Children’s wellbeing and their academic achievement: The dangerous discourse of ‘trade-offs’ in education

Abstract

Research conducted in England over the last decade has documented sustained, significant decreases in children’s wellbeing. While recent changes to curriculum policy promoting children’s wellbeing have been introduced, a notable feature of the discourse surrounding the promotion of children’s wellbeing is that wellbeing is regarded as opposed to, or in tension with, children’s academic achievement. Recently, Gabriel Heller-Sahlgren proposed that there is an inevitable ‘trade-off’ between children’s ‘wellbeing’ and their academic achievement. Using PISA 2012 data, Heller-Sahlgren argues that pupil happiness and high achievement do not go hand in hand; implying policymakers have a decision to make about which they uphold as the priority. In this article, I discuss the theoretical assumptions underpinning transnational comparisons of children’s wellbeing and review evidence from psychology and education to ascertain whether a trade-off is empirically supported. I argue that far from being incompatible, children’s wellbeing and achievement are positively associated. However, this relationship is not straightforward and requires careful disentangling of the hedonic and eudaimonic components of wellbeing. I underline four main gaps in current knowledge of the wellbeing-achievement relationship to date: the need for (1) multidimensional conceptualisation and measurement of wellbeing, (2) exploration of mediating mechanisms/constructs explaining the wellbeing-achievement relationship, (3) objective operationalisation of achievement, and (4) investigation of developmental differences. To conclude, I argue that when making policy recommendations researchers should avoid ‘all or nothing’ thinking which lures governments into false dichotomies.

Introduction: The wellbeing-achievement ‘trade-off’

In Summer 2018, an independent think tank – the Centre for Education Economics (CfEE) – released a research report entitled ‘The achievement-wellbeing trade-off in education’ (Heller-Sahlgren, 2018a). The report seemingly advances an achievement-oriented and ‘knowledge-led’ (Young and Muller, 2013) way of thinking about education (compare Association of School and College Leaders & Parents and Teachers for Excellence, 2017; Simons and Porter, 2015). In the United Kingdom, this conceptualisation of education has permeated policy; for example, Nick Gibb’s knowledge-led curriculum agenda which rowed back from the inclusion Social and Emotional Learning (SEL) in schools (Department for Education, 2015, 2017). The central premise of Heller-Sahlgren’s report is that policymakers have a decision to make about what they uphold as the priority in schools: children’s wellbeing or their academic achievement. Heller- Sahlgren surfaces an assumption many educational thinkers tacitly make about the relationship between wellbeing and academic performance: that there exists a trade-off between the two and that improvements in one will always come at the expense of the other. For instance, Hayes and Ecclestone (2009) write,

Therapeutic education draws students towards easy, unchallenging preoccupation of everyday concerns and feelings and away from a desire to rise to the challenging of pursuing truth, mastering a difficult subject or learning craft skills. (p. 156)

Similarly Craig (2007) writes,

Rather than focus on feelings and dwell on ‘all about me’ activities, redolent of American classrooms, we want teachers to help young people acquire skills. (p. 97)

What’s more by emphasising the importance of teaching social and emotional skills in the curriculum, schools are no longer seen primarily as the place where children learn academic skills and acquire knowledge. This is very likely to lead to a decline in academic standards. (p. 22)

Finally, Claxon (2005) gives expression to the idea by writing,

We might also ask whether there is at least a risk, in some situations, that too much attention to students’ emotions might make them more sensitive to disruption, more easily put off learning by their own fluctuating feelings . . . Surely, there is a complicated, momentary, intuitive balance that teachers are continually having to strike between stopping to attend to the feelings, and encouraging a child to ‘just get on with it’? (p. 17)

As representative of this line of thinking, I therefore propose to examine Heller-Sahlgren’s report in detail to question whether there indeed exists a trade-off between wellbeing and academic performance. Through my critical review of neglected research, I debunk the idea of a ‘trade-off’ altogether, instead underlining the need for a vision of education that considers how children’s wellbeing needs might interact with their achievement. As O’Shaughnessy (2015) puts it,

The old dichotomy that you can have happy children or successful children is wrong. A true education provides not one, or the other, but both. (p. 32)

Yet, the ‘old dichotomy’ referred to is resurrected in Heller-Sahlgren’s report and thus, it is important that the evidence underpinning this idea is once again revisited.

Aim and scope

In this article, I use Heller-Sahlgren’s (2018a) report as a springboard for in-depth critique of the theory and research exploring the wellbeing-achievement relationship. In his report, Heller-Sahlgren (2018) makes the following key arguments:

• If the goal of education is achievement, the most effective pedagogies are teacher-led, which are in turn ‘less joyful’, resulting in lower pupil happiness;

• ‘ . . . homework, school competition and traditional teaching methods – involve an achievement-happiness trade-off – . . . [they] increase pupils’ test scores, while decreasing pupil happiness’;

• Policymakers must take seriously the idea of ‘trade-offs’ in education, such as wellbeing versus achievement, and decide ‘which goals to promote and which ones to discard’ (pp. v–vi).

Granted, Heller-Sahlgren (2018) warns against ‘ignoring pupil wellbeing entirely’ (p. vi), yet his overriding proposition of a trade-off is one that many educationalists challenge: schools are fully capable of holding more than one goal at a time (Cigman, 2012). My overall aim in the article will be to argue for the following two points. First, assuming that schools are capable of considering multiple goals eliminates the need for a ‘trade-off’ between wellbeing and achievement; second, and relatedly, seeing schools as pursuing multiple goals calls instead for in-depth understanding of the wellbeing-achievement relationship. In support, I will make the following arguments:

1. Children’s wellbeing is multidimensional and complex. To fully engage with children’s wellbeing, researchers require an understanding of what it constitutes, including clear definitions and shared terminology.

2. Just as the clarity of definitions of children’s wellbeing is important, so is the precision of instruments used to measure it; due to its multidimensional nature. Single-item measures of children’s wellbeing lack construct validity. Rather, the use of multi-item measures assessing different sub-constructs of children’s wellbeing allow researchers to understand which particular aspects are most important for achievement.

3. Researchers and policymakers alike must question the underlying theoretical assumptions made by transnational comparisons of children’s wellbeing. Cultural differences are at risk of being overlooked by etic approaches, impacting the validity of findings.

4. Research investigating the relationship between children’s wellbeing and achievement suggests not that they are incompatible educational goals but rather, that the relationship is underexplored. Specifically, research suggests greater hedonic wellbeing is important for Primary pupils’ achievement, whereas greater interpersonal and eudaimonic wellbeing is important for Secondary pupils. Gaps in the knowledge exist due to research:

   (a) not using multidimensional measures, limiting understanding of specific aspects of wellbeing that are critical for children during particular developmental windows,

   (b) overlooking potentially important psychological mediators,

   (c) neglecting eudaimonic wellbeing.

While defining wellbeing is a focus of this article, mental health is not. This distinction is unpacked elsewhere (Keyes, 2002). ¹ Likewise, the pedagogic methods serving as ‘explanatory’ variables in Heller-Sahlgren’s report are not explored here; I focus on the wellbeing-achievement ‘trade-off’ idea, leaving matters of ‘progressive’ versus ‘traditional’ teaching to more knowledgeable colleagues. In the main, I situate this article within the UK context. However, the implications drawn from my discussion of the theoretical issues underpinning transnational comparisons of children’s wellbeing clearly have international applicability. I argue that pitching a wellbeing-achievement trade-off is especially dangerous in the UK context: an ever-pressured performativity culture and education system revolved around high-stakes achievement and wherein children experience unprecedentedly low levels of wellbeing (Rees et al., 2012).

The context

Debate and research on the relationship between children’s wellbeing and their academic achievement has been ongoing and intense over the past decade; engaging scholars across disciplines including health (Bonell et al., 2014), psychology (Gutman and Vorhaus, 2012; Suldo et al., 2011), education (Ecclestone, 2012a; Spratt, 2017), educational philosophy (Cigman, 2012; Clack, 2012) and economics (Maccagnan et al., 2019). The juxtaposing of achievement and wellbeing, or ‘enhancement’ versus ‘knowledge’ agendas (Cigman, 2012) is therefore, not new. Nor is the focus on ‘wellbeing’ in education. Educationalists have been challenging the role of school in shaping children’s social and emotional development for a decade (Craig, 2009; Ecclestone, 2012b). Theoretically however, wellbeing is an altogether different construct, and one which, following the COVID-19 pandemic, is at the forefront of educational agendas. Indeed, measuring and conceptualising young people’s wellbeing has become an area of prolific engagement in research (McLellan and Steward, 2015; Novello et al., 1992; OECD, 2014; Tomyn et al., 2015; UNICEF, 2007b; Waters, 2011). The intensified interest in wellbeing within education can be understood as the result of a renewed societal focus on wellbeing in the West, the reasons for which are manifold and vocalised from a variety of interdisciplinary standpoints. To name but a few: the incorporation of wellbeing measures into assessments of economic and social progress (Layard, 2005; Stiglitz et al., 2009), interest in post-materialist happiness in richer countries (Delhey, 2010; Inglehart, 1997) and the impact of technology (O’Keeffe et al., 2011). The school, where citizens of tomorrow spend their formative years, has consequently become a context considered critical to laying the foundations of individual wellbeing; reflected in the ‘positive education’ movement (Seligman, 2011).

Of course, very little in education is straightforward, and schools are continually balancing priorities. Whether this be delivering curricula of ‘powerful knowledge’ (Young and Muller, 2013), teacher recruitment and retainment, or ensuring pupils leave school with the minimally required grades for entrance into a competitive job market. In a school agenda overcrowded with priorities, should children’s wellbeing be considered one demand too many being added to the list? Or do we need to reconceptualise education altogether? I argue, wellbeing is as important for children’s individual development and future lives as knowledge acquisition. Schools provide the foundational soil from which children grow into adults. Neglecting to tend to the key ‘nutrients’ needed for all aspects of children’s growth (i.e. social, emotional and academic: henceforth, ‘the whole child’) fails to grasp the broader societal, social and economic responsibilities of education. While children can develop – academically speaking – from basic ‘soil’ alone, schools that nurture all aspects of children’s development should be the preferred vision.

Educationalists advocate such a vision in which schools have the goal of helping children live flourishing lives; extending beyond the knowledge agenda which has long been held the sole purpose of education in England (White, 2016). Yet, as White indicates, the UK government’s ‘stranglehold’ of high-stakes examinations currently poses a barrier to this vision. If we consider evidence from wellbeing research with children and adults however, an education system designed to nurture children’s wellbeing alongside achievement is difficult to contest.

Research illustrates that wellbeing and achievement in childhood and adulthood, are closely intertwined. Without achieving employment in adulthood, an adult’s objective and subjective wellbeing is significantly lower than their counterparts in employment (What Works Centre for Wellbeing, 2017). This wellbeing-achievement relationship comes full circle in childhood, whereby children without parents or adult role models in employment experience significantly lower wellbeing than their peers (Cusworth, 2016; Rees et al., 2011, 2015). Moreover, parental wellbeing is as important as parental employment for their children’s wellbeing; Powdthavee and Vignoles (2008) theorise the intergenerational transmission of stress from parent to child, finding that parental stress negatively impacts children’s wellbeing in both the long- and short-term. Schools have a powerful role to play which extends beyond getting pupils through examinations, to educating the whole child. Trading-off children’s wellbeing in favour of achievement would detrimentally impact children for whom school makes up for what is not provided at home.

Notably, UK government advice advocating improving pupil wellbeing as an effective means of raising achievement (Public Health England, 2014) and the introduction of ‘mental health’ and ‘emotional wellbeing’ into compulsory curricula (Department for Education, 2019) follows over a decade of accumulated evidence suggesting children have critically lower levels of wellbeing compared to their counterparts in other countries (UNICEF, 2013, 2007b). This comes at a time when a continued national decline in children’s wellbeing has been documented by longitudinal research (The Children’s Society, 2020).

Conceptualising and measuring children’s wellbeing

What’s in a name? Defining and distinguishing wellbeing, happiness and flourishing

Given the complexity of the wellbeing construct, researchers should be cautious in the terminology used, providing clear definitions reflected in subsequent measures. Heller-Sahlgren’s report uses many terms: ‘achievement-wellbeing’ and ‘achievement-happiness’ alongside ‘pupil wellbeing’, ‘pupil happiness’ and ‘human flourishing’ without definitions, suggesting synonymity to wellbeing (as per his report title). In fact, each requires separate conceptualisation in relation to the literature. First, I therefore offer a conceptualisation of wellbeing.

Wellbeing is a multi-faceted and complex state of being and theorisation of the concept dates back to the beginning of intellectual history (Aristotle, 1985). Today, wellbeing research is commonly attributed to the modern discourse of positive psychology (Seligman and Csikszentmihaly, 2000). Although this article predominantly focalises children’s wellbeing through a psychological lens, alternative disciplinary standpoints including philosophy (Noddings, 2003), sociology (Ben-Arieh, 2012) and economics (Powdthavee and Vignoles, 2008) are invaluable. Indeed, children’s wellbeing cannot be understood in isolation without consideration of what it means for life to go well (Taylor, 2012), socio-cultural influences and societal values (Inglehart, 1997).

Wellbeing has been conceptualised in both an objective and a subjective sense. Objective indicators of wellbeing include children’s physical health and educational attainment (UNICEF, 2007a), whereas subjective indicators represent children’s self-reported satisfaction with life, experiences of positive and negative affect, relationships and flourishing. Psychologists have typically focused on the latter, subjective conceptualisation. Whereas, traditionally, economists used objective indicators (such as measures of health and material possessions), an important shift occurred in the early twenty-first century owing to distinguished economists urging policymakers to shift their measurements of human development from indicators of Gross Domestic Product (GDP) to subjective indicators capturing individuals’ general quality of life (Stiglitz et al., 2009). Consequently, over the past two decades measures of subjective wellbeing have become commonplace across psychology and economics (Layard, 2005).

The shift to subjective conceptualisations of children’s wellbeing was marked by UNICEF in 2007 when their measures expanded, incorporating subjective indicators elicited by asking children directly about their experiences. Children’s self-reports were collected in addition to existing measures of children’s ‘well-becoming’ (i.e. objective indicators of what it means for children to be ‘well’, as perceived by adults: children’s behaviours, attainment, career aspirations) (UNICEF, 2007a). Importantly, significant intra-country differences emerged across objective and subjective indicators of children’s wellbeing. For instance, Greece, ranking low for children’s objective wellbeing (25th out of 29 rich countries), moved up to fifth place when subjective measures were instead considered. This is a stark presentation of the measurement discrepancies between the two different focalisations adopted (the former being ‘adultcentric’; the latter from the perspectives of children themselves). ²

What does individual wellbeing constitute? It can be understood as the extent of an individual’s combined feelings of happiness and fulfilment in life, though disagreements exist among psychologists regarding this. To appreciate the nature of these disagreements, it is important to understand the two historical paradigms of wellbeing: hedonia and eudaimonia. Hedonism dates back to Greek philosopher Aristippus, who believed life’s central goal was to experience the most pleasure possible. Psychologists conducting work aligned to the hedonic approach see wellbeing as existing on a continuum of human experience with pain and pleasure at opposite ends. The most common understanding of subjective wellbeing (SWB) according to the hedonic approach – supported by research – is through a composite measure of self-reported life satisfaction, extent of positive emotions and absence of negative emotions (Diener and Lucas, 1999). According to hedonic theorists, this measure of ‘SWB’ equates to individuals’ ‘happiness’.

Eudaimonia, or ‘leading the good life’, however, can be traced back to Aristotlean thinking; referring to a state of flourishing and self-actualisation, beyond momentary feelings of joy or negative affect (Aristotle, 1985). The eudaimonic approach proposes true happiness is achieved through committing to worthwhile endeavours. ‘Worthwhile’ endeavours can be determined at individual or societal levels; which can shift over time, as seen at the beginning of the twenty-first century (the move from objective indicators of GDP to subjective measures of wellbeing). Crucially, eudaimonic wellbeing presents undeniable paradoxes when considered in conjunction with hedonic wellbeing. This is critical when considering the worthwhileness of education in the long- and short-term, as worthwhile long-term endeavours imagined by adult society (the achievement of A*–C grades) that could positively contribute to children’s eudaimonic wellbeing may be at odds with children’s own perceptions of worthwhile endeavours in the ‘here and now’, which fulfil their short-term hedonic wellbeing. Alternatively, achievement-oriented goals may be perceived as impossible by some children, which one can conceivably imagine having a detrimental impact on their short-term hedonic wellbeing. There is thus likely to be a complex interplay between eudaimonia and hedonia in schools.

Psychologists aligned with the eudaimonic tradition theorise that psychological wellbeing (PWB) is distinct from SWB in the hedonic tradition. Ryff and Singer (1998, 2000) for instance, theorised the lifespan theory of human flourishing, using a multidimensional approach to measure PWB according to six aspects of human actualisation: autonomy, personal growth, self-acceptance, life purpose, mastery and positive relatedness. Another well-researched theorisation of eudaimonic wellbeing is Self-Determination Theory (SDT; Deci and Ryan, 2000; Ryan and Deci, 2000), which purports the fulfilment of three psychological needs result in individual wellbeing, namely, competence, autonomy and relatedness. Need satisfaction is postulated as essential to healthy child development and psychological growth (Chirkov and Ryan, 2001; Diseth et al., 2012; Miserandino, 1996; Yamauchi and Tanaka, 1998).

A separate but related construct – ‘flourishing’ – is conceptualised as tantamount to eudaimonia, yet resembles a distinct philosophical discourse of child wellbeing (Spratt, 2017). Work on human flourishing is grounded in The Capability Approach which purports that wellbeing is an issue of social justice (Sen, 2010). According to Sen (1992), wellbeing is determined by the range of opportunities afforded for individuals which allow them to develop their ‘functionings’ and live their lives fully. Likewise, Nussbaum (2011) extends Sen’s thinking to a set of 10 ‘central capabilities’ children have a right to, one being education. Both Nussbaum and Sen, therefore, see children’s education as central to their lifelong wellbeing. Crucially, from this perspective wellbeing and achievement (or ‘flourishing’) are not two separate issues; but are intertwined.

As Spratt (2017) underlines however, when it comes to children’s pursuit of accomplishments in school, there are important nuances. Specifically, schools may be considered ‘anti-hedonic’ when it comes to children’s feelings of wellbeing (e.g. encouraging children to put fun aside in favour of ‘hard work’). Yet, school facilitates the development of eudaimonic wellbeing through the pursuit of longer term goals; eventually contributing to children’s lifelong flourishing. Research supports this. When wellbeing is conceptualised in a purely hedonic sense, asking children how happy they feel at school may provoke an instinctual response that they would prefer to be doing an activity on their own terms. However, such short-term feelings may be unrepresentative of children’s wellbeing altogether (both hedonic and eudaimonic). Indeed, Csikszentmihalyi and Hunter (2003) discovered this when investigating adolescents’ moment-to-moment happiness, which was lowest when engaging in schoolwork. Yet, the longer term happiness of pupils who studied more was higher than those who studied less. Contemplating these findings, the association between hedonism and eudaimonia becomes clearer: long-term endeavours of striving at school can have a positive far-reaching impact on wellbeing. Heller-Sahlgren (2018a) offers an alternative reading:

In other words, there appears to be an achievement – well-being trade-off . . . if pupils do not attend school, or do any work, they are unlikely to learn anything – but they may be happier. (p. 13)

This misses a vital nuance in Csikszentmihalyi and Hunter’s findings (distinguishing hedonia from eudaimonia) and misrepresents the wellbeing-achievement relationship. Nonetheless, Heller-Sahlgren is not unaware of the distinction between the two, demonstrated by his treatment of Rousseau:

For Rousseau, happiness could not be equated with joy or pleasure . . . it bore greater resemblance to the classic idea of human flourishing or eudemonia. (p. 10)

Yet, this argument is at odds with Heller-Sahlgren’s own analyses which use a single-item measure of children’s experiences of momentary happiness as a proxy for ‘wellbeing’. Thus, there is a lack of genuine engagement with the differences between hedonic feelings of happiness versus positive functioning or ‘flourishing’. As shown in the literature, children’s happiness is just one part of the complex construct. Thus, basing policy decisions and making trade-offs based on Heller-Sahlgren’s analyses is unjustified.

In addition to the limitations of the Heller-Sahlgren’s own research, the research drawn upon in support of his ‘trade-off’ thesis is weak. For example, studies are referenced that do not measure (or even mention) children’s wellbeing:

. . . research also suggests that traditional teaching methods worsen, and progressive methods improve, pupil well-being and attitudes toward learning (see Algan et al., 2011; Jiang and McComas, 2015; McConney et al., 2014; Regh, 2012). (p. 15) ³

At this point, I pause to contemplate the discussion thus far. I have provided a theoretical summary of how psychologists conceptualise wellbeing, including the hedonic and eudaimonic approaches. I consequently promoted, in accordance with others, a conceptualisation of children’s wellbeing which includes both feeling and functioning. For children to be well in school, this extends beyond how happy they generally feel; it should include their relationships with peers (Chanfreau et al., 2008), perceived competence (McLellan and Steward, 2015), sense of belonging (Anderman, 2002) self-esteem and positive functioning (New Economics Foundation, 2009).

Limited measures produce limited knowledge: The need for a multidimensional approach

In this section, I argue against the use of single-item measures of children’s wellbeing and alternatively propose researchers use multidimensional instruments to improve the validity and usefulness of findings. In Heller-Sahlgren’s report, wellbeing is operationalised using a single-item measure of children’s happiness at school taken from PISA 2012 (‘I feel happy at school’). I articulate why this is problematic, with reference to the literature.

Precision is important not only in how researchers conceptualise children’s wellbeing, but also in how researchers measure it. As noted, whereas historically children’s wellbeing was understood solely through objective indicators, there is now consensus that children’s self-reported wellbeing is fundamental. For instance, as observed in the move away from ‘adultcentric’ approaches using indicators of what adults in society at large deem as constituting ‘good’ wellbeing for children (Scott, 2000), to methodological approaches reliant on asking children directly what they are experiencing (McLellan and Steward, 2015). Not only have measures developed to foreground children’s own experiences, but two decades ago researchers acknowledged the need to develop more thorough and accurate tools reflecting the multidimensional nature of wellbeing (Andrews and Ben-Arieh, 1999).

This realisation marked an ideological shift away from measuring children’s basic ‘survival’ to an approach recognising the need to understand children’s functioning across a range of domains, including social connectedness, school engagement and emotional wellbeing (positive and negative affect). A number of justifications for the multidimensional measure of children’s wellbeing have been provided, the first being that they better reflect the complexity of the construct itself compared to unidimensional conceptualisations: ‘ . . . any attempts to grasp wellbeing in its entirety must use indicators on a variety of aspects of wellbeing’ (Ben-Arieh and Frønes, 2007). A second justification is that multiple-item measures reduce the measurement error associated with single-item measures (The Children’s Society, 2019). A third justification is the growing understanding of differences between children’s hedonic and eudaimonic wellbeing – including differences in their stability over short time periods – necessitating measures that capture distinctions between children’s feelings and functioning (The Children’s Society, 2013).

Following the widespread recognition and measurement of adults’ subjective wellbeing at the beginning of the twenty-first century, psychologists began exploring what it constitutes for children. Prominent thinkers in the field began to differentiate between emotional responses versus domain-specific and global judgments of life satisfaction. It was judged that whereas measures of affect help us understand children’s feelings of happiness, this construct existed separately to their life satisfaction (Diener et al., 1999). General life satisfaction was found to be distinct from other wellbeing sub-constructs such as self-esteem, positive and negative affect (Huebner, 2004). Whereas unidimensional conceptualisations of life satisfaction assumed a single total score represented one’s overall happiness, multidimensional approaches incorporated life satisfaction as a sub-construct of higher order wellbeing models, including both the hedonic and eudaimonic aspects. One example of this multidimensional approach is Seligman’s PERMA model (2011) (Positive emotions, Engagement, Relationships, Meaning, Accomplishment). Measuring children’s wellbeing accordingly led to widespread use of multi-item and multi-domain instruments (Casas et al., 2007; PISA, 2017), with the Children’s Society designing such measures in close consultation with thousands of children to ensure validity (Rees et al., 2008, 2010). Others became concerned with indicators pertaining to the ecology of child development, allowing us to understand children’s wellbeing operating within specific contexts (Brim, 1975) and both inside and outside of school (McLellan et al., 2012).

Considering the shift witnessed over the past two decades from unidimensional to multidimensional models of wellbeing, the single-item measure of pupil happiness Heller-Sahlgren uses in his analyses limits the implications drawn. If Heller-Sahlgren were interested solely in how satisfied pupils feel at school (defined wellbeing in accordance with traditional hedonic conceptions of ‘SWB’) (Diener and Lucas, 1999), this would require additional items measuring life satisfaction and the extent of accompanied negative affect children experience. Instead, the single item used measures positive affect only (happiness). This operationalisation of ‘wellbeing’ thus lacks validity according to the earliest conceptions of SWB from positive psychology.

When it comes to Heller-Sahlgren’s (2018a) analysis (see the report, p. 16), there are a number of points to address. First, the item measuring pupils’ happiness at school was not intended to be a stand-alone measure of wellbeing. The item is from the Programme for International Student Assessment (PISA) 2012 index of ‘sense of belonging’, a multi-item measure consisting of nine self-report items collectively capturing pupils’ sense of belonging at school. This original index was not intended to measure wellbeing. In fact, it was not until PISA 2015 that the OECD explicitly attempted to conceptualise the construct of child wellbeing. The conceptual framework of children’s wellbeing developed by the OECD for PISA 2015 is multi-dimensional, consisting of five sub-constructs: cognitive, material, social, physical and psychological wellbeing (Borgonovi and Pál, 2016). The index of ‘sense of belonging’ (OECD, 2013) was retained in the 2015 wellbeing model as the ‘social’ sub-construct. The item used by Heller-Sahlgren is taken from PISA 2012’s ‘sense of belonging’ index (‘I feel happy at school’) is one of the original nine self-report items. This background is vital to our interpretation of Heller-Sahlgren’s analysis and the single-item measure of happiness used.

In another paper, Heller-Sahlgren (2018b) uses the same single-item measure as a proxy for pupils’ overall wellbeing; justifying it as a valid and reliable measure of wellbeing by referencing others using single-item unidimensional constructs. However, the studies referenced concern adults rather than children (Krueger and Schkade, 2007; Veenhoven, 2012). This is a different construct: I earlier underlined how children’s feelings towards school in general (e.g. how happy they are to be at school) is cognitively evaluated against preferred ‘fun’ activities, whereas these studies using adults asked them how happy they were about life generally. These are therefore different constructs both at the adult/child level and the domain-specific/domain-general evaluation level. Other studies Heller-Sahlgren (2018b) references warn against the limitations of global measures ‘masking important sub-constructs of wellbeing’ (Gilman and Huebner, 2003).

A second point is Heller-Sahlgren’s use of PISA 2012 data for his analysis followed by a discussion around his contentions with the PISA 2017 report (2015 data): Heller-Sahlgren’s contention being that whereas PISA data suggest a negative wellbeing-achievement relationship, the OECD’s accompanying guidance is that school climates should still facilitate both in a self-reinforcing way. It seems peculiar that Heller-Sahlgren’s uses old PISA 2012 data for his analysis then, given his contention is with more recent PISA 2015 data. In fact, close inspection of the PISA report reveals a clear warning from the OECD that their findings should not be interpreted as a trade-off, underlining nuances in the wellbeing-achievement relationship across countries:

Data from PISA 2015 show that [ . . . ] students in low-achieving countries tend to report higher levels of life satisfaction than students in high-achieving countries . . . Some countries stand out from this general pattern . . . This correlation should not be interpreted as evidence of a trade-off between high achievement and student well-being. (p. 3). (PISA, 2017)

Nonetheless, examining Heller-Sahlgren’s analysis, the models presented illustrate regressions which are non-causal, despite Heller-Sahlgren’s criticism of others (Quinn and Duckworth, 2007) finding a correlational relationship with no causal link. Regrettably, Heller-Sahlgren does not provide any values for beta coefficients or the overall adjusted R². Readers are only told the p-value (which is significant at the < .01 level) and there is no explanation of the amount of the variance separately shared between pupil wellbeing, type of pedagogy and academic achievement. Moreover, since Heller-Sahlgren’s central thesis purports a negative wellbeing-achievement relationship, it is surprising that a model incorporating the two main constructs (achievement × wellbeing) is not presented. One would expect that if there is a significant wellbeing-achievement relationship, a model illustrating this would be key source of evidence; for example, a model regressing achievement × wellbeing with an interaction effect of type of pedagogy.

So far, I have suggested how researchers can improve wellbeing-achievement research through careful construct operationalisation and valid measures. Another important consideration for researchers working in the children’s wellbeing space is cultural context. Although this article reviews research from contexts other than the United Kingdom in an effort to pool together available evidence to date on the wellbeing-achievement relationship, the ecological systems in which children develop should not be taken for granted. Socio-cultural factors shape children’s conceptions and experiences of wellbeing. In the next section, I propose some theoretical assumptions made by transnational comparisons exploring children’s wellbeing and achievement across cultures. I argue these assumptions challenge the validity and reliability of findings derived from such comparisons and require careful consideration by researchers.

Transnational comparisons and the theoretical assumptions underpinning them

Transnational comparisons of children’s wellbeing across countries are increasingly commonplace. For example, the Children’s Worlds’ International Survey of Children’s Wellbeing (ISCWeB) and the OECD’s PISA. The ISCWeB is measuring children’s wellbeing across 40 counties in its third wave, while PISA 2018 encompassed all 37 OECD countries and 42 partner countries/economies. However, the underlying assumptions of such transnational comparisons should not go unquestioned. The obvious assumption being that wellbeing can be operationalised as a universal psychological construct; comparable across cultures. Unlike objective constructs, wellbeing is subjective and socio-culturally formed, meaning it does not easily lend itself to global measurement (Tov and Diener, 2009). Indeed, educationalists have challenged the OECD’s ‘global’ approach to conceptualising wellbeing; suggesting wellbeing is deeply entrenched in different cultural ontologies and what it means to live ‘the best life’ (Rappleye et al., 2019). Given Heller-Sahlgren’s ‘trade-off’ report uses PISA data, it serves as a useful springboard for discussion of the theoretical assumptions underpinning transnational comparisons of wellbeing and achievement on the whole: provided in this section.

Children are susceptible to macro-level societal and cultural influences (Bronfenbrenner, 1979). Thus, research on children’s wellbeing requires an in-depth understanding of the cultures in which children are embedded, including how different countries have distinct conceptions of wellbeing and achievement, respectively. While research has documented some ‘core similarities’ in children’s wellbeing experiences across diverse countries (Kim and Main, 2017; Newland et al., 2019), ‘unexplained’ differences have also been found (Casas and Rees, 2015; Dinisman and Ben-Arieh, 2016) requiring further investigation. Further research is crucial, given the widespread use of transnational data by policymakers in arriving at important decisions in their respective countries. To accommodate more culturally grounded investigations, the methodology employed in transnational comparisons could be expanded – for example, into one which combines ‘etic’ and ‘emic’ perspectives – to elucidate context-specific interpretations of children’s wellbeing.

Transnational comparisons utilise an ‘etic’ approach which conceptualises wellbeing in the same way – that is, aims to measure the same underlying construct – across diverse nations, using translated language (Dinisman and Ben-Arieh, 2016; Kirkcaldy et al., 2004; OECD, 2009, 2019). Prominent psychologists of children’s wellbeing have however underlined the need for culturally grounded conceptualisations (McLellan, 2019), suggesting ‘emic’ perspectives can be combined with derived etic approaches. Emic approaches investigate children’s wellbeing as situated within children’s distinct contexts with the aim of reconceptualising the wellbeing construct; developing culturally sensitive measures and knowledge (see, for example, Camfield and Tafere, 2011; Chen et al., 2009; Gao and McLellan, 2018; Hajdukova et al., 2017; Ho and Cheung, 2006).

The benefits of an etic approach are that research is logistically easier to manage, cost-effective and affords cross-cultural comparisons to be made straightforwardly. While the emic approach – theorised by cultural anthropologists (Berry, 1969) – affords more in-depth, culturally sensitive investigations. Yet, the emic approach could be combined with the broader etic approach. Work situated within different learning cultures adopting an emic approach would allow more meaningful interpretations of the findings derived from etic transnational studies. For example, using qualitative enquiry to unpick and fully understand differences in childhoods across different cultures (Rees and Dinisman, 2015).

When it comes to researchers investigating the wellbeing-achievement relationship specifically, such a combined approach is vital. Linked to this, eudaimonic wellbeing is not yet incorporated as a sub-construct of children’s wellbeing in transnational comparisons, and, given eudaimonia is inherently bound up with achievement, this deserves more attention. Context-specific exploration of what wellbeing and achievement constitute in different cultures may reveal important overlaps between children’s experiences of eudaimonic wellbeing and academic striving.

Theoretically, Montt and Borgonovi (2018) explore the idea of drawing more culturally aware interpretations from the findings of transnational comparisons. Their study uses the same PISA 2012 data as Heller-Sahlgren to investigate the relationship between achievement and sense of belonging at school across education systems. The basic argument espoused is that education systems improve academic outcomes dependent on (a) assets and (b) effects. Assets refer to the composition of an education system (e.g. a highly skilled teaching workforce, or fewer pupils from low socioeconomic (SES) backgrounds), whereas effects pertain to the way in which assets are used (e.g. child-led pedagogy, which may be better utilised by teachers who are subject experts). Using imputation models to discover how particular countries could ‘import’ educational approaches from other countries, they find countries promoting high levels of achievement tend not to simultaneously facilitate a high sense of belonging at school; supporting Heller-Sahlgren’s findings. The difference being that while Montt and Borgonovi clearly warn against a non-nuanced approach which uses transnational data without giving due consideration to the distinct education systems and their assets/effects, Heller-Sahlgren adopts an etic model without acknowledging differences between educational ecosystems.

In Figures 1 and 2 of Heller-Sahlgren’s (2018: 16–17) report there are outlier countries – as per the OECD’s warning – who do not follow the line of best fit. If we look at the relationship modelled between pupil happiness and pupil-led teaching methods, it is not the case that a negative relationship between pupil happiness and teacher-led pedagogy exists in all countries. Japan, for instance, ranks high for pupil happiness, while having more teacher-led pedagogy, whereas the United States ranks lower for pupil happiness yet adopts more pupil-led pedagogy. Likewise, in the model depicting achievement and pupil-led teaching methods, the relationship purported – that higher achievement is positively correlated with teacher-led pedagogy – does not represent all countries. Pupils in Greece, Slovenia and Luxembourg do not achieve highly, yet these countries do adopt more teacher-led pedagogy. Therefore, the etic approach adopted risks over-generalising trends across countries of diverse compositions, providing a macro-level overview without considering cultural differences. For instance, included in the sample are Finland and Japan, documented as having significantly different learning cultures (Crehan, 2016).

Another question posed to researchers conducting transnational wellbeing-achievement comparisons is the following: what if the very idea of a ‘trade-off’ – of prioritising achievement over wellbeing or vice versa – is at odds with the entrenched cultural ideals of countries? Rappleye et al. (2019) discuss such issues in depth, challenging the in-built assumptions of the OECD’s wellbeing measures. In agreement with Rappleye et al., I argue that researching the wellbeing-achievement relationship across nations implicitly promulgates four contestable assumptions.

One, researchers assume wellbeing is ubiquitous and valued across all cultures. That is, the fact that Western researchers are interested in the relationship between wellbeing and achievement may be symbolic of values held in Western culture. The preoccupation with children’s wellbeing seen over the past decade in the West may not necessarily be shared elsewhere.

Two, if wellbeing is valued across cultures, researchers assume all countries consider achievement as equally important to wellbeing for children’s development. Rappleye et al. touch on this; noting the in-built bias of the OECD’s achievement motivation measures; assuming all cultures intrinsically value achievement over other aspects of wellbeing. This issue is particularly thorny, because if researchers consider children’s wellbeing an issue of social justice, global values of what it means for children to be ‘well’ may directly conflict with local values (Ben-Arieh and Frønes, 2007). This places researchers in a position of potential conflict when considering children’s rights.

Three, that the wellbeing construct measured across cultures is the same. Researchers often believe (consciously or subconsciously) that their perspective on reality is the only rational version of reality (Rogoff, 1991). Thus, Western researchers must be aware of the potential focalisation of Westernised ideals of what it means to live a ‘good life’, which could obscure ‘global’ conceptualisations of children’s wellbeing. Cross-cultural research reviewing conceptions of happiness across North America and East Asia clearly demonstrates there are major differences in what it means to be happy in these contexts (Uchida et al., 2004). Western conceptions of wellbeing are intrinsically wrapped-up in ideas of the self and individual experience, which presents a considerable issue when it comes to cross-cultural comparisons, since the self is context sensitive (Suh, 2007). Generally speaking, Western cultures are known to be more individualistic in their practices and overall world view, whereas Southern cultures are typically more collectivist in their orientation (Tian et al., 2017). Weisner (1984) distinguishes between individuation and individualism during child development; the former being a universal process of understanding how the self is separate to others, and the latter, pertaining to the idea of the self being autonomous, contained inside a private mind looking out to the world, which may not be universal (Shweder and Bourne, 1982). According to Weisner, individualism may not involve the same egoism, self-aggrandisement or concerns about the self (seen in Western cultures), elsewhere in the world. Such cultural distinctions of ‘self’ are likely to impact what it means to be well. Therefore, given wellbeing is relative to culture, this necessitates transnational research that is conducted by diverse research teams with lived experiences of those cultures. Moreover, etic models of children’s wellbeing should be combined with derived emic approaches; allowing researchers to develop culturally grounded conceptions of wellbeing (McLellan, 2019; Smith et al., 2013).

Finally, a fourth assumption made is that wellbeing and achievement are two separate constructs across cultures and that they are not one and the same. In certain countries, achievement is embedded in ideas about what it means to be ‘well’ and live ‘the good life’. This has ramifications for how wellbeing is conceptualised, relative to one’s achievement, rather than separate from. Since the two sub-constructs of wellbeing – hedonic and eudaimonic – have not both been fully harnessed in research, cross-cultural differences in eudaimonic wellbeing are not well-understood. Some cultures may emphasise hard work, intellectual accomplishment and knowledge acquisition as the route to self-actualisation, whereas other cultures may favour the pursuit of close interpersonal connectedness as being self-actualised. Understanding of these differences should inform interpretations of transnational comparisons.

These theoretical assumptions represent ‘question marks’ over how transnational comparisons of children’s wellbeing should be interpreted by policymakers. Certainly, policymakers must consider how evidence from transnational comparisons is treated in decision-making. While transnational comparisons are a useful tool for understanding comparative national performance, the breadth of the evidence derived from PISA – characteristic of an etic approach – should not be mistaken for country-specific depth. That is to say, making policy decisions based on transnational evidence alone without the support of additional context-specific evidence would be misguided. Therefore, using transnational data to justify a wellbeing-achievement trade-off appears dangerous. To fully understand the relationship, context is essential. For the purpose of evaluating the evidence supporting a wellbeing-achievement trade-off, I temporarily ‘park’ these assumptions. In the next section, I argue that rather than children’s wellbeing and achievement being incompatible educational goals, the relationship is underexplored.

Wellbeing and achievement: Incompatible or underexplored?

Heller-Sahlgren’s report foregrounds a timely and necessary discussion about the wellbeing-achievement relationship, yet overlooks important research meaning readers are not presented with the full picture. On wellbeing and achievement, Heller-Sahlgren (2018a) writes of their incompatibility:

It would certainly be convenient if performance and happiness were self-reinforcing goals . . . Yet there is little rigorous research presented in favour of either the theory or the prescription. (p. 3)

On the contrary, drawing from valuable research neglected by Heller-Sahlgren, I argue there is a potential positive wellbeing-achievement relationship. In this section, I illustrate the picture presented by research to date which contradicts the idea of a trade-off. There are, however, methodological limitations impeding our understanding. First, there is a paucity of research operationalising achievement objectively (i.e. attainment in standardised assessments/national-level examinations). Much research instead uses pupils’ academic self-concept, self-confidence or self-reported achievement. A second limitation pertains to the measurement of wellbeing; with research using global measures of life satisfaction only, or non-multidimensional instruments. A multidimensional approach would better enable researchers to unearth specific aspects of wellbeing that matter for pupils’ achievement.

Reviewing the literature suggests that the nature of the relationship between children’s wellbeing and their achievement is nuanced. Understanding this nuance requires careful piecing together of the evidence from a body of research wherein researchers have different operationalisations of the wellbeing and achievement constructs. However, regardless of conceptual differences in how wellbeing (or ‘happiness’, ‘life satisfaction’, ‘SWB’, ‘need satisfaction’) and academic achievement (i.e. objective achievement, self-reported achievement, self-reported academic competence), the overall picture suggests a positive wellbeing-achievement relationship. Rather than being incompatible goals therefore, I conclude by demonstrating how the wellbeing-achievement relationship is underexplored in a number of ways. Namely, gaps in existing knowledge include (a) an understanding of developmental differences and (b) mediational variables explaining how pupils’ wellbeing and achievement is related.

Hedonic wellbeing, achievement and associated academic characteristics: Pieces of a jigsaw

I begin piecing together the evidence concentrating first on the relationship between children’s hedonic wellbeing (i.e. their life satisfaction, extent of positive and negative emotions) and their achievement. Although early research demonstrates no significant relationship between pupils’ academic achievement and life satisfaction (Huebner, 1991; Huebner and Alderman, 1993), more recent studies demonstrate a positive relationship. For example, Proctor et al. (2010) found that adolescents with higher self-reported life satisfaction than their counterparts also had significantly higher self-reported academic achievement (Proctor et al., 2010). Life satisfaction is conceptualised as a sub-measure of hedonic wellbeing in the literature, but hedonic wellbeing also includes self-reported happiness. Yet, pupils reporting higher levels of happiness are also documented as achieving more highly than their peers with lower levels of happiness. For instance, in a longitudinal research project conducted in England (Smith et al., 2019), pupil happiness was measured using a single item and findings suggested that pupils aged 11–14 who self-reported as significantly happier than their peers achieved significantly higher GCSE grades. Almost half of children who were unhappy with their life ‘as a whole’ did not achieve five A*–Cs, compared to a third of their happier counterparts. This positive happiness-achievement relationship is further supported by the findings of a transnational study (Kirkcaldy et al., 2004) revealing higher pupil happiness was significantly positively associated with higher academic achievement. ⁴ Although these studies measure happiness only, Suldo et al. (2011) measured pupils hedonic and eudaimonic wellbeing in a longitudinal study which suggested that both forms of wellbeing may be important for achievement. Suldo and colleagues found that pupils with higher hedonic and eudaimonic wellbeing at time one were significantly more likely to obtain increased Grade Point Averages (GPA) at time two, 1 year later. Altogether, research suggests that children with higher hedonic wellbeing achieve higher grades than children with lower hedonic wellbeing.

Higher hedonic wellbeing has also been positively linked to other important academic characteristics including growth mindset and holding mastery-approach goals (an adaptive form of motivational goals whereby learners strive to improve their understanding and skills, rather than performance- or avoidant-approach goals, which can be maladaptive) (Cheng and Furnham, 2002; King, 2017; Wormington and Linnenbrink-Garcia, 2017). These academic characteristics have in turn been linked to higher achievement (Blackwell et al., 2007; Cury et al., 2006). Other important constructs associated with objective academic achievement ⁵ have also been documented as positively linked to pupils’ wellbeing, such as pupils’ academic self-concept (Gilman and Huebner, 2006; Suldo and Huebner, 2006), self-confidence (Cheng and Furnham, 2002) and self-efficacy (OECD, 2019).

Specifically, a recent PISA report (OECD, 2019) measuring pupils’ life satisfaction, happiness and self-efficacy found each was differently related to pupils’ achievement. The relationship between pupils’ reading achievement and life satisfaction was negative; the relationship between pupils’ reading achievement and their happiness was curvilinear; yet, the relationship between pupils’ reading achievement and self-efficacy was positive. Transnationally, therefore, self-efficacy appears most important for reading achievement; notwithstanding the issues associated with transnational comparisons discussed in the section ‘Conceptualising and measuring children’s wellbeing’. National-level research in the UK, however, also suggests both eudaimonic and hedonic wellbeing are positively associated with the engagement and frequency of pupils’ reading. Pupils who enjoy reading, read daily and think of themselves as very good readers report higher life satisfaction and self-belief, and better coping skills than peers who do not enjoy reading at all, never/rarely read and think of themselves as below average readers (Clark and Teravainen-Goff, 2018). Self-efficacy can be understood as a form of eudaimonic wellbeing, suggesting there are meaningful links between pupils’ eudaimonic wellbeing and their academic achievement (further elucidated in the following section).

Finally, evidence demonstrating clear links between how children feel about their academic abilities and their wellbeing more generally is reinforced by a study conducted by Greenspoon and Saklofske (2001) measuring pupils’ hedonic wellbeing in its entirety (life satisfaction, positive and negative affect). Greenspoon and Saklofske found that pupils with lower academic self-concept experienced decreased wellbeing. Collectively, these studies clearly document a positive association between pupils’ hedonic experiences of wellbeing and feelings of academic self-competence and confidence. Altogether, the research demonstrates pupils who feel positive also have positive academic self-regard, which is in turn linked to higher achievement. Learners with higher life satisfaction at school have significantly more positive academic experiences, while learners with lower wellbeing are at greater risk of school drop-out and behavioural difficulties (Arslan and Renshaw, 2018).

Eudaimonic wellbeing and achievement: Contributions from Self-Determination Theory (SDT)

The next piece of evidence I turn to is research underlining a link between pupils’ eudaimonic wellbeing and their achievement. Operationalised non-objectively, achievement (i.e. competence) can be conceptualised as an innate psychological need, according to SDT (Deci and Ryan, 2000). SDT suggests individuals have three core psychological needs (competence, autonomy and relatedness) that, when fulfilled, result in high wellbeing. The fulfilment of these needs is documented as important for both children’s wellbeing and academic achievement (Boncquet et al., 2020). Pupils’ experiencing greater need satisfaction are better adjusted at school (and life) and report higher achievement (Diseth et al., 2012). Research exploring SDT through measures of ‘need satisfaction’ (the extent to which pupils experience autonomy, competence and relatedness), supports a positive relationship between pupils’ eudaimonic wellbeing and their academic performance. Specifically, a study conducted by Miserandino (1996) found that pupils’ perceived competence and autonomy predicted changes in their achieved grades from the beginning to the end of one year.

The extent to which pupils need to feel autonomous is met in their learning environment has also been demonstrated as predictive of their academic motivation and wellbeing (Chirkov and Ryan, 2001). Competence and autonomy also appear positively linked to pupils’ self-esteem (Yamauchi and Tanaka, 1998). Another innate need of SDT is relatedness – or ‘social wellbeing’ – which research suggests is essential for pupils’ achievement (Dix et al., 2020). Children who belong to peer groups wherein bullying is prevalent not only have significantly lower SWB, but also lower achievement, compared to children in positive peer groups (Gutman and Feinstein, 2008). Understanding children’s experiences of interpersonal connectedness is an area requiring particular attention in the UK. This is because research has long documented the enduring issue of children struggling to have the psychological need for relatedness met (Office for National Statistics, 2018a, 2018b; Rees et al., 2012; UNICEF, 2007a). Children’s positive feelings about their friendships also showed a significant decrease in the recent Good Childhood Report (The Children’s Society, 2019). Schools vicariously absorb much responsibility for children’s wellbeing, especially when it comes to their socialisation and sense of connectedness.

Since children spend the majority of their lives at school, educational goals should include both the academic and introspective; supporting children to live fulfilled lives. In sum, research conducted in alignment with SDT documents a positive relationship between pupils’ eudaimonic wellbeing and achievement.

Developmental differences in the wellbeing-achievement relationship

Research exploring the wellbeing-achievement relationship across Primary- and Secondary-aged pupils presents mixed evidence. Some studies suggest there is no significant relationship between wellbeing and achievement for Primary pupils, whereas others suggest there is. Once again, discrepancies may be explained by differences in how researchers conceptualise wellbeing and achievement. For Secondary pupils, evidence generally indicates a positive relationship between eudaimonic wellbeing and achievement. Girls and Primary pupils experience higher school wellbeing overall. Most research reviewed exploring age differences uses a multidimensional wellbeing approach, providing rich evidence.

Major research conducted in England measuring children’s multidimensional wellbeing (Gutman and Vorhaus, 2012) found significant differences between younger and older pupils pertaining to specific wellbeing sub-constructs that are important for achievement. Higher emotional wellbeing (aged 7) significantly predicted higher achievement from KS1 to KS2, whereas higher school wellbeing (aged 13) significantly predicted higher achievement from KS3 to KS4. These are important developmental differences in terms of pupils’ hedonic and eudaimonic wellbeing. Emotional (hedonic) wellbeing (e.g. feelings of anxiety, sadness, happiness) was important for younger pupils’ achievement, but not older pupils’, whereas school (eudaimonic) wellbeing (e.g. enjoying school, feeling motivated to complete schoolwork, engaged at school) was important for older pupils’ achievement, but not for younger pupils.

The importance of hedonic wellbeing for younger pupils’ achievement prevails in research elsewhere. For example, van Batenburg-Eddes and Jolles (2013) found lower emotional wellbeing (e.g. feeling less good about life) was associated with an increased risk of academic underachievement in 12-year-olds. ⁶ Moreover, in England, research with 11-year-olds demonstrated that positive emotions (towards school specifically) also significantly predicted achievement (Riglin et al., 2013). ⁷ These findings are consistent with researchers’ theorising a significant association between pupils’ positive feelings of school connectedness or ‘belonging’, their wellbeing and achievement (McLaughlin and Clarke, 2010; Roeser et al., 2000).

If hedonic wellbeing is more important for Primary pupils’ achievement, one would expect children’s own values to corroborate this, which is what research conducted in Spain by López-Pérez et al. (2016) found. Children valued more hedonic conceptualisations of wellbeing at school (positive feelings) whereas adolescents valued more eudaimonic forms (achievement). Interestingly, however, later work conducted by López-Pérez and Fernández-Castilla (2018) discovered that eudaimonic wellbeing played an important role in relation to Primary pupils’ objective achievement; pupils who conceptualised happiness as ‘learning’ had significantly higher achievement, but not happiness. Whereas, Secondary pupils conceptualising happiness as ‘being with friends’, ‘being praised’, ‘helping’ and ‘having no leisure time’ had higher achievement and happiness. ⁸ Internationally, the picture is not therefore straightforward. There are differences between younger and older pupils’ experiences of eudaimonic wellbeing in terms of what it constitutes (learning/being with friends/helping/lack of leisure time), and whether these experiences are positively associated with achievement.

The importance of pupils’ objectively achieved grades may be more important for Secondary pupils’ wellbeing than Primary. Research conducted in Italy (Tobia et al., 2019) revealed no significant relationship between Primary pupils’ wellbeing and their grades, yet Secondary pupils’ wellbeing was significantly positively associated with their grades. However, Primary pupils had higher wellbeing overall than Secondary: also found in England. ⁹ Tobia et al. further document large declines comparing Secondary to Primary pupils’ ‘school wellbeing’ (gratification obtained from school grades); suggesting there are different concurrent predictors for Primary compared to Secondary pupils’ wellbeing. Specifically, objective achievement appears important for Secondary but not for Primary pupils’ wellbeing. Primary pupils on the other hand felt significantly more positive at school when concurrently feeling they were doing well academically. This is supported by research conducted with Primary pupils in England (Clark and Teravainen-Goff, 2018). Elsewhere however, research conducted in Northern Ireland contradicts these findings. Miller et al. (2013) found Primary pupils’ with greater wellbeing had significantly higher achievement, controlling for the effects of gender and deprivation. ¹⁰ However, Miller et al. operationalised wellbeing as a single amalgamated score, meaning the unique contributions of different sub-constructs (feelings of competence or self-worth) are potentially being masked.

Summarising the developmental differences

Proportionate attention should be paid to different aspects of wellbeing that are important for children of different ages. Altogether, existing research suggests that during the developmental window of late Primary to early Secondary school (ages 7–12), hedonic wellbeing is particularly important for pupils’ achievement. When it comes to Secondary pupils, eudaimonic wellbeing appears important for achievement. Despite this need for Secondary pupils’ to experience greater eudaimonic wellbeing than Primary pupils, cross-sectional research in England observes eudaimonic wellbeing increases from Years 3 to 6 (Primary) but steadily decreases at Secondary, with Year 10 pupils experiencing the lowest levels (McLellan and Steward, 2015). Further research is needed to understand developmental differences, which could be explained by a variety of biopsychosocial factors. Namely, as children become adolescents they experience biological, cognitive and social changes (Gutman et al., 2010). Likely explanations include, therefore, the unique period of adolescence when self-concept is in flux and decreased positive affect can occur (Moneta et al., 2017), particularly in the context of school as adolescents juggle academic, peer and extra-curricular aspects of their identities (Bosma and Jackson, 1990). Despite Primary pupils being earlier in their school careers however, they are still under pressure to perform in high-stakes exams (McLellan et al., 2012; Stobart, 2009). Therefore, it is conceivable that even Primary pupils’ wellbeing can be affected by academic pressures.

Existing evidence and unexplored land

It is important to take stock of the evidence and the ‘unknowns’ in the wellbeing-achievement relationship. Heller-Sahlgren’s report suggests the wellbeing-achievement relationship is straightforward: aside from the demographic variables controlled for ¹¹ ‘type of pedagogy’ is the only explanatory factor. A wider understanding of the home and school environments in which pupils are embedded, and how they are meeting children’s psychological needs, is required.

There is a tension between the psychological need for autonomy (purported in SDT) and Heller-Sahlgren’s idea that child-led pedagogy is conducive to children’s wellbeing, but not achievement. Research exploring Secondary pupils’ experiences of wellbeing in school has proposed that pupils’ perceived opportunities for playfulness in lessons has a significant indirect effect on their life satisfaction through interpersonal wellbeing (Clarke and Basilio, 2018). Thus, playful learning environments facilitating interactions with others may be important for fostering pupils’ wellbeing. In England, findings from The Children’s Society (2019) also suggest that while children’s feelings about school are in decline, their feelings about schoolwork are stable: highlighting the distinctness of school climate. Research conducted by Steinmayr et al. (2018) highlights a wider range of factors (including pupils’ mood, life satisfaction, self-efficacy, test anxiety and school climate) that are significant predictors of achievement. A range of contributory factors, including school climate as a whole rather than pedagogy alone, should therefore be accounted for in research. This is supported by existing evidence demonstrating school environments that foster kindness or adopt a whole-school socio-emotional ethos are associated with higher academic engagement and achievement (Banerjee et al., 2014; Datu and Park, 2019). Futhermore, a recent review of school-based wellbeing programs found they had a small to moderate positive impact on pupils’ academic achievemen in both literacy and Maths, compared to pupils in control groups engaged in their typical activities with general academic performance (Dix et al., 2020).

There are four main gaps in existing knowledge; owing either to methodological limitations or oversights. First, only recently have researchers begun conceptualising and measuring wellbeing as a multidimensional construct; this is paramount to fully understanding pupils’ experiences. A second related gap is the potential differences between younger and older pupils and the specific aspects of wellbeing that are critical for achievement during different developmental windows. Understanding age differences would allow teachers, CAMHS and parents to focus on particular aspects of children’s wellbeing at specific ages. Third, since achievement in national exams is the ultimate currency of many societies, more studies operationalising achievement as objective grades are needed (Brown and Eklöf, 2018). Qualitative research directly engaging children is further required to understand the role achievement plays in pupils’ school wellbeing experiences. Pupils’ eudaimonic wellbeing in particular is underexplored (McLellan, 2019), especially in relation to academic pressures. Fourth, research indicates certain sub-constructs of wellbeing play a pivotal role in achievement, but we have little understanding of how or why.

Taking the wellbeing-achievement relationship forward for more in-depth scrutiny in future research is imperative for elucidating an understanding of mediating psychological mechanisms that underpin it. Some unanswered questions being: Why do children with higher eudaimonic wellbeing achieve higher grades? What are the mediating or moderating mechanisms by which pupils’ wellbeing is associated with achievement? What are the underlying psychological processes that characterise pupils with higher wellbeing and achievement? The field of motivational psychology represents fruitful yet unexplored land for future researchers. Theory of Self (Dweck, 1999; Dweck and Leggett, 1988) offers a cognition-affect-behaviour model suggesting individuals’ ‘implicit theories’ (deeply held beliefs) have important implications for their psychological functioning (Dweck et al., 1995). According to the model, children’s wellbeing and achievement are likely to be associated with the mindsets they adopt and learning goals they endorse (Duckworth et al., 2012; Howell, 2016; Yeager and Dweck, 2012). Children’s Theory of Self internally manifests as one of two types of mind-set: a fixed mindset (whereby, children believe they are made up of fixed traits), or a growth mindset (whereby, children believe that they are made up of changeable traits that they are capable of improving). Research demonstrates that the two belief patterns – the ‘helpless’ fixed mindset and the ‘flexible’ growth mindset – impact children’s wellbeing (Schleider et al., 2015). ¹³ Mindsets and goals could, therefore, serve as mediators in the wellbeing-achievement relationship. Particular learning cultures may foster mindsets that are harmful for pupils’ wellbeing (e.g. if I do not pass my exams, I am a failure) compared to cultures cultivating growth mindsets (e.g. all I can do is try my best).

When plugging these gaps, researchers should take into account national-, home- and school-level influences on children’s wellbeing and achievement. School climates constitute (but are not limited to): approaches to learning and assessment, pedagogy, accountability measures, interpersonal relationships, commitment to socio-emotional learning and school connectedness. Existing evidence demonstrates children’s wellbeing is affected by a range of determinants including parental stress (Powdthavee and Vignoles, 2008), the quality of their friendships (Gutman and Feinstein, 2008) and educational assets deployed within countries (Montt and Borgonovi, 2018). There is no way of fully understanding children’s wellbeing in relation to their achievement without a broader awareness of the respective school, home and societal ecosystems in which they are nested. Individual-level factors also impact pupils’ wellbeing and achievement including Special Educational Needs, SES, child-parent relationships, gender (Gutman et al., 2010) and non-cognitive skills (Gutman and Shoon, 2013). These factors should not be overlooked by researchers; there is a clear need for sophisticated research designs and methodologies that accommodate the complexities of the wellbeing-achievement relationship. It would be wise for researchers to consider all these factors before planting the idea that policymakers could discard children’s wellbeing in favour of prioritising achievement. Such ideas – should they be acted upon – have the potential to cause serious, long-term damage. The young people most at risk of being negatively impacted by such a ‘trade-off’ are those from disadvantaged or low socio-economic households; a group pinpointed as having significantly lower wellbeing and achievement (OECD, 2019; The Children’s Society, 2020).

Conclusion

Heller-Sahlgren’s report perpetuates much-needed conversations about the goals of education. I have argued however, that coining the wellbeing-achievement relationship as a ‘trade-off’ is negligent. Importantly, the false dichotomy pitched by Heller-Sahlgren represents a wider current of thinking in education whereby researchers pit achievement goals against the goal of wellbeing. This kind of discourse characterises pupils’ wellbeing as a rivalrous force in direct opposition to academic achievement. On the contrary, in this article, I have refuted, with the support of research from psychology, the need for such a trade-off between children’s wellbeing and achievement. Rather, evidence suggests a positive relationship, but this is not straightforward and calls for nuance; requiring researchers to contemplate different aspects of children’s wellbeing (hedonic and eudaimonic).

Recent UK research indicates that performativity cultures may be harmful for pupils’ wellbeing (Roome and Soan, 2019) and internationally, the increasingly competitive education systems observed elsewhere are likely to exhibit similar problems (Chen et al., 2009; Wang et al., 2015). Given the current context – that is, the humongous value society attaches to children’s academic achievement, and the simultaneous experiences of young people reporting low levels of wellbeing and mental health – I urge all those in the field to adopt a more balanced view that takes both of these realities into account. In the UK, our education system must avoid blindly promoting academic achievement goals without recognising how pupils’ accomplishment of these goals is empirically linked to their wellbeing. Reimagining schools as places that can foster pupils’ wellbeing and learning should be our default position, rather than assuming one must trump the other.

To conclude, Heller-Sahlgren’s proposition that positive pupil wellbeing and high achievement cannot co-exist in schools, and that they are goals in direct opposition, is not warranted by evidence. Furthermore, imagining an education system with the single agenda of knowledge acquisition is a reductionist vision. Not only would such a trade-off be harmful to the wellbeing of future citizens, it underestimates the transformative power of education to prepare children to live informed and worthwhile, emotionally textured, fulfilled lives. When making policy recommendations researchers should avoid ‘all or nothing’ thinking which lures governments into false dichotomies.