The 34 eminent scientists within the field of emotions who were surveyed produced themes illustrating that emotions such as happiness have great importance in the motivating and organisation of cognition and action. National governments also have become interested in the study of positive emotions, having initiated research programs directed at encouraging national wellbeing. Nicholas Sarkozy commissioned a nationwide piece of research to devise an index of the happiness of French citizens in order to combine with measures of GDP in creating a broad index of the nation’s well-being (Easterlin, 2011). Similarly, the UK Office of National Statistics under David Cameron was commissioned to measure national happiness within the United Kingdom (Weijers & Jarden, 2013; Layard, 2011) which has led to the continued measurement of national happiness alongside GDP today. Continuing on a practical level, it has been seen that interventions devised through the study of positive constructs such as subjective well-being have indeed proven significant in reducing pathologies such as depression. (Sin & Lyubomirsky, 2009). In more social terms, the extensive adoption and development in recent years of psychologically researched methods such as ‘scaffolding’, ‘circle time’ and ‘mindfulness’ further demonstrate the impact and practical implications of studies relating to subjective and mental wellbeing (Huppert & Johnson, 2010). Subjective well-being is an emotion or group of emotions which is thus not only of individual, but of social, political and broad academic importance today, in various economic and psychological spheres of study. A separate construct which has been studied vigorously alongside happiness is that of ‘strengths use’.

The benefits of strengths use and development of a scale of measurement.

As incidences and awareness of mental illness grow, the study of psychological human strengths has been supported empirically as an area which assists with improvement of well-being. Published research consistently suggests that people who use their strengths are happier (Linley, Nielsen, Gillett, Biswas-Deiner, 2010). Specifically, identification and use of psychological strengths have been supported in their ability to foster subjective well-being, resilience and in better coping with stress. (Martinez-Marti & Ruch, 2017; Seligman, Steen, Park & Peterson, 2005; Proctor, Maltby, & Linley, 2010). It was further found in 2010 that use of strengths vocationally and in a volunteering capacity related to heightened levels of well-being (Littman-Ovadia and Steger, 2010). Similarly, use of strengths at work (Harzer & Ruch, 2013) led to increased positive experiences. Thus, a wealth of empirical research supports the concept that strengths use relates to increased levels of wellbeing (Emmons & McCullogh, 2003; Otake, Shimai, Tanaka-Matsumi, Otsui & Fredrickson, 2006; Linley, Nielsen, Gillett, Biswas-Diener, 2010). Such research is consistently placed within a highly practical context.

With a focus upon prevention of pathology and in light of emboldening research as to the influence of strengths use upon wellbeing, researchers continue to examine the implications of strengths use in the development of interventions and counselling techniques. Underlying the concept of research questions such as examined in the present study, is the idea that usage of such positive traits, termed character strengths, serve as resources for people to use in the improvement of personal health and wellbeing (Proctor, Maltby & Linley, 2010). Many therapists today use strengths-based interventions to improve wellbeing and buffer against stress. Thus, various clinical and coaching methods have consequently been developed relating to the development of strengths and their uses over the last 15 - 20 years. Given the implications of strengths use upon subjective and other areas of wellbeing, alongside the existence of a vast array of related therapeutic techniques, it is argued that strengths analysis should therefore be part of well-being and health assessments (Wright & Lopez, 2002).

The starting point and basis for the aforementioned array of research was a cross-cultural examination of the predominant positive values found world-wide in 2004 by Peterson and Seligman (Peterson & Seligman, 2004). The study found 6 virtues which were broadly recommended across various populations: wisdom, temperance, justice, humanity, courage and transcendence. Under each of these 6 virtues was then arranged 4 character strengths such as hope, curiosity etc. thus comprising a list of 24 such character strengths in total. This forms the Values In Action – Inventory of Strengths (VIA-IS; Peterson and Seligman, 2004), developed in order to identify and measure an individuals strengths, providing a DMV style assessment of such character traits. Ensuing empirical cross-cultural studies incorporating the VIA-IS linked character strengths cross culturally with subjective well-being. (Park et al., 2004). It would be challenging to find anything distasteful or culpable in the promotion of such exemplary traits as for example ‘temperance, ‘courage’, ‘justice’ wisdom and love. However, the pursuit and promotion of these virtues has been criticised by some researches as encouraging selfishness (Layard, 2016). Disapproval lies in the concept that said strengths or virtues are often sought or promoted with a focus on ‘subjective’ wellbeing or eudemonia, as a benefit to the self rather than due to the wider benefits of the acts themselves. In other words, nurturing a just or empathetic act for the sake of one’s own wellbeing rather than for the sake of the result of the act, is perceived as a less virtuous precursor. However, the debate over the potentially selfish motives of promoting positive and widely beneficial traits in order to produce a goal of subjective wellbeing, is outside of the realms of this discussion. Extant research into the identification and promotion of character strengths has nonetheless formed a sturdy basis for the following work into practical application of character strengths and their effects.

Following on from the work on basic analysis and identification of character strengths, the strengths use scale (as seen in Appendix C), was developed in 2007 for the purpose of research by Govindji and linley (2007) in which researchers found that self-esteem and strengths use both predicted unique variance in subjective well-being. Govindji & Linleys research underlined the importance of measurement and application of strengths use vs mere knowledge of strengths. A finding of note therein and one which impacted following studies into strengths, was that whilst identification and knowledge of strengths was not found to independently predict subjective well-being, strengths use did. It has therefore been deemed as appropriate over the last decade to examine strengths use as a separate and more influential construct in the prediction of happiness, than mere knowledge or identification of strengths. Studies since, involving the active use of strengths have supported not only increases in wellbeing but also in lower levels of depression (Mongrain & Anselmo-Matthews, 2012). Important longitudinal research has also suggested that subjective well-being has been seen to increase in students who had strengths based exercises integrated into their curriculum, versus those students who did not, over time (Proctor, Tsukayama, Wood, Maltby, Eades & Linley, 2011). Whilst numerous links between strengths use and subjective well-being have been empirically supported, there is a dearth of research into how and why these relationships may occur. In relation to this, Govinji and Linley suggested in 2007 that future research might wish to examine the existence of an interactive effect between strengths use and self-esteem. This study looks specifically at the interactive influence of self-esteem, in relation to the well-established predictive role of strengths use upon subjective wellbeing.

Self-esteem and its relationship with positive constructs.

In exploring the aforementioned hypothesis regarding self-esteem, explanations of this heavily studied construct have varied greatly over the last century. James defined it in 1892 as the ratio of ones’ successes over ones attempts. In 1967 Coopersmith defined it as a self-belief of capability, significance, success and worthiness (Harter, 1993). In reference to various definitions, self-esteem is commonly measured validly today using Rosenbergs (1979) self-esteem scale, as seen in Appendix C. This scale is frequently used in cross cultural research and has therefore been translated into many languages. Whilst many have defined self-esteem as a largely domain-specific construct (Harter, 1993), studies have also shown the idea of general self-esteem to be a relevant dispositional variable (Judge, locke, Durham & Kluger, 1998) with general application to many other psychological variables, as contemplated herein. Scientists have also been historically divided on the role of self-esteem, with some regarding its position as purely descriptive and non-causal, versus others bestowing a central and explanatory function (Branden, 1984, p.12). Further, whilst research has linked self-esteem to scholastic achievement in children, research by Harter (1993) found that scholastic achievement itself had the capacity to cause self-esteem, rather than vice versa. Given much conflict then regarding the placement and role of this construct, in 2003 the American Psychological Society Task Force on Self-esteem published a report (Baumeister et. al., 2003) on the state of this construct. The position of self-esteem according to Baumeister (2003, p.2) is as an individual’s perception, rather than a reality. High self-esteem and low self-esteem could be for example either accurate and well-founded or a distorted, even pathological sense of self. It is possible therefore for a person with high self-esteem to have a self-concept which is dramatically better than other peoples evaluation. Consequently, in light of the much contested role of self-esteem, several hypotheses from extant research should be borne in mind when considering this variable alongside positive constructs.

Firstly, on the basis that self-esteem has been described as ‘perception’ and not ‘reality’ (Baumeister, 2003) it is not surprising that self-esteem correlated as highly as .85 with self-report ratings of physical attractiveness (Harter, 1993). As it is unlikely that self-esteem and physical attractiveness are in fact measuring the same construct, as such a high correlation is said to indicate, it is likely rather that those with high self-esteem tend to speak highly of themselves across many domains. In support of this, a 1995 study (Deiner, Wolsic & Fujita) found that ratings by others of physical attractiveness correlated as little as 0.6. with the subjects self-esteem. These contrasting correlations together illustrate that self-esteem is less about reality than it is about thinking or speaking highly of oneself and implies a very high level of ‘subjectivity’ of this construct. (Baumeister 2003). When this concept therefore is applied to its impact upon happiness it may be of note that the contribution of self-esteem may further increase the subjective element of subjective wellbeing – highlighting or perhaps increasing a potential delusional element of perception rather than a real indication of the impact of self-esteem upon certain individuals happiness.

Secondly, in analysing self-esteem alongside positive constructs, researchers have also attended to the possibility of circular reasoning. Correlations between self-esteem and happiness have been seen to be extremely robust. A Cross-national study using a large student sample showed self-esteem to be the strongest predictor of happiness when compared to multiple other predictive variables, showing a .47 correlation (Deiner & Deiner, 1995). Another study using retired participants in the US similarly resulted in a high correlation between self-esteem and happiness of .58. It is hard however to pinpoint directionality relating to a construct which has been seen to operate frequently as both as an antecedent and an outcome. The question of whether self-esteem causes happiness, or happiness causes self-esteem, and how this comes about is an important consideration in research of such highly correlated variables.

Thirdly, researchers have traditionally held that low self-esteem is a trait which leads to aggressive behaviour and that those at the low end of the self-esteem score spectrum are more likely to display aggression (Horney, 1950). Conversely, Baumeister in 2003 found support for the theory that where the self concept is linked to narcissism, egotism and inflated self-esteem, threat to this perception by others is more likely to cause aggressive retaliation than low self-esteem would. However, subsequently research by Donnellan, Trzesniewski, Robins, Moffitt & Caspi (2005) found that low self-esteem caused heightened levels of aggression, and that this was irrespective of narcissism. Findings that negative constructs such as aggression surface at both ends of the self-esteem spectrum have very much swung back and forth in research over the last 10 years,  maintaining debate over the influence of self-esteem as a ‘body builder’ construct. Thus, Findings in this sphere must be considered accordingly regarding the established effect of self-esteem upon the concept of a distinctly positive paradigm such as happiness. Self -esteem based aggression could conceivably lower scores of subjective well-being at the upper end of the correlation between self-esteem and subjective well-being. This could create an inverted ‘U’ association between these two variables, under Baumeister’s theory. Similarly, Grant and Schwartz (2011) have opined upon such variables providing ‘too much of a good thing’ and ultimately producing negative outcomes if possessed to a high degree. This debate should thus be borne in mind in analysing results for those with very high as well as very low self-esteem. Baumeister further suggests that in light of this, the potential for fostering narcissistic or egotistical traits should be considered in the development of interventions aimed at boosting self-esteem. There are suggestions for example that such methods should aim to achieve a more balanced and accurate level of self-esteem which combines appreciation of strengths as well as limitations, rather than simply aiming for indiscriminate growth. (Baumeister, 2004). In addition to elements of self-esteem research which may inform the present research questions, other aspects of this pivotal construct are closely attuned to the study of positive variables.

Very few studies have monitored self-esteem alongside positive constructs historically (Baumgardner & Corothers, 2009). Over the past 6 years however it has found a place more strongly embedded within positive psychological studies. This could have been due to debates as outlined above which may complicate studies in a burgeoning field such as strengths use. It has been after all a challenge to determine whether self-esteem should be examined as a variable which affects behaviour or is produced as a consequence of it. Researchers and clinicians now however see self-esteem in a context highly compatible with studies relating to wellbeing (Millner, 2012). In relation to this for example, expansion theory is also significant. Alongside the buffering, self-regulating function of self-esteem (consistency theory), the other perceived function of this construct in relation to helping the self to grow or actualize (expansion theory) (Pyszczynski, Greenberg, Solomon, Arndt & Schimel, 2004) is decidedly connectable to positive psychology principles. Self-esteem is frequently  understood as a positive, predictive construct which also maintains interesting links to areas such as age and gender (Heppner, Kernis, Nezlek, Foster, Lakey & Goldman, 2008). Self-esteem is found to consistently correlate with happiness as a distinct variable and appears to be innately compatible with the study of subjective wellbeing. This with the caveat that extant theory should be considered, particularly relating to extremes of self-esteem, circular referencing and the deeply subjective or perceptional nature as set out above. In theorising therefore upon the location of this capacious paradigm, a firm grasp of extant theory is required.

The placement of self-esteem as a mediator or moderator in the relationship between strengths use and subjective well-being.

In attempting to unravel or refine the involvement of self-esteem in the relationship between strengths use and subjective well-being, there is both cross sectional and longitudinal research to guide theory in placing self-esteem as a moderator or mediator. In beginning with the proposed predictor variable, strengths use has been linked to self-esteem in that heightened levels of self-esteem were found in those who actively used their strengths (Minhas, 2010). Additionally, Proctor (2011) found that students who carried out self-esteem exercises experienced greater levels of subjective well-being. This correlation between strengths use and self-esteem has been supported in longitudinal findings (Wood et al., 2011) demonstrating that strengths use leads to greater self-esteem over time.

In approaching the link beyond the well-established path from strengths use to self-esteem as examined convincingly by Wood et al (2011), much support is found for a connection leading from self-esteem to subjective wellbeing. Self-esteem has been found to correlate highly with life satisfaction amongst a large sample of college students (Diener & Diener, 1995). Other studies have similarly found self-esteem to be a predictor of happiness related concepts (Zhang & Leung, 2002; Chen, Cheung, Bond & Leung 2006; Joshanloo & Afshari, 2011). Importantly, longitudinal methods have in addition demonstrated the predictive ability of self-esteem upon satisfaction with life, a large component of subjective well-being (Ye,Yu & Li, 2012). Despite concern over circular reasoning, temporal precedence has been proposed by longitudinal findings in support of the present research question which places self-esteem in between strengths use and subjective well-being, as outlined later herein. In further suggestion of such an arrangement Govindji and Linley found in 2007 that both strengths use and self-esteem predicted subjective well-being (Govindji and Linley, 2007). Research subsequent to this by Proctor  et al. (Proctor, Maltby and Linley, 2010) found in a study of undergraduate university students that those who use their strengths, experience greater wellbeing and additionally that subjective well-being is related to both physical and mental quality of life. Proctors work expanded upon previous findings to suggest that strengths use was a unique predictor of subjective well-being when controlling for self-esteem and advised additional research to improve generalizability.

Whilst certain variables can be identified as predicting others in a linear model with regression analysis, such as that between strengths use and subjective well-being (Govindji, 2007; Proctor, Maltby & Linley, 2011) this study sought to analyse in greater depth two potential functions of self-esteem in this relationship, based on its suggested location between strengths use and subjective well-being. This manuscript examines firstly a potential ‘additive affect’ of self-esteem upon strengths use causing increased subjective well-being (moderation), and secondly, the eventuality that the relationship between strengths use and subjective well-being might itself be ‘explained’ by self-esteem (mediation). In undertaking this analysis, researchers also sought to recruit a wide socio economic spectrum of subjects. Both self-esteem and subjective well-being have been shown in previous studies to be impacted heavily by demographic aspects such as age, gender, relationship satisfaction and wealth or income (Headey & Wooden, 2004; Deiner & Biswas-Deiner, 2001 Erol & Orth, 2017; McMullin & Cairney, 2004). Previous research into both subjective well-being and self-esteem has acknowledged limitations in studies due to limited diversity in potentially important socio demographic figures. Many for example have utilised all student samples (Proctor, Maltby & Linley, 2011; Douglas & Duffy, 2015; Macaskill & Denovan, 2014; Govindji & Linley, 2007). The broader spectrum of participants herein may reduce skew due to lack of availability in data of such factors and provide a more representative cross section.

This study therefore set out to utilise a broad spectrum of adult participants per suggestions in extant research and diversify away from the trend in related work utilising student-based samples (Douglas and Duffy, 2015). It sought to fill a gap in current research in asking not only if, but how strengths use and self-esteem have an impact upon happiness or subjective well-being. Consequently, in examining the potential reasons for numerous findings that strengths use benefits happiness, it analyses two research questions:

1) To what degree would the study variables of strengths use, self-esteem and subjective well-being correlate within a more diverse data set?

2) Would the link between strengths use and subjective well-being be moderated or mediated by self-esteem?

2.  Method

Research Design

 A cross sectional study was conducted amongst a diverse adult sample in order to examine the role of self-esteem in the relationship between strengths use and subjective wellbeing.

Sample

Participants were 123 adults aged between 18 – 65 years (75 females, 48 males) from across the UK designed to allow for robust analyses of study variables. The mean age category of participants was 3.09 (SD = 1.14, range = 35 – 44 years) as shown in Appendix A. Participants were volunteers who completed the study questionnaire after being sent a link to the survey either via social media or by researchers’ contacts. The sample encompassed a wide variation in levels of education, income, marital status and employment (Appendix A: Descriptive Analyses). The sample comprised of mostly white British adults (71%), and included white European (9%), other white background (9%), Indian (3%), White Irish (2%), Other Mixed Ethnic background (2%), Chinese (1%), White Asian (1%), Caribbean (1%) and Chinese (1%). The mean completion time of the questionnaire was 9.6 minutes.

Materials

The below psychological scales and socio-demographic questions were used in a ‘Survey Monkey’ questionnaire (as seen in Appendix C) sent to participants using an email link to the survey website and via social media such as ‘Instagram’.

Strengths Use Survey (Govindji & Linley,2007)

strengths use is a 14 item, self-report, 7-point Likert scale that measures individual strengths use with answers ranging from strongly disagree to strongly agree. For example: “I always play to my strengths” and “Using my strengths comes naturally to me”. The internal consistency was found to range from 0.95 to 0.97 with a test – retest reliability of 0.84 (Wood et al., 2011). In strengths use instrument development studies the alpha has been reported at 0.95 and has been shown to correlate with appropriate criterion measures such as the Rosenberg self-esteem scale (r = 0.56) and other constructs such as subjective well-being (r = 51) (Govindji and Linley, 2007).

Life Satisfaction (Diener et al., 1985)

The satisfaction with life is a 5 item 7-point Likert scale with answers ranging from strongly disagree to strongly agree. Questions include “The conditions of my life are excellent” and “so far I have gotten the important things I want in life”. Instrument development studies show the satisfaction with life to have internal consistency of 0.87 and test – retest reliability of 0.82. The test has been found in many studies to correlate in expected directions with appropriate variables such as self-esteem (Arrindell et al). Note: subjective well-being = (positive affect + satisfaction with life) -negative affect (Diener et al., 1985):

Positive and Negative Affect Schedule (PANAS, Watson et al., 1988)

This is a 20 item, self-report measure made up of two subscales: ten Positive and ten Negative items(PANAS: Watson et al., 1988). The positive subscale asks to what extent an individual has felt a certain way, such as “strong”, “determined”, “inspired” (Positive Affect) or “distressed”, “upset”, guilty” in the past week. Answers are on a 5-point Likert scale and range from very slightly to very much. PANAS has been shown to correlate win expected directions with relevant constructs. For PA, internal consistency has ranged from 0.86 to 0.90.

Subjective Wellbeing Measurement

Composite subjective well-being variables were calculated by subtracting the negative affect subscale of PANAS, from global judgements of life satisfaction (satisfaction with life), per previous research (Proctor, Maltby, Linley, 2011)

Rosenberg self-esteem scale (Rosenberg, 1965)

The Self-esteem is a 10 item, self-report, 4-point Likert scale asking questions such as “I certainly feel useless at times” with answers ranging from strongly disagree to strongly agree. Internal consistency reliabilities range from 0.80 to 0.92 with test – retest correlation reported at 0.82 (Sam, 2000).

Procedure

The Survey Monkey App was used to create an on-line questionnaire and to record data prior to transfer to SPSS. A wide variation in participants was achieved by recruiting adults from personal contact lists and randomly via social media. A consent form, alongside privacy and confidentiality information was agreed to by each participant within the online the survey prior to being permitted to progress to the questionnaire section (Appendix C ). Completed questionnaire data was transferred to SPSS. 5 items of the Self-esteem scale were subsequently reverse coded and a new mean self-esteem variable was created in order to insure that a high score would indeed signify high self-esteem. Mean scores were then computed for all other relevant variables. Positive Affect and satisfaction with life were then added and from this negative affect was subtracted in order to compute the subjective well-being variable (Sheldon & Elliott, 1999; Deiner and Lucas, 1999) and standardised. After data preparation, initial analysis of descriptives, box plots, scatter plots and data cleaning was carried out (Appendix A). Analysis of reliabilities using Cronbach’s alphas and intercorrelations was undertaken (Appendix B). Following this moderation and mediation analyses were completed in order to study the nature of involvement of self-esteem in the relationship between strengths use and subjective well-being. Whilst It has been possible traditionally to conduct both moderation and mediation analysis in SPSS using macros or by following a manual process, the PROCESS dialogue box was used to conduct multiple regression analyses and simultaneously assess for moderating and mediating relationships, as illustrated in the Analysis section later herein. Finally, further regressions were carried out for the purposes of comparison with previous studies which used slightly different methodology as traditionally required in standard moderation and mediation analysis (Appendix B).

Ethics

Permission was granted by the BPP ethics committee after full proposal review. The survey was structured such that participants were only permitted to proceed from the survey information page to the questionnaire itself once they had acknowledged all patient information re anonymity, data use and right of withdrawal, and given relevant consent, as seen in Appendix C.

Analysis

In analysing the potential mediating or moderating role of self-esteem, the PROCESS custom dialogue box file (Hayes, 2012) was installed and integrated into SPSS software. This dialogue box automates and integrates the traditional multi step, manual method for carrying out both moderation and mediation analyses. This software thus initially conducts linear regressions to analyse predictive qualities between variables and then assesses the existence of either an ‘additive’ effect (moderation), or an ‘explanatory’ effect (mediation) upon the relationship. Due to recent developments in analytical method and as the distinction between moderation and mediation models is often perplexed, both the concepts and analytical methods used in this research are clarified herein.

Moderation

Whilst previous research which contemplates relationships between the present variables utilises regression analyses to look only at predictive relationships, moderation analysis examines the ‘combined’ effect of multiple variables on an outcome variable. When a new predictor variable such as self-esteem is included and potentially ‘combines’ with strengths use to have a joint effect upon subjective well-being, a moderating effect is observed, as conceptualised in Figure 1.1. If it is found for example at low levels of self-esteem that there is a small association between strengths use and subjective well-being, but at high self-esteem that there is a greater association between strengths use and subjective well-being, then we would say that self-esteem moderates the association between these two constructs. Conceptually therefore, a moderation model will assess if and the degree to which the relationship between strengths use and subjective well-being changes, as a function of self-esteem.

Moderation analyses in SPSS traditionally involved manual creation of new variables via the ‘compute’ command, followed by centring or transforming variables into deviations around a fixed point to avoid uninterpretable predictors associated with the interaction effect. Interaction variables would customarily then need to be computed manually via multiplication of the predictor variable and the moderator (strengths use x self-esteem). Regression outcomes of the strengths use (predictor) and self-esteem (moderator) upon subjective well-being would then be computed alongside the interaction outcome. If the interaction of strengths use x self-esteem is found to be a significant predictor of subjective well-being then a moderating effect has occurred. In practice, moderation analysis occurs using various regression analyses. The practical (or statistical) moderation process as just discussed is thus illustrated in Figure 1.2, with a slightly different diagram than is used to highlight the conceptual effect. The next step traditionally was then to refine the exact nature of the moderator, by assessing the impact of self-esteem as the moderator at various levels or strengths, using slopes analysis ( Aiken & West, 1991). However, whilst the statistical procedure remains the same today the PROCESS custom dialogue box (Hayes, 2012) was used to carry out these traditionally manual tasks by automatically a) centring predictors, b) calculating the interaction term and c) computing simple slopes analysis.

Mediation

Rather than simply producing a ‘combined’ effect upon an outcome variable, in mediation the relationship between a predictor and an outcome is instead ‘explained’ through their involvement with a third ‘mediating’ variable. The present study thus also examined whether or not self-esteem is a variable which operates as a potential mechanism through which strengths use can alter subjective well-being. Importantly when the effect of self-esteem is removed therefore, the relationship between subjective well-being and self-esteem may even disappear. Figure 1.3 illustrates how a mediated relationship functions, in that a positive relationship between strengths use and subjective well-being can be explained by the presence of self-esteem. This model would potentially suggest that the relationship between strengths use and subjective well-being is not a direct effect, but operates instead through a reduction in self-esteem.

Mediation statistical analyses has also historically utilised a set of regressions, but requiring 4 conditions in order to demonstrate mediation. These are: a) strengths use must predict subjective well-being; b) strengths use must also predict self-esteem; c) self-esteem must in turn predict subjective well-being and d) The relationship between strengths use and subjective well-being must be ‘reduced’ when self-esteem is included in the model. Prior to the availability of PROCESS These 4 conditions were tested manually using 3 regression models (Baron & Kenny, 1986): 1) prediction of subjective well-being from strengths use; 2) prediction of self-esteem from strengths use; 3) prediction of subjective well-being from both strengths use and self-esteem. Mediation would then be seen if the outcome is predicted less strongly in regression model 3, than in model 1. Following this a Sobels test is carried out to determine significance. However, manual method just described of comparing the strength of prediction between the indirect vs indirect pathways is criticised (Field, 2012). This study therefore utilised the more recent method of employing the PROCESS dialogue box within SPSS, in lieu of a manual or macro based approach (Preacher and Hayes, 2008). The PROCESS dialogue box enabled analysis of mediation regressions and comparisons automatically (Hayes, 2012). It is noteworthy that in place of a Sobels test, the PROCESS dialogue box calculated confidence intervals for the indirect effect based on bootstrapped standard error (Efron & Tibshirani, 1993). The method used herein was elected as it is deemed preferable due to its increased focus upon examining the ‘degree’ of mediation found, rather than simply looking for significance as within a Sobels test. (Field, 2012).

The present study conducted both moderation and mediation analysis in order to examine whether self-esteem varied the strength or direction of the relationship between strengths use and subjective well-being, or whether a positive relationship found between these variables can be explained by (or operates through) the presence of self-esteem.

3.  Results

Primary Analysis

Outliers were assessed using box plots (Appendix A) and normality was analysed using descriptive statistics and visual histogram inspection (Appendix A). Variables showed few extremes and no outliers were removed. Non-demographic data was deemed parametric in nature, showing only minimal skewness and kurtosis, as shown in Appendix A and Table 1, with no detection of violations of assumptions for proposed analyses. Statistics were reported with a two-tailed significance and with alpha levels set at 0.05, unless stated otherwise. Scatter plots were produced in order to visually assess the nature of initial relationships between study variables (Appendix A) and revealed no notable deviations from extant theory.  Cronbach’s alphas were calculated and all study variables were shown to have high reliabilities as seen in Table 2. Subsequently the two research questions regarding 1) correlations and 2) mediation and moderation properties of self-esteem, were addressed.

Table 3.1

 Descriptive statistics

The strengths use, satisfaction with life, positive affect and self-esteem scales all had high reliabilities when tested within the present study. All Cronbach’s a > 0.86 as in Table 1.

Table 3.2

Reliabilities

3.1.  The relationship between strengths use, self-esteem and subjective well-being.

Following preliminary analysis, Pearson’s correlation coefficients between study variables were calculated within a matrix using SPSS as shown in Table 2. Strong correlations were found between all variables, with subjective well-being and self-esteem (self-esteem) showing the strongest correlation with each other (r = 0.73, p = 0.00).

Table 3.3

Intercorrelations between study variables

All correlations are significant at the 0.01 level

3.2.  Moderation and mediation of self-esteem in the relationship between strengths use and subjective well-being.

Having found support both within our dataset that strengths use and self-esteem are both significant predictors of subjective well-being, and also that strengths use predicts self-esteem, moderation and mediation analyses were then used to explore how these constructs may interact. The next research question asked whether self-esteem combined or interacted with strengths use to alter either strength or direction of the relationship in a moderating role. It then examined whether strengths use actually operated via self-esteem itself, in order to explain levels of subjective well-being via a mediating role.

3.2.1.       Moderation analysis

Moderation analysis using the PROCESS custom dialogue box for SPSS revealed that the relationship between strengths use and subjective well-being was not moderated by self-esteem. self-esteem did not itself affect the direction or strength of the relationship which strengths use has with subjective well-being in our sample. b = 0.098, self-esteem = 0.227, t = -0.434, p = 0.665 as seen in Table 3.4.

Table 3.4

Linear model of predictors of subjective well-being

3.2.2 Mediation analysis

Mediation analysis found that self-esteem did act as a mediator between strengths use and subjective well-being. As bootstrapped confidence intervals did not contain 0, there was deemed to be a significant indirect effect of strengths use upon subjective well-being through relationship commitment by self-esteem, b = 0.5567, BCa CI (0.357, 0.779). This represents a relatively sizeable mediating effect, k2 = 0.309, 95% BCa CI (0.204, 0.426) as shown in Table 3.5. The model suggested that the relationship between strengths use and subjective well-being is therefore not a direct effect but operates instead through a reduction in self-esteem. As noted previously, within a mediation model, where an indirect effect increases (the effect including the mediator) the new direct effect between strengths use and subjective well-being will necessarily decrease. Our results demonstrate therefore that strengths use significantly affects subjective well-being through self-esteem, with self-esteem acting as a substantial mediator.

Table 3.5

Indirect (mediated) effect of self-esteem upon the relationship between strengths use and subjective wellbeing

Supplemental regression analyses were also carried out separately in order to compare and verify findings with previous research which used the manual method of mediation analysis, prior to creation of the automated process developed by Preacher & Hayes (2012), as seen in Appendix B.

4.  Discussion

Whilst previous studies have examined the predictive abilities of strengths use and self-esteem upon subjective well-being, the present research examined not only correlations between variables, but also utilised contemporary analytical methods (Hayes, 2012) to assess the specific role of self-esteem within this relationship. In relation to the first research question as to the degree to which study variables would correlate within a more diverse data set, elevated scores of strengths use were unambiguously connected with greater levels of self-esteem. Strengths use and subjective well-being similarly showed a strong positive correlation as seen in previous research (Govindji & Linley, 2007; Proctor et al, 2011). The present study thus emphasises the strengths of correlational links between these three variables and continues in addressing a likely reason as to how these relationships may function.  

In consideration of extant research, the second research question then addressed a possible moderating or mediating role of self-esteem upon the existing relationship between strengths use and subjective well-being. Whilst no moderation effect was found, mediation results established that self-esteem partially mediates the link between strengths use and subjective well-being to a meaningful degree. In interpreting these results, findings indicate it unlikely that high levels of strengths use automatically cause levels of happiness (subjective well-being) irrespective of self-esteem. Findings suggest rather that strengths use leads to other affects (which include self-esteem) which if not present can undermine the effect of strengths use on subjective well-being. This model would suggest that the relationship between strengths use and subjective well-being is not a direct effect therefore, but that it operates instead through a reduction in self-esteem. A potential rationale for the relationship between strengths use and subjective well-being as partially explained by self-esteem, is perhaps that individuals who are able to utilise their strengths more often, experience a greater level of positive feeling towards themselves and their abilities. This may then lead to superior levels of self-esteem. In turn, self-esteem contributes to and indeed partially explains greater levels of happiness. There are several further implications and considerations which should be highlighted in digestion of these results.

Findings bring to the fore the importance of self-esteem in the study of positive constructs such as happiness and other well-being measurements. Extant theory regarding the ability of strengths use to lead to higher levels of self-esteem (Minhas, 1020; Proctor, 2007) is reinforced and the predictive nature of self-esteem upon subjective well-being (Chen, 2006; Joshanloo, 2011) is emphasised. Accordingly, the important place of self-esteem in relation to various positive constructs and studies relating to happiness and wellbeing is underlined. On a more conceptual level increasing support has been found herein for the role of self-esteem within human psychology as having an explanatory and central capacity (Branden, 1984) versus the purely descriptive, non-causal role which other researchers have suggested is provided by this construct.

Given the recent return of both science and government (Weijers & Jarden, 2013; Layard, 2011) to the original dual focus of not only the remedying of dysfunction but also the promotion of optimum human functioning, the practical applications of findings such as these within several domains are broad. In clinical practice this knowledge can inform interventions for patients suffering with pathologies such as depression (Sin & Lyubomirsky, 2009). Additionally, counselling and coaching psychologists often use character strengths and self-esteem methods as resources to help individuals deal with adversity and buffer against stress. Combined with extant theory and future research, this study adds to knowledge of central pathways leading to happiness and the crucial involvement of self-esteem in such therapeutic practice. Treatments can thus be devised which synthesize with these known pathways in aid of improved mental wellbeing and the creation of preventative therapies. Such practical implications can in turn embolden the return to a dual factor model of mental health assessment and improve upon the current ‘disease approach’ to mental wellbeing (Greenspoon & Skolfske, 2001: Keyes, 2007).

In addressing considerations relating to self-esteem, future research may also wish to explore the idea that the relationship which self-esteem has with subjective well-being may be concave down (inverted ‘U’ shape). Our findings did not support a negative impact upon subjective wellbeing of those with high self-esteem scores. However a larger sample size which includes a presence of extremely high self-esteem individuals may enrich Grant and Schwartz’s ‘too much of a good thing’ theory (2011). In further considering the concept of ‘body builder’ constructs, there may be merit in exploring strengths theory, in virtues or strengths which whilst in moderation assist with positive well-being, if possessed to a disproportionately high degree may have negative impact upon happiness or mental health. In this line of study, the distinction between well-being and subjective well-being would be crucial. Given the subjective nature of both happiness and self-esteem, it may be necessary to measure in future not only subjective well-being but also other less subjective areas of well-being to avoid ‘delusion’ impacting well-being results (Baumeister, 2004).

Another area of implication which this study may have is in the development of current happiness scales. Self-esteem has been seen herein to hold an important mediating or ‘explanatory’ role between strengths use and happiness. These findings may thus provide a stimulus for a closer re-examination of a rationale for inclusion of concepts such as self-esteem within its measurement questions. Kashdan (2004) criticised the oxford happiness scale heavily for its inclusion of items which incorporated without rationale, distinct concepts such as self-esteem, agreeableness, autonomy and extraversion. Given confirmation herein of the explanatory role of one of these variables (self-esteem) in conjunction with happiness, researchers may wish to examine the mediating properties of remaining variables as aggregated within the oxford happiness questionnaire. Such future results may proceed to embolden theory relating to integration of such constructs previously criticised within the oxford scale.

The present research had several limitations. Firstly, due to the studies’ cross-sectional nature it was not possible to determine strict causal relations. In relation to this concern, Maxwell et al (2011) opined that mediation effects seen within cross sectional data may not be reliable or indeed significant when tested longitudinally, despite previous cross sectional findings suggesting cautious interpretation until longitudinal testing. That said however, the sequential structuring of variables and basis for this research is founded upon several pieces of longitudinal research. For example, Wood (2011) found that strengths use led to increased self-esteem over time and Ye (2012) established longitudinally that self-esteem has a predictive effect upon satisfaction will life, the cognitive component of subjective well-being, thus perhaps allaying scepticism over the present cross sectional methodology. A longitudinal design replicating this investigation would lend more weight to these findings and provide an even firmer platform from which to examine other combined effects from constructs which may also impact the important outcome variable of happiness.

Secondly, whilst the sample size was adequate in relation to the quantities of variables and methodologies used, a larger cross section would have been desirable to ensure an even greater power and diversity amongst the sample. Whilst the sample was slightly skewed in terms of gender towards females, with little assortment regarding ethnicity, there was good range in terms of age, levels of education, income, marital status and employment. This data set has therefore permitted good generalizability amongst the present findings and could be used to further analyse relationships between age and wage level as suggested in past research (Govinji & Linley, 2007). However, a larger cross section may allow for further mediation analysis pertaining to relevant demographic variables.  Given that past research has cited self-esteem and subjective well-being as constructs vulnerable to age, income, relationship contentment and job satisfaction (Headey & Wooden, 2001; Deiner & Biswas-Deiner, 2001, Erol & Orth, 2011; McMullin & Cairney, 2004) on-going research could examine the operation of this relationship further by using a ‘moderated mediation’ approach to examine how such important demographic variables may interact in the link between self-esteem and subjective well-being. Equally, as diversity of socio-demographic data is present herein, further analysis could be carried out which controls for variables already gathered such as age, income, gender in order to gauge the impact of such social factors upon various positive constructs. The importance of socio-demographic variables amongst self-esteem and subjective well-being highlights one of the strengths of this research, as being its’ participant diversity. Previous research using predominantly student samples (Proctor, Maltby & Linley, 2011; Douglas & Duffy, 2015; Macaskill & Denovan, 2014; Govindji & Linley, 2007) may thus be subject to very homogeneous age, income and relationship status. As these are variables known to impact both self-esteem and subjective well-being the broader spectrum of participants in the present study may have produced higher mean self-esteem and subjective well-being scores herein, based upon higher levels of age and income (Headey & Wooden, 2001; Deiner & Biswas-Deiner, 2001). Interestingly, this may be exemplified in the supplemental regressions carried out as shown in Appendix B which resulted in higher coefficients than those studies using the same variables with students alone. This may be noteworthy on a practical level as further examination could lead to identification of interventions which may assist particular groups within society in achieving greater mental well-being.

Thirdly, the reliability of self-report measures in relation to these constructs can often be questioned due to social desirability response bias (Van de Mortel, 2008). In addition, self-esteem is often a skewed concept due to the very nature of people having high self-esteem potentially inflating their own self view (Baumeister, 2005). Strengths use may also be open to less accurate self-reporting given that people do not always know what their strengths are to begin with (Seligman, 2002). Not necessarily having carried out strengths assessment interventions prior to the survey may mean it less likely that individuals can accurately report their strengths use. It is possible that future studies involving strengths use may benefit from prior inclusion of the VIA-IS exercise in order to first highlight an individuals strengths. As a 240 item exercise, including the VIA-IS would involve a considerably lengthier questionnaire. Whilst the strengths use scale has been shown to have good criterion validity, high internal consistency and strong test-retest reliability, a pilot study which compares strengths use responses both before and after completing the VIA-IS could determine the need for inclusion of the VIA-IS as an additional pre-measurement in strengths use studies. Additionally, despite criticisms of self-report measures and of the deeply subjective nature of happiness, the subjective element has been seen as crucial in the accurate research of subjective well-being, with self-report methods acting as a vital element of the detailed study of happiness (Macaskill, 2015).

5.  Conclusion

These findings pave the way for further evaluation of other mediators or moderators affecting the demonstrably important link between strengths use and happiness. The importance of understanding and promoting subjective well-being at both an individual and societal level has been expounded and acted upon by various national governments – the implications of happiness and mental well-being upon GDP for example having been well understood by various leaderships. Thus, the demand for in depth analysis of contributing variables as provided within this study, and hopefully within proceeding research, will enable mental health practitioners, coaches and therapists to better understand pathways to subjective well-being. This will in turn allow them to devise the most effective interventions in support of wellbeing across various socio-demographic groups. The present study’s findings could be helpful to clinicians in understanding the mechanisms which contribute to lower subjective well-being in combatting pathologies such as depression. The introduction of psychologically researched concepts and activities such as circle time, scaffolding and mindfulness into schools and institutions worldwide has improved aspects of mental welfare in recent history. It is vital therefore that happiness findings such as those herein continue to be translated into tangible tools which can then be brought forward into domains where society can benefit. Happiness is a construct which has been seen to affect both physical and mental welfare and this research sought to refine contributing factors to enable development of not only curative but preventative techniques. This manuscript contributes to knowledge of positive concepts examined within fields of personality, counselling, clinical and social psychology and draws a spotlight upon the importance of self-esteem in consideration of such positive paradigms. This research further supports the model that happiness and related variables should be measured and evaluated in conjunction with mental well-being assessment and should involve consideration of factors relating to self-esteem.

REFERENCES:

Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and interpreting interactions. Sage.

Andrews, F. M., & Withey, S. B. (1976). Americans’ Well-Being: Specific Life Concerns. In Social Indicators of Well-Being (pp. 249-281). Springer, Boston, MA.

Arrindell, W. A., Heesink, J., & Feij, J. A. (1999). The satisfaction with life scale (satisfaction with life): Appraisal with 1700 healthy young adults in The Netherlands. Personality and individual differences, 26(5), 815-826.

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of personality and social psychology, 51(6), 1173.

Baumeister, R. F., Campbell, J. D., Krueger, J. I., & Vohs, K. D. (2003). Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles?. Psychological science in the public interest, 4(1), 1-44.

Baumgardner, S. R. & Crothers, MK (2009). Positive psychology.

Biswas-Diener, R., Kashdan, T. B., & Minhas, G. (2011). A dynamic approach to psychological strength development and intervention. The Journal of Positive Psychology, 6(2), 106-118.

Branden, N. (1995). The six pillars of self-esteem. Bantam Dell Publishing Group.

Brdar, I., & Kashdan, T. B. (2010). Character strengths and well-being in Croatia: An empirical investigation of structure and correlates. Journal of research in personality, 44(1), 151-154.

Bruni, L., & Porta, P. L. (2001). 2007. Handbook on the Economics of Happiness.

Chen, S. X., Cheung, F. M., Bond, M. H., & Leung, J. P. (2006). Going beyond self‐esteem to predict life satisfaction: The Chinese case. Asian Journal of Social Psychology, 9(1), 24-35.

Weijers, D., & Jarden, A. (2013). The science of happiness for policymakers: An overview. Journal of Social Research & Policy, 4(2).

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Pub.

Diener, E., & Chan, M. Y. (2011). Happy people live longer: Subjective well‐being contributes to health and longevity. Applied Psychology: Health and Well‐Being, 3(1), 1-43.

Diener, E., Diener, M., & Diener, C. (1995). Factors predicting the subjective well-being of nations. Journal of personality and social psychology, 69(5), 851.

Diener, E., & Oishi, S. (2005). The nonobvious social psychology of happiness. Psychological Inquiry, 16(4), 162-167.

Diener, E., Wolsic, B., & Fujita, F. (1995). Physical attractiveness and subjective well-being. Journal of personality and social psychology, 69(1), 120.

Donnellan, M. Brent, Kali H. Trzesniewski, Richard W. Robins, Terrie E. Moffitt, and Avshalom Caspi. "Low self-esteem is related to aggression, antisocial behavior, and delinquency." Psychological science 16, no. 4 (2005): 328-335.

Douglass, R. P., & Duffy, R. D. (2015). Strengths use and life satisfaction: A moderated mediation approach. Journal of Happiness Studies, 16(3), 619-632.

Easterlin, R. A., Angelescu, L., & Zweig, J. S. (2011). The impact of modern economic growth on urban–rural differences in subjective well-being. World development, 39(12), 2187-2198.

Emmons, R. A., & McCullough, M. E. (2003). Counting blessings versus burdens: an experimental investigation of gratitude and subjective well-being in daily life. Journal of personality and social psychology, 84(2), 377.

Erol, R. Y., & Orth, U. (2017). Self-esteem and the quality of romantic relationships. European Psychologist.

Field, A. (2014). Discovering statistics using ibm spss statistics+ spss version 22.0. Sage Publications.

Forest, A. L., & Wood, J. V. (2012). When social networking is not working: Individuals with low self-esteem recognize but do not reap the benefits of self-disclosure on Facebook. Psychological science, 23(3), 295-302.

Govindji, R., & Linley, P. A. (2007). Strengths use, self-concordance and well-being: Implications for strengths coaching and coaching psychologists. International Coaching Psychology Review, 2(2), 143-153.

Grant, A. M., & Schwartz, B. (2011). Too much of a good thing: The challenge and opportunity of the inverted U. Perspectives on Psychological Science, 6(1), 61-76.

Greenspoon, P. J., & Saklofske, D. H. (2001). Toward an integration of subjective well-being and psychopathology. Social Indicators Research, 54(1), 81-108.

Harter, S. (1993). Causes and consequences of low self-esteem in children and adolescents. In Self-esteem (pp. 87-116). Springer, Boston, MA.

Harzer, C., & Ruch, W. (2013). The application of signature character strengths and positive experiences at work. Journal of Happiness Studies, 14(3), 965-983.

Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, mod- eration, and conditional process modeling [White paper]. Retrieved from http://www.afhayes.com/ public/process2012.pdf

Headey, B., & Wooden, M. (2004). The effects of wealth and income on subjective well‐being and ill‐being. Economic record, 80, S24-S33.

Heppner, W. L., Kernis, M. H., Nezlek, J. B., Foster, J., Lakey, C. E., & Goldman, B. M. (2008). Within-person relationships among daily self-esteem, need satisfaction, and authenticity. Psychological Science, 19(11), 1140-1145.

Hills, P., & Argyle, M. (2002). The Oxford Happiness Questionnaire: a compact scale for the measurement of psychological well-being. Personality and individual differences, 33(7), 1073-1082.

Horney, K. (1950). The Collected Works of Karen Horney: Self analysis. Neurosis and human growth (Vol. 2). WW Norton.

Izard, C. E. (2011). Forms and functions of emotions: Matters of emotion–cognition interactions. Emotion Review, 3(4), 371-378.

Joshanloo, M., & Afshari, S. (2011). Big Five personality traits and self-esteem as predictors of life satisfaction in Iranian Muslim university students. Journal of Happiness Studies, 12(1), 105-113.

Judge, T. A., Locke, E. A., Durham, C. C., & Kluger, A. N. (1998). Dispositional effects on job and life satisfaction: The role of core evaluations. Journal of applied psychology, 83(1), 17.

Kashdan, T. B. (2004). The assessment of subjective well-being (issues raised by the Oxford Happiness Questionnaire). Personality and individual differences, 36(5), 1225-1232.

Judge, T. A., Locke, E. A., Durham, C. C., & Kluger, A. N. (1998). Dispositional effects on job and life satisfaction: The role of core evaluations. Journal of applied psychology, 83(1), 17.

Keyes, C. L. (2005). Mental illness and/or mental health? Investigating axioms of the complete state model of health. Journal of consulting and clinical psychology, 73(3), 539.

Layard, R. (2011). Government’s role should be to increase happiness and reduce misery. Policy analysis must be recast to reflect outcomes in terms of changes to happiness. British Politics and Policy at LSE. Retrieved from http://blogs.lse.ac.uk/politicsandpolicy/2011/10/17/happines-and-misery/ retrieved April 2018.

LAYARD, R. (2016). Promoting Happiness Ethics: The Greatest Happiness Principle. WORLD HAPPINESS, 50. Retrieved from https://www.badishams.com/wp-content/uploads/2017/11/World-Happiness-Report-2016.pdf#page=52 April, 2018

Linley, P. A., Nielsen, K. M., Gillett, R., & Biswas-Diener, R. (2010). Using signature strengths in pursuit of goals: Effects on goal progress, need satisfaction, and well-being, and implications for coaching psychologists. International Coaching Psychology Review, 5(1), 6-15.

Littman-Ovadia, H., & Steger, M. (2010). Character strengths and well-being among volunteers and employees: Toward an integrative model. The Journal of Positive Psychology, 5(6), 419-430.

Lyubomirsky, S., Tkach, C., & DiMatteo, M. R. (2006). What are the differences between happiness and self-esteem. Social Indicators Research, 78(3), 363-404.

Macaskill, A. (2012). A feasibility study of psychological strengths and well-being assessment in individuals living with recurrent depression. The Journal of Positive Psychology, 7(5), 372-386.

Martínez-Martí, M. L., & Ruch, W. (2017). Character strengths predict resilience over and above positive affect, self-efficacy, optimism, social support, self-esteem, and life satisfaction. The Journal of Positive Psychology, 12(2), 110-119.

McGillivray, M., & Clarke, M. (2006). Human well-being: Concepts and measures (pp. 3-16). United Nations University Press.

McMullin, J. A., & Cairney, J. (2004). Self-esteem and the intersection of age, class, and gender. Journal of aging studies, 18(1), 75-90.

Millner, U. C. (2012). Review of Positive psychology: The science of happiness and human strengths.

Minhas, G. (2010). Developing realised and unrealised strengths: Implications for engagement, self-esteem, life satisfaction and well-being. Assessment and Development Matters, 2(1), 12.

Mongrain, M., & Anselmo-Matthews, T. (2012). Do positive psychology exercises work? A replication of Seligman et al.(). Journal of clinical psychology, 68(4).

Otake, K., Shimai, S., Tanaka-Matsumi, J., Otsui, K., & Fredrickson, B. L. (2006). Happy people become happier through kindness: A counting kindnesses intervention. Journal of happiness studies, 7(3), 361-375.

Park, N., Peterson, C., & Seligman, M. E. (2004). Strengths of character and well-being. Journal of social and Clinical Psychology, 23(5), 603-619.

Peterson, C., & Seligman, M. E. (2004). Character strengths and virtues: A handbook and classification (Vol. 1). Oxford University Press.

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior research methods, 40(3), 879-891.

Proctor, C., Maltby, J., & Linley, P. A. (2011). Strengths use as a predictor of well-being and health-related quality of life. Journal of Happiness Studies, 12(1), 153-169.

Proctor, C., Tsukayama, E., Wood, A. M., Maltby, J., Eades, J. F., & Linley, P. A. (2011). Strengths gym: The impact of a character strengths-based intervention on the life satisfaction and well-being of adolescents. The Journal of Positive Psychology, 6(5), 377-388.

Rosenberg, M. (1965). Rosenberg self-esteem scale (Self-esteem). Acceptance and commitment therapy. Measures package, 61, 52.

Pyszczynski, T., Greenberg, J., Solomon, S., Arndt, J., & Schimel, J. (2004). Why do people need self-esteem? A theoretical and empirical review. Psychological bulletin, 130(3), 435.

Seligman, M. E., Steen, T. A., Park, N., & Peterson, C. (2005). Positive psychology progress: empirical validation of interventions. American psychologist, 60(5), 410.

Sims, A., Barker, C., Price, C., & Fornells-Ambrojo, M. (2015). Psychological impact of identifying character strengths in people with psychosis. Psychosis, 7(2), 179-182.

Sin, N. L., & Lyubomirsky, S. (2009). Enhancing well‐being and alleviating depressive symptoms with positive psychology interventions: A practice‐friendly meta‐analysis. Journal of clinical psychology, 65(5), 467-487.

Van de Mortel, T. F. (2008). Faking it: social desirability response bias in self-report research. Australian Journal of Advanced Nursing, The, 25(4), 40.

Weijers, D., & Jarden, A. (2017). Wellbeing Policy. Wellbeing, Recovery and Mental Health, 35.

Wood, A. M., Linley, P. A., Maltby, J., Kashdan, T. B., & Hurling, R. (2011). Using personal and psychological strengths leads to increases in well-being over time: A longitudinal study and the development of the strengths use questionnaire. Personality and Individual Differences, 50(1), 15-19.

Wood, A. M., & Tarrier, N. (2010). Positive clinical psychology: A new vision and strategy for integrated research and practice. Clinical psychology review, 30(7), 819-829.

Wright, B. A., & Lopez, S. J. (2002). Widening the diagnostic focus. Handbook of positive psychology, 26-44.

Ye, S., Yu, L., & Li, K. K. (2012). A cross-lagged model of self-esteem and life satisfaction: Gender differences among Chinese university students. Personality and Individual Differences, 52(4), 546-551.

Douglass, R. P., & Duffy, R. D. (2015). Strengths use and life satisfaction: a moderated mediation approach. Journal of Happiness Studies, 16(3), 619-632.

Zhang, L., & Leung, J. P. (2002). Moderating effects of gender and age on the relationship between self-esteem and life satisfaction in mainland Chinese. International Journal of Psychology, 37(2), 83-91