The relationship between subjective well-being in school and children ’ s participation rights: International evidence from the Children ’ s Worlds survey

This paper considers the relationship between children ’ s subjective well-being at school and the fulfilment of their participation rights. Our research focuses on the association between children ’ s involvement in decision-making in school and their subjective well-being (SWB) using international evidence from the Children ’ s Worlds survey. The analysis uses data from the third wave of the Children ’ s Worlds survey collected from 12-year-olds within 13 EU and former EU countries. We focus on the school in response for calls for more domain-specific analysis of children ’ s SWB, and in light of the considerable amount of time children spend in school and the compulsory nature of schooling. We identify an association between subjective well-being and participation in decision-making at school, across all countries included in this study, except Malta. The paper also finds that relationships with teachers and other peers (as measured by bullying) also impact upon children ’ s school-based SWB.


Introduction
While much research has been conducted on both children's participation at school and their well-being at school, very little research has considered the relationship between the two (Casas et al., 2013;Lloyd and Emerson, 2017;Kutsar et al., 2019).This is because these literatures tend to be related to different fields, with human rights scholars and education researchers primarily focusing on the former, and the latter mainly emanating from the psychological literature on the mental health of school-children (Lundy, 2014).This research aims to respond to calls (Lundy, 2014) for further connections between the two by considering how children's subjective well-being (SWB) at school is related to their perception that they are able to participate in decisionmaking about issues that are of importance to them.It does this using international evidence from the Children's Worlds survey, to consider the role of culture in the relationship between the two.

Children's well-being
Well-being is recognised as a complex and 'multifaceted' concept (Ben-Arieh et al., 2014;Casas & Frønes, 2020).In the field of social and political science, it was first used to provide a broader and more nuanced picture of 'standards of living' than Gross Domestic Product (GDP) (Casas & Frones, 2020).Thus, the concept can include a wide-range of indicators, including objective measures such as family income and access to material goods, as well as subjective indicators such as life satisfaction and happiness (Camfield, Streuli & Woodhead, 2009).Early research on children's well-being focused almost exclusively on 'objective' measures of well-being, which cast children as 'passive objects who are acted on by the adult world' (Ben-Arieh, 2008, p.7).However, more recently, there has been a growing focus on children's subjective wellbeing, which recognises that their own experiences should be considered alongside 'objective' measures, such as family income and deprivation (Casas and Frønes, 2020).In addition, the insufficiency of models of subjective well-being designed for adults have been highlighted, leading to attempts to produce measures designed for children, which are contextualised within particular domains (Strózik et al., 2016), and based and rooted in their own lives and experiences (Long et al. 2012).This has also been sparked by a move away from a deficit-based approach to childhood (Suldo et al. 2006), and an attempt to improve children's well-being, as opposed to focusing on their 'well-becoming', which considers how children are progressing towards being successful and useful adults (Bradshaw et al. 2013;Huebner et al. 2014).
Models of subjective well-being tend to include two or three key components: most use an evaluative measure such as happiness or life satisfaction, and experiential measure based on positive and negative affect, with some also using an eudemonic measure of 'achieving rewards in life independent of pleasure' (Bradshaw et al. 2013, 622).Many studies draw on the model of child subjective well-being developed by Huebner (1991) and Huebner & Dew (1996) in the early 1990s.These are based on the tripartite model of subjective well-being developed by Diener (1984), which includes life satisfaction, positive affect and lack of negative affect.
The mediating role of culture and place in the construction of subjective well-being have also been considered (White, 2008;Bradshaw et al., 2011) alongside personality (Diener, 2000).The potential impact of cultural and environmental factors on well-being is complex; a key factor may be people's perception of their circumstances relative to their peers (Bradshaw et al., 2011;White, 2008).This may have more or less influence than absolute differences in life circumstances.Drawing on evidence around the contribution of personality and environmental factors to subjective wellbeing, Bradshaw et al. (2011) suggest that wellbeing is 'a result of the dynamic interactions between the top-down (e.g.personality traits) and bottom-up (objective circumstances) factors, and these interactions (or processes) vary according to a specific time and place' (p.549).

Relationship between SWB & rights
Connections are often made between well-being and children's rights (Camfield et al., 2009;Laevers and Declercq, 2018), and it has been suggested that the United Nations Convention on the Rights of the Child (UNCRC) offers a 'normative framework' for understanding the wellbeing of children (Ben-Arieh, 2008, p.5) in a way that positions children as agentic and competent beings.Bradshaw et al. (2007) argue that 'well-being can be defined as the realisation of children's rights and the fulfilment of the opportunity for every child to be all she or he can be' (p.135).However, while there are a number of commonalities between rights and well-being, they are distinct concepts (Lundy, 2014;Tisdall, 2015).
While the well-being literature draws heavily on the language of needs, rights are entitlements which place legal obligations on states to deliver for children in a wide range of substantive areas (Lundy, 2014).However, as states are legally accountable for realising them, children's rights can only include what states are able to deliver.Thus, while the fulfilment of rights may enable the conditions under which happiness and well-being can flourish, concepts such as happiness and love cannot be included in the concept of rights (Ibid).As a result, well-being is often discussed in terms of 'ideals', while rights may be seen as limited, as they set only minimum standards (Lundy, 2014;Tisdall, 2015).However, rights do go further than promising children basic standards of livingthey place a duty on states to ensure that children's civil and political rights are protected, and that their right to participation in all areas that affect them is asserted under article 12 (Lundy, 2014;Quennerstedt, 2010).Further, a crucial aspect of rights is that they are empowering and that they provide children with a legal entitlement: without rights, children and their advocates could not make demands of the State, and would instead be dependent on the goodwill of sympathetic parties to have their interests met (Freeman, 2007;Lundy, 2014).
Participation rights are complex and difficult to measure, partly because of the cross-cutting nature of the Convention on the Rights of the Child (Lundy, 2014).Studies which aim to measure children's participation rights often only consider children's 'participation in activities rather than in decision-making about their own lives' (Lundy, 2014: p. 83).Further, research on participation often neglects to consider the interdependence and indivisibility of the rights in the Convention (Lundy, 2007).According to the framework for the fulfilment of participation rights developed by Lundy (2007), the right to participation is only realised when the right to information (article 13) the right to non-discrimination (article 2), the right to have decisions made in the child's best interests (article 3) and the right to guidance from adults (article 5) are also fulfilled.As Lundy (2007) reminds us, 'voice is not enough': children's participation rights are not fulfilled simply by providing them with opportunities to speak.Their views must also be given due weight, so that they are 'listened to and acted upon as appropriate' (Lundy, 2007: p. 933).
Previous research has found that knowledge of child rights or the Convention on the Rights of the Child have weak associations with SWB (Kosher & Ben-Arieh, 2017;Casas et al., 2018).However, perceiving that adults respect rights and listen to them has a much greater effect, which may be because relationships and interaction between children and adults is more important than knowledge of rights (Casas et al., 2018), and also that knowledge of rights does not mean that rights are being fulfilled.In fact, children who are aware of their rights might be better placed to recognise violations of them, potentially lowering their SWB.

Factors related to children's subjective well-being
A number of factors relating to children's subjective well-being have been identified in the literature.In general, boys report higher levels of subjective well-being than girls (Bradshaw et al., 2011;Kaye-Tzadok et al., 2017).Age is also an important factor, with research indicating that subjective well-being tends to decline by age (Casas and González-Carrasco, 2019;Bradshaw & Rees, 2017).In addition, material deprivation is associated with lower levels of subjective well-being across nations, though this relationship is not straightforward and the relative importance of this varies by country (Bradshaw et al., 2011;Bradshaw & Rees, 2017).Furthermore, research has indicated that while there is not a significant association between country-level material resources (e.g.GDP), there is an association between SWB and children's access to child-specific material resources (Main et al., 2019).
Similarly, Lee and Yoo (2017) found that freedom to choose and sense of self were the most important factors related to global life satisfaction in their analysis of 12 year-olds from 14 countries participating in the 2013-2016 international Children's Worlds survey.Whereas, happiness with amount of choice in life was the second strongest association with overall subjective well-being in Rees et al.'s (2010) study.

Role of school satisfaction in children's subjective well-being
A number of studies have considered the contribution that school satisfaction makes to children's overall subjective well-being, with mixed findings.Lawler et al. (2015;2017) found that school satisfaction was the second most important factor for predicting life satisfaction (after family relationships) for both 10-year-olds (2017) and 12-yearolds (2015).However, in a study in the UK, Rees et al. (2010) found that school and schoolwork were one of the aspects of children's lives that were least strongly associated with SWB.Small effects have been found in other studies as well: in a US study, aspects of school climate relating to children's social experiences at school explained 14% of the variance in global life satisfaction (Suldo et al., 2008), and in an international study analysing pilot study data from the International Survey Of Children's Well Being (ISCWeB), Lee & Yoo (2015) find that 9% of the variation in children's subjective well-being is explained by factors relating to school life.
There are indications that the importance of school-related wellbeing to overall subjective well-being is cultural and varies by national contexts.Huebner et al. (2014) highlight that much of the research in this area has been conducted in the US, where research tends to show that children's self-reported satisfaction with schooling makes a statistically significant but modest contribution to their overall life satisfaction, when compared with other factors such as satisfaction with friendships and family relationships.In the Korean context, research has shown that school-related satisfaction is far more important for the overall life satisfaction of Korean school students than for their counterparts in the US (Park & Huebner, 2005;Huebner et al., 2014).Bradshaw et al. (2011) considered the contribution of well-being at school to children's overall subjective well-being as part of a wider study that investigated the factors that affected children's subjective wellbeing at both a micro and macro level, to discern whether there were national differences in subjective well-being.Children's SWB was assessed on three components at the macro level: personal well-being, relational well-being, and well-being at school.At the micro level, these factors were considered alongside neighbourhood well-being.The study found that at the macro level there was no association between being well-being at school and subjective well-being.However, at the micro level (using data from England) there was an association between subjective well-being and well-being at school.
Using the Health Behaviours of School Aged Children survey 2009/ 10, Bradshaw et al. (2013) found that subjective well-being is associated with various objective domains of well-being (including, material, health, education, behaviour, housing and environment) but that subjective educational well-being is not associated with any objective measure (including education).However, this study used a composite measure of feeling pressured by schoolwork and liking school, and our research aims to investigate this further, using the more holistic measure of educational-well-being included in the Children's Worlds survey.
Research has also considered the impact of different aspects of school experiences on children's overall subjective well-being.In a literature review that focused on the academic correlates of children's and adolescents' life satisfaction, Suldo et al. (2006) identified three core aspects of schooling which were found to be related to life satisfaction in the literature: (1) perceiving themselves to do well academically (2) feeling that their teachers cared about them and were supportive (3) having generally positive experiences of school.

Well-being at school
Well-being at school has been given a significant amount of attention in the literature on domain-specific subjective well-being, because of the considerable amount of time children spend in school and the mandatory nature of schooling (DeSantis King et al., 2006;Long et al.,2012).This body of research has identified a number of important factors, including relationships with teachers and pupils (Casas et al., 2013;Liu et al., 2016;Kutsar & Kasearu, 2017), school belonging (Tian et al., 2015;2016) and academic attainment (Steinmayr et al., 2016;Huebner & Gilman, 2007).

Well-being at school and participation
Recent research has started to consider the link between subjective well-being and participation in particular domains, such as school, home and the community (Casas et al., 2013;González et al., 2015;Lloyd & Emerson, 2017;Kutsar et al., 2019;John-Akinola and Nic-Gabhainn, 2014).In a large-scale study of children in Spain, Casas et al. (2013) found that there was a relationship between children's reported wellbeing and their perception of the extent to which they were able to participate in decision-making at school.In an Irish study John-Akinola and Nic-Gabhainn ( 2014) identified an association between participation in school-decisions and rules, and children's health and well-being outcomes.Further, in their study of children's well-being across eight European countries, Kutsar et al. (2019) found that there was a positive association between eight-year-old children's subjective well-being and their perceptions that parents and teachers listen to them and take their views into account.In a Northern Irish context, Lloyd & Emerson (2017) identified an association between children's subjective well-being and participation rights in school and in the community.One of the key conclusions of this study was that the relationship between children's participation rights and the subjective well-being is facilitated by the 'social/relational aspects of both participation', (p.604) and wellbeing.This study aims to investigate this further, focussing on domainspecific well-being within the school context, using international data to discern whether the relationship between subjective well-being and participation rights is maintained across national contexts.

This paper
This paper builds on the analysis of previous authors (e.g.Kutsar and Kasearu, 2017), drawing on recent data to address the question of the relative importance of participation and rights on children's subjective well-being within the school environment.

Sample
This analysis uses data drawn from the third wave of the Children's Worlds studyan international survey of children's subjective wellbeing, funded by the Jacobs Foundation.The survey design is split into three age groups (8-, 10-and 12-year-olds) and collected responses from 128,184 children from 35 countries (Rees et al., 2020).This analysis focuses on responses collected from those in the oldest age group (mean age = 12.2 years) from 13 EU and former EU countries.Table 1 provides the number of responses collected from each of these countries, along with the overall and slight variation in mean age and gender composition. 1  As we are interested in the role of national contexts in subjective well-being, we have selected countries which are comparatively homogenous on a number of measurese.g. in terms of relative global affluence, and the age at which compulsory education begins and ends.In none of the countries sampled are children expected to leave school 1 Whilst the authors recognise that gender is more complex than the simple binary presented in the statistics here, the survey was designed such that respondents self-identified themselves as either girl or boy (hence the use of these terms in this paper).

R. Mari Barrance and J. May Hampton
and enter the workforce prior to age 16, and as a result, childhood is generally perceived as a protected period of the lifecycle without adult responsibilities.Previous research has indicated that there is a relationship between SWB and participation in decision-making at school in some European countries, including Northern Ireland (Lloyd & Emerson, 2017) and Spain (Casas et al., 2013).In this paper we investigate whether this relationship is consistent across the European countries in our sample.

Ethics
Whilst the Children's Worlds Survey has a centralised team, each country had its own research team responsible for data collection and obtaining associated ethical permissions.All countries included in the study had ethical approval granted by their relevant organisational body's ethical committee.

Dependent variable
As an indicator of children's well-being in their school life, a single item from the Brief Multidimensional Student Life Satisfaction Scale (Seligson et al., 2003) was used: Satisfaction with life as a student.
Children were asked to indicate their satisfaction with their life as a student along a Likert-type scale, ranging from 0: not at all satisfied to 10: totally satisfied.
Although presenting a skewed distribution, the dependent variable Satisfied life as a student was treated as linear, allowing multivariate linear regression to be performed.Treatment of happiness and life satisfaction variables as satisfying linear assumptions is an accepted and well-established method in subjective well-being research literature (Bradshaw & Rees, 2017), with Ferrer-i-Carbonell & Frijters ( 2004) concluding that little difference to is made to findings whether these measures are treated as ordinal or cardinal.2

Independent variables
A number of survey items measured aspects of the school environment (social rather than physical).Importantly for this analysis, these included an item measuring the extent to which children perceive that they can participate in decision making 'At school I have opportunities to make decisions about things that are important to me' and another item measuring the extent to which their opinions are seriously heard and considered 'My teachers listen to me and consider what I have to say'.Alongside these, were additional questions about their teachers ('My teachers care about me'; 'If I have a problem at school my teachers will help me') and perceived safety of the school ('I feel safe at school').Responses to all of these items were originally collected along a 5-point Likert-type scale: Do not agree, agree a little, agree somewhat, agree a lot, totally agree.Following Kutsar & Kasearu (2017), these were each transformed into binary indicator variables, indicating total agreement or otherwise.
Furthermore, detail was sought about children's relationships with fellow schoolmates, namely the extent to which they experienced bullying.Three items measured the frequency that children had experienced other children at school being unkind to them, excluded them, and hit them over the previous month, with four substantive response options provided along a scale: never; once; two or three times, four or more times.A three-level summary factor variable was constructed in an attempt to capture the frequency of bullying experienced.This was constructed by summing the responses to the three bullying items and then recoding to none (0), infrequent (1-3), and frequent (4-9).
The final independent variable included in the analysis measured children's satisfaction with the things they had learned.As with the dependent variable, this was originally collected along a Likert-type scale from 0: not at all satisfied to 10: totally satisfied.Unlike the dependent variable, this was transformed into a binary indicator variable, indicating total satisfaction or otherwise, to remain in keeping with the other independent variables.

Control variables
Items measuring demographic and socioeconomic characteristics of respondents included age, gender, and material deprivation.The latter allows assessment of the socioeconomic status of individuals without having to rely on measures of income (which may be unknown, as well as fluctuate).In the Children's Worlds study, material deprivation is measured using a specially developed child-centred series of indicators measuring whether children have access to good clothes, money for school trips, access to the internet, equipment for sports and hobbies, pocket money, two pairs of good shoes, a mobile phone, and the equipment necessary for school.The majority of children (70%) were missing none of the items, with 22% indicating that they were missing one.To capture the very most deprived, a binary indicator variable was constructed which indicated whether a child was missing two or more items (8.1% of the total sample).
Although the same questions were largely asked across the sample countries, some characteristic questions were omitted from individual countries analysis and so were not included in the current analysis.These include both ethnicity and disability variables.

Analytic design
The Children's Worlds surveys followed a complex sample design which was accounted for by recognising the clustering by school within each country and the use of weighting for single age group analysis. 3hese were both declared using the svy function in Stata.
Although more common to use an independent t-test when considering bivariate relationships between a continuous dependent variable and binary independent variable, in order to maintain the survey structure in the analysis and to allow for sub-population estimation, simple linear regression was conducted to determine the statistical significance of the main independent variables of interestschool decisions and teachers listento the variance of satisfaction with life as a student.These bivariate analyses were also run for other the independent variables.As well as calculating these relationships for the sample as a whole, sub-population analysis was conducted (through the svy subpop command) to calculate estimates for each country.
Following estimation of the bivariate relationships between the dependent and independent variables, multivariate linear regression was used to assess the relationships between independent variables of interest and the dependent variable, whilst controlling for the relative effect of one another, along with demographic and socioeconomic characteristics.Separate multivariate linear regression analyses were run for each country as a subpopulation of the EU sample by creating a series of dummy variables for each country.The use of subpopulation estimation enables the accurate use and application of the declared survey design.In this way, regression coefficients and associated significance values were produced for each of the variables for each of the countries, allowing comparisons of the relative importance of the independent variables of interest within and between each country.
All analyses were conducted in Stata 14.

Dependent variable: Satisfaction with life as a student
Mean values of satisfaction with life as a student ranged from 8.51 (Romania) to 7.14 (Hungary), demonstrating some variation between countries (as detailed in Table 2 and Fig. 1).Although differences between mean values do not seem overly large in absolute terms, they were found to be statistically significant, with five countries having mean satisfaction scores statistically significantly higher than the average (Romania, Norway, Belgium, Finland, Spain) and four countries having scores significantly lower than the average (Estonia, Poland, Wales, Hungary).The standard errors also point towards variation, of a greater or lesser extent, within countries subsamples.

Independent variables
Table 3 shows proportion of children who declared that they 'totally agreed' with statements about their school environment (including school culture, social relationships and learning activities) identified as potentially contributing to overall well-being of pupils.Considering the responses as a whole, there are relatively similar levels of total agreement with satisfaction with things learned (31%), teachers caring (32%), and teachers listening (34%).Slightly more pupils totally agreed that they could participate in decisions in school (38%).The only area in which over half the pupils agreed was the safety of the school, with 52% totally agreeing that they felt safe in school.
These overall values mask the great deal of variation between countries, with Malta consistently scoring above average across the measures, Italy consistently scoring below average on all measures, and with less systematic variation across the other countries.Variation was observed in descriptive analysis of the main independent variables of interest: opportunity to be involved in school decisions and teachers listening.Malta reported the highest levels of total agreement for both, with more than half of children indicating this (62.3% and 50.7%, respectively).Conversely only around a quarter of children in Italy totally agree to these statements (24.0% and 25.2%, respectively).
In terms of relationships with peers, 39.0% of pupils indicated that they had not experienced any of the three forms of bullying (being hit, called unkind names, nor excluded by others) over the past month, 39.9% had experienced 1-3 events, and 21.1% had experienced 4 or more.Variation between countries is visible in Fig. 2, showing the Finnish subsample to be country with least instances of this kind of behaviour at just less than half of pupils in this country reporting this, whilst Wales demonstrates the highest levels, both overall and in terms of the most intense category (4 or more events), with over 70% reporting at least one instance.

Bivariate analysis
To preliminarily test whether a formal relationship exists between the two independent variables of interest and the dependent variable, bivariate analyses were conducted.These indicated that a positive, statistically significant relationship does exist between the dependent variable and independent variables of interest, such that total agreement corresponds with a similar increase in values of satisfaction with life as a student for both.Total agreement with involvement in decisions explained a significant amount of the variance in satisfaction with life as a student, F(1, 309) = 603.42,p < 0.001, R 2 = 0.085, with the regression co-efficient indicating that total agreement corresponded to an average increase of odds by 1.27 on the satisfaction score, B = 1.27, 95% CI [1.17,1.38].Similarly, total agreement with feeling teachers listen to them explained around 8% of the variance in satisfaction in school, F(1, 309) = 671.36,p < 0.001, R 2 = 0.084, with the regression co-efficient indicating that total agreement corresponded to an average increase of odds by 1.30 on the satisfaction score, B = 1.30, 95% CI [1.20,1.40].These relationships held true across both for the sample as a whole and for each country when treated as subpopulations, despite variation in strength.Multivariate analysis.
Multivariate linear regression was conducted to examine the relative influence of the independent variables on pupils' satisfaction with their lives as students.When considered as a whole, the multivariate analysis established that the independent variables were all statistically significant to levels of satisfaction with their lives as students (see bottom row Table 4).Together, these variables explained 28% of the variance in the dependent variable, F(10, 300) = 144.97,p < 0.001, R 2 = 0.28.
As with the descriptive statistics, some variation is evidenced between countries.The relative importance of school decisions, teachers listening, teachers care, low levels of bullying, safety of the school, gender, age, and experiencing material deprivation all vary between countries.In terms of these latter three variables, for those few countries where these were found to be important, increasing age and material deprivation were found to be associated with lower levels of well-being.Gender appears to have a more varied effect, with girls reporting lower well-being than boys in Malta but higher in Croatia and Norway.Conversely, satisfaction with things learned and high levels of bullying remain statistically significant across all countries, with the former showing by far the highest coefficients (i.e. the strongest positive effect on school well-being) and the latter showing the lowest coefficients (i.e. the strongest negative effect on school well-being).Perceived safety of the school was also shown to be reasonably consistently important, being statistically non-significant for Italy only.
In terms of the independent variables of particular interest to this study, total agreement with school decisions was associated with statistically significant increases in school well-being for all but Malta, whilst total agreement with teachers listening was only statistically significant for Italy and Hungary.Teachers caring was important in more countries, showing statistical significance for Belgium, Estonia, Hungary, Malta, Norway, and Poland.

Discussion
Previous research has shown that autonomy in time use is important, and may explain why children in relatively affluent countries, such as South Korea, report comparatively low levels of SWB (Lee and Yoo, 2017).As Lee and Yoo suggest, this is likely due to a demanding academic culture which leaves little free time for students outside of private tutoring and schooling.In this study, we examine the question of autonomy from a different perspectivewe focus on the importance of opportunities to participate in decision-making within school.The research brings international evidence to support the previous findings of national studies conducted in Spain (Casas et al., 2013) and Northern Ireland (Lloyd & Emerson, 2017), which indicated that there was a relationship between participation rights and subjective well-being at school.The analysis presented in this paper has shown that there is an association between subjective well-being and participation in decisionmaking at school, across all countries included in this study, except Malta.This suggests that participation has an important role to play in SWB within the European context.
Our study found that teachers caring and listening were important for subjective well-being, supporting the findings of previous studies (Suldo et al. 2006).However, involvement in school decisions makes the bigger contribution to subjective well-being than either of these, suggesting that participation is important in and of itself.The importance of school decision-making, teachers caring and teachers listening to children's SWB lends weight to previous research (Fattore et al., 2009;Lloyd and Emerson, 2017) that has suggested that the relationship between SWB and the fulfilment of participation rights is facilitated by the 'social and relational aspects of both' (Lloyd and Emerson, 2017: p.604).As Lloyd and Emerson suggest, the supportive relationships between teachers and children may play a key role in enabling children to claim their participation rights.This is in line with scholarship on the right to participation (Lundy, 2007), which has highlighted the interdependence  R. Mari Barrance and J. May Hampton between the right to participation and other rights in the Convention, and emphasised the key roles of the rights to information (article 13) and guidance from adults (article 5) in the fulfilment of children's participation rights.
The results suggest more broadly that relationships with others make a key contribution towards SWB, not only between teachers and children, but also amongst pupils.The research found that bullying was negatively associated with satisfaction with life as a student across all countries, and the greater the number of bullying incidences, the stronger this relationship was.This echoes the findings of previous research on earlier Children's Worlds datasets (Savahl et al., 2019), which found that bullying made a significant contribution towards SWB, which was strengthened when physical and psychological bullying were combined.Linked to this, 'feeling safe at school' made an important contribution to children's SWB.This supports previous findings (Bradshaw et al., 2017;Klocke et al., 2014) which show the detrimental impact of bullying on global life satisfaction, and the importance of relationships with peers.
As highlighted above, our sample consists of European countries which are relatively homogenous in terms of affluence (when compared globally), and in terms of educational provisioneducation is compulsory in all selected countries until at least 16.The research found a number of commonalities across countries sampled, particularly in terms of the importance of bullying and participation to SWB.Where differences are identified, these are likely due to differences in educational policies or social and political cultures.For example, being a girl is associated with a statistically significant increase in satisfaction in life as a student in Norway, which is widely recognised as having a strong culture of gender equality, and regularly ranks highly in worldwide indexes of gender equality and development (World Economic Forum, 2022;OECD, 2022).Conversely, in Malta there is a negative relationship between being a girl and SWB at school, and international comparisons place Malta as one of the lowest ranked European countries in terms of the Gender Gap Index 2022 (World Economic Forum, 2022, p. 24).However, in other cases, the relationship between socio-economic factors and SWB is not as clear.For example, it is difficult to determine why material deprivation is such a strong predictor of SWB at school in Wales and Malta.This may be partly explained by high levels of inequality in these countrieswhile there is no individual score for Wales, the UK has one of the highest Gini coefficients in Europe, and is highest amongst sampled countries (World Bank, 2022).This would appear to support the findings of previous research, which suggest that it is relative, rather than absolute deprivation that makes the most important contribution to SWB (Bradshaw et al., 2011;White, 2008).However, it is clear that there is a more complex interplay of factors at work here, since while Malta has a comparatively high score of 31, it is lower than though other countries in the sample, including Spain, Romania and Italy (where material deprivation did not make a statistically significant contribution).Moreover, material deprivation is also a predictor of SWB at school in Finland, which has one of the lowest Gini coefficients in the sample.It is clear that further investigation is needed to understand the role of social and political cultures in relation to SWB at school in the European context.

Limitations
Although this paper provides useful information about domain specific well-being, and the importance of the opportunity for participation in decision making in the school environment, it is not without limitations.We know that children's well-being is multi-faceted (Ben-Arieh et al., 2014) experienced across, as well as within, various aspects of their lives.There is little evidence on the interplay between well-being in various domains, although research has indicated that social actors from various domains play a key role in levels of well-being but that this effect is mediated by experiences within school (Oriol et al., 2017), highlighting the complex nature of the phenomenon.Although an aspect of social relationships within school was included in the analysisnamely, negative social relationships in the form of bullyingthe current study overlooks the potential importance of wider social relationships (e.g.other friends and family) and the role that they may play, directly or indirectly, on children's well-being within school.
We are also aware that while the Children's Worlds survey asks children questions about their participation in decision-making, that there are limitations to this data, given that it is difficult to determine from survey data about the kinds of decision's children are allowed to participate in, and the extent to which their views are acted upon as appropriate.It must also be acknowledged that, despite our focus on the fulfilment of children's participation rights, the research itself is not participatory.Whilst the questionnaire was designed and piloted using focus groups and interviews with children, children were involved in the development of the research instruments or analysis of the data.Currently, very little quantitative research in the area of child wellbeing and child indicators is participatory (Lundy, 2014).There is a need for future research which responds to this call using children's rights based research methods (Lundy and McEvoy, 2012), which are empowering to children and young people, and enable them to claim their rights.

Conclusion
A key finding of this research was that low proportions of children feel listened to in school, feel that their teachers care about them, and feel that they are able to participate in decision-making at school.This raises questions and concerns about the nature of schooling in the countries included in the survey, and indicates that schools tend not to be empowering and rights-based institutions for all children.Such findings suggest that this is an essential area for future research.This research should consider whether particular groups of children are more likely to have positive experiences of school-decision making and report better relationships with teachers, as this is a key equalities issue and has important implications for the fulfilment of the right to nondiscrimination (article 2 of the UNCRC).Future research could build upon the findings of our study by considering whether the aspects of school decision-making which children are able to participate in makes a difference to their SWB.We know from previous studies that children are rarely involved in decisions about curriculum and assessment at school (Barrance & Elwood, 2018;Elwood, 2012;Elwood and Lundy, 2010) and that much of the literature around student voice indicates children's general dissatisfaction with the opportunities available to them (Alderson, 2000;Wyse, 2001;Osler, 2008).Asking about different types of school participation will also help us develop a better understanding of what children mean when they report that they are involved in decision-making about things that are important to them at school and so that we can identify which aspects are particularly valuable to them.A children's rights approach to such research (Lundy and McEvoy, 2012) would involve developing the options in collaboration with children, so that they are based on children's knowledge and experience.As well as empowering children and enabling them to claim their rights, this information would provide key information to policy-makers and schools which they could use to fulfil their duties to children.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 1 .
Fig. 1.Satisfaction with life as a student, mean scores by country.

Table 1
Number of responses per country.

Table 2
Satisfaction with life as a student, mean scores by country.

Table 3
Total agreement with school environment questions, by country.

Table 4
Multivariate regression coefficients predicting satisfaction with life as a student.Reference categories: those who did not totally agree to school decisions, teachers listen, teachers care, school safe, things learned; none for bullying; boys; missing < 2 items.Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001 R. MariBarrance and J. May Hampton