Subjective Well-being of Brazilian Children Over Time: Comparing Children's Worlds 1st and 3rd Wave of 10 and 12-year-olds Samples.

This study aims to verify differences in the subjective well-being of 10 and 12 years old Brazilian boys and girls over time using data from the First (2012) and the Third (2019) Wave of Brazilian data collection on Children's Worlds research. Participants of the first wave were 2,338 children (M = 11.08 years old, 55.3% girls) and of the third wave, they were 1,787 children (M = 11.33 years old, 55.3% girls). We performed a Multivariate Analysis of Variance (MANOVA) with subjective well-being scales (OLS, PWI-SC and BMSLSS) used as dependent variables, and age, gender and times of measurement (wave of data collection) were used as independent variables. Main results present that there are significant reductions of all subjective well-being means over time (from 2012 to 2019) regardless gender and age. These results denounce the profound changes in contexts associated with diminished investment in education, access to health, social care and children's rights' policies overall in Brazil and reinforce the perspective that children's participation in society is essential for the development of public policies that represent their current needs.

Introduction

The interest in the study of children’s well-being increased in the last couple of decades (Ben-Arieh, 2008; Dell’Aglio et al., 2011; Sarriera & Bedin, 2017). To evaluate children’s well-being from a positive perspective of childhood, it is necessary to investigate health, education, demography, social services, but also to go beyond those aspects and aim at children's satisfaction with services and aspects of their lives, interests and activities (Casas, 2010). It is essential to continuously access children’s intakes on their lives in order to have indicators not only about children's development, but also about the impact of public policies in their lives over time (Bradshaw, 2019; Casas, 2010).

The United Nations have chosen the promotion of well-being of all ages as one of the goals for the 2030 agenda for sustainable development (UN, 2015). Children’s Worlds Project, from the International Society for Child Indicators, focuses on the study of children’s well-being, considering children’s own takes on their lives. Subjective well-being (SWB) is a key concept when it comes to studying children’s lives, as it is a predictor of mental health (Bradshaw, 2019 It consists of an overall evaluation, with a cognitive and an affective component (Diener, 1984). The first refers to an overall cognitive appreciation on one’s life, considering specific domains, and the second one considers the inner experience of life satisfaction both on positive and negative affects (Casas, 2010; Diener, 2000; Moreira et al., 2021). This study takes interest on the cognitive aspect of subjective well-being.

The measurement of life satisfaction for children is still a challenge, considering the abstract character of the concept itself and the need to evaluate social indicators that are equally relevant (Bradshaw et al., 2011; Bradshaw, 2019). Lee (2014) states that child well-being indicators are the representation of well-being levels, within domains, through statistical measures. From this perspective, studies focusing on indicators of child well-being should consider children as their main informants and measures that respect their ability to evaluate their lives, considering language, form of measurement and their main interaction contexts. Children’s participation is key not only to promote their rights, but as a path to ensure reliable intervention programs and policies (Casas & Bello, 2012).

The ecological-contextual framework takes interest in the influence of context to one’s life. It proposes that social structures and social processes are interdependent. Social structures, such as policies, procedures and events attribute meaning to the elements of the system. On the other hand, social processes are related to the interpersonal exchanges within and between structures. A social structure is composed of personal resource potential, resources, settings, and boundaries, while social processes are the actions on these structures. Processes illustrate how a system responds to entropy, through the concepts of adaptation, reciprocity, networking and boundary spanning (Kelly et al., 2000).

Children’s lives and relationships take place in their systems, such as within their families, school and friends. Children are assigned roles and perform activities, which are their social structures. The relationships in between these contexts, the social processes, mutually affect each other and children’s perceptions and evaluations (Casas, 2010; Sarriera, 2010). Therefore, in this perspective, we propose an approach that recognizes the interaction among social processes and structures to study the contribution of multiple social levels on the SWB of children. Previous studies on the influence of psychosocial variables in children’s relationships and well-being have established the influence of gender, type of school and context of housing (Santos et al., 2018; Vieira et al., 2015).

Currently, studies on children’s well-being point to certain tendencies, and some of them concern age and gender. Younger children and boys tend to have higher rates of well-being (Bedin et al., 2020; Kim et al., 2019; Rees et al., 2020; Sarriera et al., 2014). This phenomenon may be related to psychosocial influences on subjective well-being, or even expected developmental changes. Lawler et al. (2016) have also found family and peer relationships, school, and neighborhood quality to be strong predictors of SWB. Parental stress and psychopathology, as well as lower social capital also contributed to poorer outcomes in children’s mental health (Caetano et al., 2021). And, children from countries with more consistent family policies presented higher indexes of SWB, supporting the hypothesis of state level decisions and their impact on children’s lives (Mínguez, 2017).

Historically, Brazil faces strong challenges when it comes to guarantying children’s rights, that goes from child mortality, poverty, violence as disciplinary actions (Caetano et al, 2021), to economic crisis and unequal distribution of resources. As children from privileged backgrounds have access to means that are unavailable for vulnerable children, these differences are also relevant when it comes to child well-being. A cross-cultural study with eight countries including Brazil, found a strong relationship between children’s perception of access to material resources and subjective well-being, especially in contexts of extreme deprivation (Sarriera et al., 2015).

Major challenges are reported both on school and family contexts. In 2019 the National Scholar Health Survey (PeNSE – IBGE, 2021) revealed that close to 12% of students between 13 and 17 years old have not attended school for not feeling safe on their commute to school. The survey also illustrates some of the social inequalities faced by children and adolescents from public and private schools. Around 50% of students from public schools had access to computer, against around 90% of students from private schools, and 82% of public school children had access to cellphones, against 95% of private schools. This difference is also apparent when it comes to balance school and work, while 59% of public-school children affirm to study and work simultaneously, whilst 49% of private schoolers balance work and study. This difference is also racial, with higher incidence of black students that study and work simultaneously (55,2%) (IBGE, 2021). These results indicate not only a lack of state policies that effectively address these situations, but a fail to attain the current legislation regarding protecting children and adolescents, such as the Statute of Child and Adolescents (Law 8060, 1990) and the antibullying Law No. 13, 185 (2015).

The Statute of Child and Adolescents (ECA) (Law nº 8060, 1990) states that “it is the duty of the family, the community, society in general and of the public authorities to ensure, with absolute priority, the realization of rights referring to life, health, food, education, sport, leisure, professionalization, culture, dignity, respect, freedom and family and community life”. Although this is provided for in the law, the application that is observed in the precariousness of the education system, and even in the basic assistance policy for the population, through income redistribution programs, goes in another direction. Although in the past 30 years Brazil has been on a considerable reduction of child mortality rates in every region of the country, since 2015 this reduction has slowed down on face of financial and political crisis. This situation culminates in cuts on income distribution programs, such as Bolsa Família, that targets families in extreme poverty, with children until 17 years old, pregnant or lactating women (Leal et al., 2018; Marinho et al., 2020).

The last decade in Brazil has also been of diminished investments in education (Arelaro, 2017) and the failure to provide rights has been evident. There were expressive changes in Brazilian children’s contexts, such as alterations in the allocation of resources, changes on mortality rates and political policies (Caetano et al., 2021; Marinho et al., 2020), which could implicate in their relationship with families, friends and school, and therefore in children’s well-being. Thus, we wish to broaden this perspective and evaluate the difference in children's well-being in a 7-year time range.

Background

This study is part of the International Society for Child Indicators’ research “Children’s Worlds”. The research focuses on the gathering of data on children’s well-being. First-wave of research was collected between 2011–2012 and consists of data from 13 countries and 34,000 children, aged between 8 and 12 years old, with 3,328 children from Brazil. After this stage, researchers compared data and adapted instruments as well as added questions regarding diverse life situations, which led to the second wave, from which 61,000 children from 18 countries participated, and Brazil did not take part at this stage of research. In 2019, a new-wave surveyed 128,000 children from 35 countries, of which 2,676 from Brazil (Bruck & Ben-Arieh, 2020). In Brazil, research was conducted by the Research Group of Community Psychology from the Universidade Federal do Rio Grande do Sul (GPPC-UFRGS).

This is a sequential design study, as proposed by Schaie (1965) for studying pattern changes in two groups, considering two times of measurement. Considering these implications, this study aims to verify differences in the subjective well-being of 10 and 12 years old Brazilian boys and girls from the First (2012) and Third (2019) Wave of Brazilian data collection on Children's Worlds research.

Method

Participants

In the first wave, we examined a sample of 2,338 children (M = 11.08 years old, 55.3% girls) from public (61%) and private (39%) schools of Rio Grande do Sul state, Brazil. In the third wave, we examined a sample of 1,787 adolescents (M = 11.33 years old, 55.3% girls) from public (72%) and private (28%) schools in five Brazilian states: Rio Grande do Sul, Santa Catarina, Paraná, Rio de Janeiro and São Paulo. The schools were randomly selected from a list made available by the Department of Education of each state and children from 5th grade (mean age of 10 years old) and 7th grade (mean age of 12 years old) were invited to participate.

Instruments

For the present study, we used one single-item scale (overall life satisfaction) and two SWB scales with multiple items (PWI and BMSLSS) described below. The measures were adjusted for a scale ranging from 0 to 10 (0 = totally dissatisfied and 10 = totally satisfied) in order to facilitate the understanding of children (Casas et al., 2012). The questionnaire also contained sociodemographic variables such as age and gender.

Overall Life Satisfaction (OLS)

Consists of a single item of overall life satisfaction (OLS), in which participants respond to the question: At the present, to what extent are you satisfied with your life, considered as a whole? The importance of using a single item in the assessment of SWB was emphasized by Campbell et al. (1976).

Personal Wellbeing Index- School Children (PWI-SC)

The PWI-SC was originally developed for adults by Cummins et al. (2003) in order to assess the well-being of different groups of the population. The original instrument consists of seven items of satisfaction, and each corresponds to a domain (health, quality of life, accomplishments, feeling safe, community, security for the future, and interpersonal relationships). The items are answered from a question of satisfaction with life in general. This study uses the version proposed by Cummins and Lau (2005) for school-age children, with adapted and simplified language. Scale items are: “How satisfied are you with all the things you have?”, “How satisfied are you with your health?”, “How satisfied are you with the things that you want to be good at?”, “How satisfied are you with your relationship with people in general?”, “How satisfied are you about how safe you feel?”, “How satisfied are you about doing things away from your home?” and “How satisfied are you about what may happen to you later in your life?”. For the version adapted by Schütz et al. (2018) it was found a Cronbach's alpha of 0.70. For this sample, we found Cronbach's alpha of 0.77.

Brief Multidimensional Students’ Life Satisfaction Scale (BMSLSS)

The short version of the Multidimensional Students’ Life Satisfaction Scale was developed by Seligson et al. (2003). In its original form, the BMSLSS consists of five items, and a general item on satisfaction with life. For the present study only the five domain items are used. The reliability and validity of the scale were evaluated in the study of Benjamin, Funk, Huebner and Valois (2006), by which it is recommended the one-factor solution, and the obtained internal consistency was 0.75. For the version adapted for Brazilian children by Schütz et al. (2018), authors found a Cronbach's alpha of 0.69. The internal consistency for this sample was 0.74. The items are: “How satisfied are you with your family life”, “your friends”, “your experience in school”, “yourself”, and “the area where you live, in general”.

Procedures

The selected schools in Brazil were contacted and, once authorization was obtained through the Institutional Agreement Term, children were invited to participate in the research, after clarifying the objectives and procedures involved. The Informed Consent Forms were delivered to take home, and those responsible, as well as the Assent Forms to the students who agreed to participate signed the term. For the application of the questionnaire, a day was defined with the schools’ coordinators, upon return of the terms of consent and assent, being applied collectively, in a room provided by the schools, with an approximate duration of 45 min. At least two trained researchers from the team were present in the application, to clarify doubts and assist in filling in the data. All ethical procedures foreseen for research with human beings were ensured and the research ethics committee approved the project.

Data Analysis

We performed a Multivariate Analysis of Variance (MANOVA) to verify differences in the SWB of 10 and 12 years old Brazilian boys and girls in 2012 and 2019. Subjective well-being scales (OLS, PWI-SC and BMSLSS) were used as dependent variables, and age, gender, and times of measurement (wave of data collection) were used as independent variables.

Results

Descriptive Analysis

Table 1 presents the means of the children’s subjective well-being, considering their age, gender, and different times of measurement for the instruments (OLS, PWI-SC and BMSLSS). It is possible to verify that means are higher for all SWB means in 2012 than for 2019, indicating a decrease in SWB indexes over time. The means of 10-year-old children are higher than those of 12-year-olds, as well as the means of boys are higher than girls for the three well-being scales.

Table 1

Means and standard deviations of the OLS, PWI-SC and BMSLSS by time of measurement, age and gender

Wave		Boys – M (SD)			Girls – M (SD)			Total – M (SD)
	10 years	12 years	Total	10 years	12 years	Total	10 years	12 years	Total
OLS	First wave (2012)	9.46 (1.22)	9.02 (1.66)	9.26 (1.46)	9.43 (1.38)	8.81 (1.89)	9.17 (1.64)	9.44 (1.31)	8.91 (1.79)	9.21 (1.56)
	Third wave (2019)	9.01 (2.13)	8.58 (2.31)	8.79 (2.24)	8.87 (2.32)	7.82 (2.76)	8.35 (2.60)	8.93 (2.24)	8.17 (2.59)	8.54 (2.45)
PWI-SC	First wave (2012)	8.93 (1.08)	8.71 (1.15)	8.83 (1.12)	9.02 (0.98)	8.59 (1.29)	8.83 (1.15)	8.98 (1.03)	8.65 (1.23)	8.83 (1.13)
	Third wave (2019)	8.63 (1.42)	8.26 (1.56)	8.44 (1.51)	8.49 (1.61)	7.84 (1.76)	8.17 (1.71)	8.55 (1.53)	8.03 (1.68)	8.29 (1.63)
BMSLSS	First wave (2012)	8.93 (1.20)	8.63 (1.22)	8.79 (1.22)	9.05 (1.08)	8.50 (1.31)	8.81 (1.21)	8.99 (1.13)	8.56 (1.27)	8.80 (1.22)
	Third wave (2019)	8.74 (1.59)	8.07 (1.81)	8.39 (1.74)	8.60 (1.67)	7.58 (1.95)	8.09 (1.88)	8.66 (1.64)	7.80 (1.91)	8.23 (1.83)

Differences Between Children’s Well-Being by Age, Gender and Time of Measurement

We performed multivariate analysis of variance (MANOVA) and analysis of variance (ANOVA) to verify differences between the means of the children's well-being (PWI-SC, BMSLSS and OLS) considering age, gender and time of measurement. The dependent variables were the means of the PWI-SC, BMSLSS and the single-item measure OLS, considering that these instruments are representative of SWB and present correlations of 0.62 (between BMSLSS and OLS), 0.76 (between BMSLSS and PWI-SC) and 0.61 (between PWI-SC and OLS).

Table 2 shows the MANOVA results’, considering age, gender, and time of measurement as independent variables. Considering instruments as a unit, we observe age, gender and time of measurement had a significant effect. The interactions between age and time, age and gender, time and gender are significant, but the interaction of the three variables altogether is not.

Table 2

MANOVA by age, gender and time of measurement

	Wilks’ Lambdaa	F	df	Error df	Sig
Age	0.956	57.125	3.00	3691.00	0.000
Gender	0.995	6.626	3.00	3691.00	0.000
Time	0.962	48.084	3.00	3691.00	0.000
Age * Gender	0.997	3.848	3.00	3691.00	0.009
Age * Time	0.994	7.311	3.00	3691.00	0.000
Gender * Time	0.997	3.633	3.00	3691.00	0.012
Age * Gender * Time	0.999	1.130	3.00	3691.00	0.335

aDependent variables (DVs): PWI-SC, BMSLSS and OLS.

Through ANOVAs we find significant differences between gender, age and time of measurement for the three dependent variables (PWI-SC, BMSLSS and OLS). Furthermore, there are significant differences for the interaction between gender and age, gender and time of measurement for the three DVs. For the interaction between age and time of measurement, it was significant for BMSLSS and PWI-SC, but not for OLS, and for the interaction of the three IVs relations are not statistically significant as seen in Table 3. Main results present that there are significant reductions of all SWB means over time (from 2012 to 2019) regardless gender and age, as seen in Table 1.

Table 3

ANOVA by age, gender and time of measurement

	Dependent variable	Sum of squares	df	Mean Square	F	Sig
Age	BMSLSS	350.722	1	350.722	164.404	0.000
	PWI-SC	135.735	1	135.735	85.339	0.000
	OLS	349.838	1	349.838	92.503	0.000
Gender	BMSLSS	21.639	1	21.639	10.144	0.001
	PWI-SC	19.529	1	19.529	10.840	0.001
	OLS	71.651	1	71.651	18.946	0.000
Time	BMSLSS	244.656	1	244.656	114.684	0.000
	PWI-SC	224.092	1	224.092	124.394	0.000
	OLS	328.725	1	328.725	86.920	0.000
Age * gender	BMSLSS	18.911	1	18.911	8.865	0.000
	PWI-SC	12.564	1	12.564	6.947	0.008
	OLS	35.705	1	35.705	9.441	0.002
Gender * time	BMSLSS	20.858	1	20.858	9.778	0.002
	PWI-SC	14.853	1	14.853	8.245	0.004
	OLS	24.327	1	24.327	6.433	0.011
Age* time	BMSLSS	39.959	1	39.959	18.529	0.000
	PWI-SC	7.529	1	7.529	4.179	0.041
	OLS	9.371	1	9.371	2.478	0.116
Age * gender * time	BMSLSS	0.524	1	0.524	0.246	0.620
	PWI-SC	0.289	1	0.289	0.160	0.689
	OLS	10.699	1	10.699	2.829	0.093

Discussion

This study had the objective of verifying differences in the SWB of 10 and 12 years old Brazilian boys and girls from the First (2012) and Third (2019) Wave of Brazilian data collection on Children's Worlds research. Our results point to significant reductions of all SWB means over time (from 2012 to 2019) regardless gender and age, and dissent from the UN's goal for the promotion of well-being for all ages. Considering the importance of the subject to indicate children’s quality of life currently (Bradshaw, 2019) and the global priority, these results are aligned with the ecological, relationship-based model of children’s subjective well-being, that states the importance of relational, school and neighborhood satisfaction (Lawler et al., 2016). These results also contribute to the number of studies considering children as main informants, especially in Latin American context.

Differences in children’s SWB over time may indicate profound changes in contexts associated with diminished investment in education, access to health, social care, and children’s rights’ policies overall. Our results raise questions on the sorts of adaptation children from 2019 had to take on social systems, when compared to the 2012 sample. For Kelly et al. (2000), adaptation is an active attempt to change processes of a system in response to external environmental demands.

Since the Constitutional Amendment 95 (EC95) from 2016, the investments in social areas, such as health and education, have been frozen for the next 20 years, which indicates a progressive decline in the formation of teachers and education professionals and on the overall quality of education (Arelaro, 2017). Schools represent an important area in children’s lives, a space where they can thrive, learn new skills, and develop emotionally and socially. It can also be a source of protection and a place to attain basic needs, such as being fed. It is expected that lack of investments in education and on training of teachers would make a difference on children’s resources and on their SWB. Satisfaction with school is a predictor of SWB, especially when considered along with children’s relationships (Lawler et al, 2016; Schütz et al., 2016). On the other hand, interventions on children’s SWB occurring in school’s grounds have previously been effective, demonstrating increases on children’s SWB (Sarriera & Bedin, 2017; Schutz, & Stum, 2017).

Marinho et al. (2020) argued that a slowdown in the annual reduction of children mortality was already seen prior to the budget restrictions, in 2015, and the expectation for the years following EC95 was tangible negative effects on children’s health. With the limitations on investments to conditional cash transfer programs, such as Bolsa Família, poverty levels have also been increasing (Leal et al., 2018). Changes in the investment in mental health should also be considered, once it is known that not only children with higher parental stress, parental psychopathology, and lower social capital tend to present more behavioral, developmental or mental health disorders (Caetano et al., 2021), but also that parents’ SWB might have an impact on their children’s (Bedin & Sarriera, 2014). The perils of these reductions are also the effects on community’s assistance programs that have been showing good results between 1990–2015, with major implications on public policies operationalized through Brazilian’s both Unified Health System (SUS) and Social Assistance Unified System (SUAS), with direct repercussions on family’s lives.

These lack of investments and proper management from Brazilian government are not restricted to health and assistance. Ministry of Education has also suffered from the non-continuance of programs and superficial initiatives (Arelaro, 2017). These measures might expose a larger number of children to more violence and victimization, which are risk factors for their subjective well-being (Fernandes et al., 2020).

Changes in family policies might contribute to children’s life satisfaction and specifically to children’s satisfaction with family life. Previous studies have argued that investment on early childhood care and education services, as well as assistance to work-family balance (through maternity leave assistance and the promotion of quality time spent with family) and efficient measures to verify these investments are key to promote child SWB (Mínguez, 2017). These are opposites to Brazilian governments’ actions in the last years and could reflect on children’s current life satisfaction, but also on their overall feeling of belonging to the country and to their ability to promote social changes.

In our study, children differed significantly from each other in all well-being instruments, considering their age, gender and time of measurement. Younger children, boys and children from 2012 showed higher well-being measures. The results considering age and gender are expected and confirm a tendency that has been previously reported in literature (Bedin et al., 2020; Kim, et al., 2019; Rees et al., 2020; Sarriera et al., 2014). Since these differences in well-being measures are consistent in the first wave and in the third wave, we believe the interaction between age and time of measurement, as well as gender and time of measurement reinforces the downfall on children’s well-being as a consequence to external, larger social factors.

Final Consideration

Reductions in children’s SWB over this period of time suggest the effect of public investment in childhood policies on children's lives and support the hypothesis of the importance of social processes and structures on the evaluation of subjective well-being. Kelly et al. (2000) states the need for the creation of social structures that encourage people’s participation in processes and allow them to have a meaningful influence on the system promoting emancipation. These results reinforce the perspective that children’s participation in society is essential for the development of public policies that represent their current needs. Therefore, it is essential to continuously access their intakes on their lives including on surveys and on the knowledge production, considering that children are citizens now, not only in the future.

This study is not without limitations. Despite using two times of measurement, this is a transversal study and children from the first wave could also present higher well-being measures on the present. Even though this effect is not likely due to a developmental effect or a change in generation, this should be considered. As Brazilian children did not participate in the second wave of research, we don’t have access to a third time of measurement that could help explain this downfall on SWB. Our method, even though carefully performed, does not allow us to make causal inferences.

Future studies should focus on qualitative perspectives, and longitudinal data collection to address some of these methodological challenges. Interviews on children’s perception of larger government initiatives in their lives, as well as on their knowledge on their basic human rights could point to other important directions. Since the last data collection, COVID-19 has affected the world and has ongoing implications, though it is possible to assume an even sharper drop on Brazilian children’s SWB, considering not only the large mortality rates in Brazil, but also the effects of the necessary social actions regarding the pandemic, as socially distancing. These situations could amplify vulnerabilities with children’s lack of access to schools and their possibilities of social support from this context (Christoffel et al., 2021). Social isolation poses as a challenge as it exposes vulnerable children to more violent situations (Loades et al., 2020).

With over 600,000 lives lost, setbacks have been immense in times of pandemic in Brazil, where the posture of the federal government was to deny scientific knowledge (Anjos & Pereira, 2021). The collective grieving will also pose as a challenge for children that lost significant family members and studies have shown that sudden losses of family member are particularly impacting for children (Albuquerque & Santos, 2021) and will probably pose as risks for their subjective well-being.

During the pandemic, other actions were taken by Brazilian government administration that increased vulnerabilities and violence, especially against indigenous people, homeless people, and overall human rights, increasing social inequality. Future studies should also focus on comprehend the implications of these state negligence and what has been called a necro policy from Jair Bolsonaro’s administration (Rapozo, 2021).

Hence, it is necessary to build, expand and strengthen forms of resistance to mitigate the injustices aggravated by the measures taken from current and previous federal administrations. The role of social movements as ways of provoking discussions, problematizing public policies and pedagogical actions developed, as well as putting pressure on governments in defense of children and Early Childhood Education, has a strategic importance in fighting the pandemic, but also in combating the deepening of inequalities (Anjos & Pereira, 2021). As UN’s goals for sustainable development aim, in “a world with equitable and universal access to quality education at all levels, to health care and social protection, where physical, mental and social well-being are assured” (UN, 2015) children would necessarily be heard.