1 Introduction

Life satisfaction is a subjective evaluation of overall quality of life and is considered to be the key indicator of subjective well-being (Proctor et al. 2009; Gilman and Huebner 2003). Subjective well-being is synonymous with happiness and refers to how people feel and think about their lives (Diener 1984; Diener et al. 2002). According to the tripartite model (Diener 1984), subjective well-being comprises both emotional and cognitive components. The two emotional components of subjective well-being include the experiences of positive and negative affects in one’s life, the cognitive component relates to evaluative judgments of life quality, that is, a global life satisfaction or satisfaction within a specific domain. Although the structure of subjective well-being still remains unclear (Busseri and Sadava 2011), substantial evidence indicates that the emotional components are primarily determined by reactive domains of personality and that subjective well-being has a strong cognitive component (Davern et al. 2007; Schimmack et al. 2004). Because the nature of emotional components of subjective well-being is relatively well understood, the focus of the present study is on the cognitive component, or life satisfaction, which is often used interchangeably with subjective well-being and happiness (Busseri and Sadava 2011; López-Pérez et al. 2016; Proctor et al. 2009).

A vast majority of studies on subjective well-being and/or life satisfaction has been conducted with adults. At first, research has been focused on socio-demographic factors, such as gender, age, marital status, education, occupational status, and income (Diener et al. 1999). However, the findings showed that these variables accounted for only a small amount (less than 10%) of overall variation in adult subjective well-being (Diener 1984; Diener et al. 2002). It has also been found that socio-demographic variables may have different impacts in different countries (Diener et al. 2002). Because life satisfaction, like personality traits, shows substantial continuity over time (Lucas and Donnellan 2007), the focus of research has shifted to personality traits that determine how a person perceives events and circumstances, to explain subjective well-being (Schimmack et al. 2002). The meta-analysis by DeNeve and Cooper (1998) showed that personality was equally predictive of life satisfaction, happiness, and positive affect, but significantly less predictive of negative affect. The most studied personality domains are extraversion and neuroticism (Diener et al. 2003; Gomez et al. 2012; Steel et al. 2008); research with the five factor model has shown that neuroticism and conscientiousness are the strongest predictors of life satisfaction; extraversion and agreeableness may also make a contribution, the results for openness are inconclusive (DeNeve and Cooper 1998; Haslam et al. 2009; Hayes and Joseph 2003; McCrae and Costa 1991; Steel et al. 2008). Russian studies have yielded similar results for neuroticism and extraversion and have highlighted the role of temperamental characteristics such as social tempo and ergonicity (Shamionov and Grigor’eva 2017), for which there may be no directly comparable personality counterparts.

Whereas adult subjective well-being is an extensively researched phenomenon, child subjective well-being and/or life satisfaction has received considerable attention more recently. Although early research on child well-being was largely focused on the child quality of life from adult perspective, expert or parental, contemporary studies emphasize the importance of measuring child’s thoughts and feelings about their life (Ben-Arieh 2010; Bradshaw et al. 2011; Rees et al. 2012). Most findings on child and adolescent life satisfaction parallel those for adults. It has been shown that child and adolescent subjective well-being tends to be moderately stable across time (Luhmann 2017) and that, similar to adults, most children and adolescents around the world view their overall lives positively, although there are notable differences across cultures (Gilman and Huebner 2003; Proctor et al. 2009). Studies in Russia suggest that the level of life satisfaction in Russian adolescents and young adults was lower than in many other countries (Balatsky and Diener 1993; Currie et al. 2012). For example, in the 2009/2010 Health Behaviour in School-aged Children Survey (HBSC) conducted in 39 countries, the Russian Federation was in the bottom of the rankings for the prevalence of high life satisfaction (a score of 6 or more on a scale ranging from 0 to 10) in 11- and 13-year-olds, although it was in the middle of the rankings for 15-year-olds (Currie et al. 2012).

With respect to demographic predictors, similar to adult findings, child and adolescent life satisfaction differed only slightly across age and gender; it tended to decline in adolescence (Proctor et al. 2009) and in some countries was somewhat higher in males (Gilman and Huebner 2003). Life satisfaction is only weakly, albeit significantly, correlated with socio-economic status (parental education, occupational status and income) and family composition (parental marital status, a number of other children and adults living in the home). Several studies have shown that life satisfaction is higher in children from higher socio-economic groups and in children living with married parents and that, overall, parental separation, divorce, and remarriage are associated with lower child and adolescent life satisfaction (Bjarnason et al. 2012; Gilman and Huebner 2003; Proctor et al. 2009; Dinisman et al. 2012).

Apart from the findings on the role of demographic and socio-economic characteristics in defining subjective well-being, much of the research with children has evaluated social relationships with adults and peers and other family variables such as parenting practices. According to the bioecological model of human development (Bronfenbrenner 1999), the former are distal environmental factors that the child does not directly encounter, while the latter are proximal environmental influences that that directly impinge on the child. Consequently, it is suggested that the effects of more distal factors operate through the influence of more proximal factors which directly shape child development (Bornstein 2012). Empirical evidence favors this assumption showing that family experiences, relationships and parenting style are more strongly related to child and adolescence life satisfaction than demographic and socio-economic characteristics (Holder and Coleman 2009; Huebner 1991b; Rees et al. 2012).

For example, in a national study of 9-year-old children in Ireland, family stressors were found to explain more than twice the variance in the children’s happiness than explained by socioeconomic status (McAuley and Layte 2012). The preceding review indicates that poor parental relationships and lack of paternal involvement has a greater negative effect on adolescent life satisfaction than family status (Proctor et al. 2009). The findings from of a representative sample of 11–13-year olds in Norway showed that parental emotional well-being measured by a scale, which contains items about depression and anxiety, and parenting practices may mediate the association between family socioeconomic status and child mental health (Bøe et al. 2014). Overall, research in Western and East Asian cultures has shown that child and adolescent life satisfaction is related to parents’ life satisfaction, perceived parental characteristics and the quality of family relationships; it is positively associated with warm and supportive parenting and negatively with dysfunctional parenting (Casas et al. 2008; Chang et al. 2003; Proctor et al. 2009; Suldo and Huebner 2004).

Although adult findings indicate that personality accounts for most of the variance in subjective well-being and/or life satisfaction (Diener et al. 2003; Steel et al. 2008), child studies rarely address individual characteristics, apart from gender and age. Existing evidence suggests that happiness in children and adolescents is negatively related to neuroticism and to a lesser degree and positively, extraversion (Gilman and Huebner 2003). Given that extraversion and neuroticism are essentially equivalent to positive emotionality and negative emotionality (Markon et al. 2005; Steel et al. 2008), it should perhaps come as no surprise. The findings on the effect of other personality traits are more scarce. A few recent studies using self-reports on the Big Five traits show that while neuroticism is the strongest predictor of adolescent life satisfaction, all other traits may also make a contribution (Goswami 2014; Suldo et al. 2015; Weber and Huebner 2015). In addition, child and adolescent life satisfaction has been consistently associated with other personality characteristics such as self-esteem and locus of control (Gilman and Huebner 2003; Proctor et al. 2009).

Life satisfaction can be reliably measured in children from 8 years of age (Huebner 1991a) and child self-report measures of life satisfaction display substantial validity (Gilman and Huebner 2003; Casas et al. 2013), however, there is a lack of research on life satisfaction of preadolescent children. A few studies comparing adolescents with younger children have found both similar and different correlates (Gilman and Huebner 2003). For example, perceived social competence was a strong predictor of life satisfaction in Chinese adolescents, but not in primary schoolchildren (Chang et al. 2003). Overall, family factors, such as parenting, parental mental health and family relationships may be more important for younger children than for adolescents (Holder and Coleman 2009; Gilman and Huebner 2003; Parkes et al. 2016). Evidence from the studies of adult and adolescent life satisfaction suggests that personality traits should also play an important role, but, as far as we know, no study has examined personality effects on life satisfaction of preadolescent children. To address this gap, this study aims to examine the contribution of family environment and personality to the prediction of life satisfaction in primary schoolchildren. Because adult findings suggest that inclusion of lower-level traits could increase the knowledge base of specific personality influences on life satisfaction (Schimmack et al. 2004), we studied personality traits at the higher- and lower-order levels. Addressing these issues in the Russian cultural setting, this study will contribute to a deeper understanding of child life satisfaction within a cross-cultural perspective.

Regarding family factors, we hypothesized that living with both biological parents, higher parental education, affluence, good family relationships and parental warmth and involvement would be positively related to child life satisfaction, whereas high levels of parental stress, harsh parenting and poor supervision would be negatively related. We also expected that the contribution of proximal family factors would be larger than that of distal family factors. Regarding personality, we expected that neuroticism would be negatively related to child life satisfaction, whereas extraversion would be positively related. Because of the scarce data, no hypotheses were made about other personality traits and about the relative contribution of personality and family environment. We explore these issues by gathering data from parents and children within a large community sample. This study also aims to add to the growing literature on child well-being in different countries providing evidence from the Russian cultural context.

2 Method

2.1 Participants and Procedure

Data were collected in two Siberian cities. Most of the participants (68%) were from Novosibirsk, Russia’s third largest city with a population of more than 1.5 million, and the economic and academic capital of Siberia. Others were from a smaller city, Novokuznetsk, a coal mining and industrial center with a population of about 0.5 million. This study used school-based sampling method. Following approval from the Institute of Physiology and Basic Medicine Ethics Committee, non-selective municipal schools were invited to participate. Five agreed, and named a teacher responsible for administration. Parents of primary schoolchildren (grades 1–4) were informed about the study, invited to participate, and provided with questionnaires. The following questionnaires were administered to parents: the Inventory of Child Individual Differences-Short version (ICID-S), the Alabama Parenting Questionnaire-Brief Form (APQ-BF), the Self Reporting Questionnaire (SRQ) and Family Life Style Questionnaire. With parental permission, children were asked to complete Huebner’s Student Life Satisfaction scale (SLSS). Parents completed questionnaires at home, children were assessed during the school day at school. Data were collected on 705 children, the response rate was about 66% of the eligible sample. Participants represented diverse socioeconomic backgrounds.

Average age of the sample was 9.1 (SD 1.2), 96% of children were from 7 through 10 years, 51% were female. Parent reports were available for 503 children, most data came from mothers (86%), fathers rated 11% of children, and other close adults rated the rest. Most of the children (70%) lived with both biological parents, 19% with a single mother, 10% with a mother and a stepfather, and the rest with other carers; 58% had one or more siblings. Twelve percent of the mothers and 12% of the fathers had ten years of schooling, 25% of the mothers and 30% of the fathers have completed vocational college, and 61% of the mothers and 54% of the fathers had university degree. For occupation, the parents ranged from unskilled to professional workers, 17% of the mothers and 3% of the fathers were unemployed.

2.2 Missing Data

Life satisfaction data were available for 649 children, 92% of the original sample. The reasons for missing SLSS data were the child missing school, refusals or invalid responses. We performed a missing value analysis using the expectation maximization algorithm in SPSS Version 22.0. Little’s (1988) Missing Completely at Random (MCAR) test was not statistically significant, χ2 (35) = 46.87; p = .087, therefore, it may be assumed that SLSS values were missing randomly. In addition to using Little’s MCAR test, potential differences between participants with and without SLSS data for all of parent-reported measures were examined via independent t-tests. The differences were nonsignificant, with the exception of age: children with SLSS data were slightly younger (M = 9.1, SD 1.2 years) than those without (M = 9.6, SD 1.2 years, t(703) = 3.24, p = .001. These findings also suggest that there were no systematic reasons for missing data. Because the sample size was sufficiently large, the level of missing data was low and the SLSS data may be assumed to be MCAR, the present study included only participants with no missing SLSS data. Of these, parent reports were available for 503 children (79%); exact sample sizes for each measure are given in Table 1. In the regression analyses, missing data were deleted listwise, resulting in a sample of N = 424.

2.3 Measures

2.3.1 Life Satisfaction

In this study, a 7-item Huebner’s Students’ Life Satisfaction Scale (SLSS, Huebner 1991a) has been used as a measure of child life satisfaction. The SLSS purports to measure global life satisfaction in children ages 8–18; it consists of the following items: (1) My life is going well; (2) My life is just right; (3) I would like to change many things in my life; (4) I wish I had a different kind of life; (5) I have a good life; (6) I have what I want in life; (7) My life is better than most kids. Children are asked to respond to each item on a six-point scale ranging from ‘strongly agree’ to ‘strongly disagree’. Possible scores for the SLSS range from 7 to 42, with a higher score indicating a greater level of satisfaction in life. Findings provide evidence for good psychometric properties of the SLSS, demonstrating convergent, discriminant and criterion-related validity and internal consistency reliability in the .70–.80 range across all age groups (Huebner et al. 2005). The SLSS statements were translated from English into Russian for the purposes of this study, adhering to the recommended procedures (Peña 2007) that include translation followed by back-translation, analyses of discrepancies and subsequent revision of the Russian version; in the present study alpha for the SLSS was .74.

2.3.2 Family Environment

Life Style Questionnaire (Slobodskaya et al. 2008) was designed to assess socio-economic status (SES) and family factors associated with child mental health and well-being. Items on demographic characteristics included child’s gender, age, family size and structure (step-, lone- or two-parent), parental education (1 = secondary school, 2 = vocational college, 3 = university degree, 4 = postgraduate degree) and occupation (0 = unemployed, 1 = unskilled workers, 2 = skilled manual workers, 3 = specialists, 4 = professional workers, 5 = administrative staff). Fathers’ and mothers’ education and occupation scores were averaged to generate parental education and occupation scores. Family income was measured as subjective assessment of purchasing power on a 5-point scale ranging from ‘buying food, clothes and shoes causes difficulties’ to ‘can buy everything’.

Family cohesion was measured by a 5-item scale (α = .65) including one item about parent–child relationships rated on a 4-point scale ranging from ‘not so good’ to ‘very good’, and four items from the General Functioning Scale of the MacMaster Family Assessment Device (FAD-GFS, Miller et al. 1985) rated on a 4-point scale ranging from ‘strongly agree’ to ‘strongly disagree’: “Planning family activities is difficult because we misunderstand each other”; “In times of crisis we can turn to each other for support”; “We don’t get along well together” and “We confide in each other”. The original FAD-GFS scale has good psychometric properties (Miller et al. 1985); the abbreviated Russian version used here has been shown to have good levels of reliability and validity correlates with measures of child mental health (Slobodskaya et al. 2008). Domestic violence was measured by two items asking whether the child has witnessed any quarrels between the adults in the family and if so, have the quarrels involved physical aggression; responses were rated on a 3-point scale ranging from ‘no quarrels’ to ‘quarrels with physical aggression’.

Parental stress was measured by the Self Reporting Questionnaire (SRQ, WHO 1994) including 20 yes/no items about anxiety and depression (α = .79). The SRQ has been shown to have good psychometric properties in studies from around the world (WHO 1994); a study of Russian child mental health provided support for the validity of the Russian version in a stratified random sample of 7- to 14-year-old schoolchildren (Goodman et al. 2005). Parenting was measured by the Russian version of the Alabama Parenting Questionnaire-Brief Form (APQ-BF, Scott et al. 2011) that includes 15 items covering five empirically identified aspects of positive and negative parenting practices for school-age children: positive parenting, involvement, corporal punishment, poor supervision and inconsistent discipline (of 3 items for each). Responses are rated on a 5-point scale ranging from ‘never’ to ‘always’. The Russian version of the APQ-BF has been validated in a community sample (Loginova et al. 2016), supporting a five-factor structure and good internal consistency, discriminant and criterion validity of the scales. In the present study alphas for APQ-BF scales ranged from .40 to .73 with a mean of .58.

2.3.3 Personality

The Inventory of Child Individual Differences-Short version (ICID-S, Slobodskaya and Zupančič 2010) is an age and culture neutral instrument for assessing child personality in terms of the five factor model. The ICID-S for parents was developed from the full instrument (Halverson et al. 2003) in English (Deal et al. 2007) and the Slavic languages (Slobodskaya and Zupančič 2010). The Slavic version of the ICID-S includes 62 items measuring five higher-order and fifteen lower-order traits: Extraversion (sociable, activity level, positive emotions), Neuroticism (fearful, shy, negative affect), Conscientiousness (achievement orientation, organized, compliant, distractible), Agreeableness (antagonism, strong willed, considerate), and Openness (open to experience, intelligent). The Slavic ICID-S has been validated, supporting good reliability and validity of the scales. In the present study alphas for ICID higher-order scales ranged from .70 to .91 with a mean of .86, alphas for lower-order scales ranged from .70 to .84 with a mean of .78.

3 Results

3.1 Child Life Satisfaction

We tested the structure of the Russian version of the SLSS using confirmatory factor analysis (CFA) from AMOS.17 software (Arbuckle 2008). The results have shown that the one-factor model with seven indicators and one correlated residual (between negatively worded items 3 and 4) demonstrated good fit to the data: χ2 = 33.55; df = 13; p = .001; CFI = .981; RMSEA = .049. The internal consistency (Cronbach’s alpha) of the scale was α = .74. This supports the use of the SLSS scores to measure child life satisfaction. In our sample, scores ranged from 11 to 42 with a mean of 33.4 (SD 6.7). Seventy-two children (11%) had the highest score of 42, whereas 14 children (2.2%) scored lower than 16 and could be considered outliers. However, in this paper, we were interested in the whole range of Student Life Satisfaction scores and retained these low scorers in the sample. We screened for multivariate outliers in the regression analysis, using Cook’s test with a cutoff point of 1 (Stevens 1984). Mean Cook’s distance was .003 (range .000–.035), indicating that there were no potentially influential outliers.

The effect of demographic variables was assessed using the analyses of variance; the effect size was estimated by Cohen’s d. Results showed that there were significant gender differences in child life satisfaction, F = 10.35, p < .001, with girls (M = 34.14, SD 6.22) scoring higher than boys (M = 32.49, SD 7.10). Cohen’s d = 0.25 indicated that the effect of gender on child life satisfaction was small; effects of age and gender–by-age interactions were not significant. Family type was also a significant factor, F = 5.88, p < .01: children from intact families (M = 34.27, SD 6.26) scored higher than children from step-parent (M = 31.80, SD 7.33) and single-parent families (M = 32.17, SD 7.27), the effect sizes were small (Cohen’s d = 0.36 and 0.31, respectively).

3.2 Relations of Child Life Satisfaction with Family Environment and Personality

Table 1 presents descriptive statistics for family and personality variables; Table 2 presents correlations between all the study variables. Child life satisfaction was positively related to parental education, income and family cohesion, and was negatively related to domestic violence, parental stress, corporal punishment and poor supervision. Child life satisfaction was also associated with all Big Five personality traits; correlations with extraversion, conscientiousness and openness domains were positive, whereas correlations with disagreeableness and neuroticism domains were negative. Correlations with lower-order traits followed the same pattern as the factor they defined, but were generally smaller. Child life satisfaction was not significantly correlated with traits of sociable and open to experience.

Table 1 Descriptive statistics for family and personality variables
Full size table
Table 2 Correlations between the study variables
Full size table

3.3 Contribution of Family Environment and Personality to Child Life Satisfaction

We examined the contribution of family factors and child personality to the prediction of child life satisfaction in a series of hierarchical multiple regressions. Multiple regression analysis is a best choice when “a quantitative variable, the dependent variable (Y), is to be studied as a function of, or in relationship to, any factors of interest, the independent variables (IVs)” (Cohen et al. 2003, p. 1). It has been shown “to be peculiarly appropriate for the behavioral sciences in its capacity to accommodate … the multiplicity and correlation among causal influences” (Cohen et al. 2003, p. 18). The multiple regression equation is Y = a + b1x1 + b2x2 + ··· + bkxk, where Y is a dependent variable; a, b1, … bk are constants, x1, … xk are predictors, and k is the number of predictors. In each of the models, gender and age were entered first as fixed variables; for the subsequent predictors, stepwise method was used. In the first two models, distal family factors (family type, parental education and occupation, and income; Step 2) and proximal family factors (family cohesion, domestic violence, parental stress and parenting; Step 3) were followed by personality (Big Five in model 1 and lower-order traits in model 2; Step 4). In models 3 and 4, personality traits (Big Five in model 3 and lower-order traits in model 4) were entered as Step 2 variables, distal family factors were entered next (Step 3) and proximal family factors were entered last (Step 4). Thus, in models 1 and 2 we were able to test whether personality would predict child life satisfaction beyond family environment, whereas in models 3 and 4 we investigated the predictive power of family variables above and beyond personality. The contribution of child personality was estimated both on the higher- (models 1 and 3) and the lower- (models 2 and 4) level of the hierarchical structure.

Table 3 presents the findings from the multiple regression analyses in which family factors and child personality were used to predict child life satisfaction. In all models, both distal and proximal family environment made a significant contribution to child life satisfaction. The significant predictors were income, parental stress, and poor supervision, together accounting for 7–12% of the total variance. In models 1 and 2, proximal family factors accounted for more of the variance in child life satisfaction than distal family factors (6 and 4%, respectively); in models 3 and 4, however, after personality has been taken into account, their contribution was approximately equal (3%). In all models, personality traits from the domains of neuroticism and conscientiousness contributed significantly to the prediction of child life satisfaction, accounting for 4–9% of the total variance. In models 1 and 2, after family environment has been taken into account, lower-order traits of achievement orientation and shyness made a significant contribution. In models 3 and 4, however, the significant predictors were the other two traits from the same personality domains, compliance and fear. Overall, the models explained 14–15% of the total variance in child life satisfaction.

Table 3 Hierarchical multiple regression indices predicting child life satisfaction
Full size table

4 Discussion

The findings of this study suggest that the Russian version of the Students’ Life Satisfaction Scale (SLSS, Huebner 1991a) is both reliable and valid. The results of confirmatory factor analyses of 7-item SLSS supported the unidimensional structure of global life satisfaction in Russian children aged 7–10 years, consistent with the findings for the original SLSS (Huebner 1991a). It is also worth noting that, contrary to earlier findings for the satisfaction with life scale in Russian adolescents (Balatsky and Diener 1993), internal consistency reliability of the Russian version of the SLSS (.74) was in the range reported for the original scale (Huebner et al. 2005). Our findings therefore confirmed that this instrument has good psychometric properties and may be used as a brief measure of child global life satisfaction in large-scale studies, thus allowing cross-cultural comparisons and greater generalization of findings.

Similar to previous findings from other countries (Gilman and Huebner 2003; Huebner et al. 2005; Proctor et al. 2009) and contrary to previous research with Russian adolescents and young adults (Balatsky and Diener 1993; Currie et al. 2012), most Russian children in our study were satisfied with their lives: more than 80% of 7–10-year-olds scored in the positive range (28 or more). However, around 7% were, on average, moderately to mildly dissatisfied with their lives (scored 21 or less). The mean SLSS score in our study was higher than in a Suldo and Huebner’s (2004) study of 11–19-year-olds from South Carolina, USA; however, although both studies used the same scale, the findings are perhaps not directly comparable, given age differences between the two samples. While cross-cultural differences in child life satisfaction await further research, we turn now to the factors which contributed to the overall level of life satisfaction in our study.

In line with a variety of other studies (Currie et al. 2012; Gilman and Huebner 2003; Proctor et al. 2009), the effects of gender and age were not that substantial, accounting for less than 2% of the variance in primary schoolchildren’s life satisfaction. Family structure was a slightly more influential factor: similar to children and adolescents from other countries (Bradshaw et al. 2011; Dinisman et al. 2012; Rees et al. 2012), Russian 7–10-year-olds living with both biological parents were more satisfied with their life than children living in other family arrangements. These findings are consistent with our predictions and the robust evidence that children living with both biological parents tend to have better outcomes than children in step-parent or single-parent families (Thomson and McLanahan 2012), although effect sizes are small. Confirming our hypotheses, we found that parental education and family income were also associated with child life satisfaction, thereby replicating and extending previous findings (Casas et al. 2013; Currie et al. 2012; Parkes et al. 2016; Rees et al. 2012) to Russian primary schoolchildren. The strength of the association between income and life satisfaction in our study was close to that reported for adults in meta-analyses and large-scale surveys (Lucas et al. 2008). It should be noted, however, that some studies have reported no differences in child and adolescent life satisfaction between socio-economic groups (Gilman and Huebner 2003).

The relations of family cohesion, domestic violence, parental stress, and parenting practices with child life satisfaction were largely consistent with our hypotheses and the existing research on the role of proximal family environment. The prominent role of parental stress in child life satisfaction and well-being is in accordance with findings already presented in the literature (Bøe et al. 2014; McAuley and Layte, 2012). Our findings suggest that, among other aspects of parenting, supervision and monitoring of child activities is the most important contributor to life satisfaction in Russian primary schoolchildren. This is consistent with previous research that has also found that neglectful parenting with poor supervision is an important negative predictor of child subjective well-being (Parkes et al. 2016; Suldo and Huebner 2004) and is associated with a plethora of adverse outcomes for children (Crouter and Head 2002). It is also notable that, contrary to expectations, positive parenting and parental involvement did not appear to be associated with child life satisfaction. Although one study from Hong Kong did report an association between parental warmth and life satisfaction, both in primary schoolchildren and in adolescents (Chang et al. 2003), evidence supporting a link between quality of parenting and child subjective well-being (Proctor et al. 2009) largely comes from studies investigating the role of different parenting types based on the levels of two main dimensions of warmth and control (Smith 2011). Therefore, it is unclear, whether child life satisfaction is related to parental control or warmth, or some combination of these two dimensions of parenting. The strength of the present study is its relative comprehensiveness in measuring parenting. Still, the findings await replication and the specific contribution of parenting dimensions and practices to child life satisfaction in different cultures remains to be investigated.

It is widely held that family relationships are more important for subjective well-being and/or life satisfaction than demographic and socio-economic factors at all ages, but particularly in children (Diener 1984; Diener et al. 2002; Proctor et al. 2009; Rees et al. 2012, but see also Lucas et al. 2008). In this study, family structure and parental education did not explain significant variance in child life satisfaction when more proximal family factors were taken into account, supporting the bioecological model wherein distal environmental factors exert their influence on the developing child through the effects of more proximal factors. Family income, however, remained a significant predictor even when proximal environment and personality were taken into consideration. This is consistent with findings on adult and adolescent subjective well-being. Although social relationships are often considered to be the most important predictors of happiness, whereas affluence is considered to be relatively unimportant, Lucas and colleagues (2008) reviewed the empirical evidence and concluded that the effect size of social relationships on adult subjective well-being is often similar to the effect size of income. In a national survey of 10–15-year olds in Great Britain, material deprivation had the largest effect on young people’s life satisfaction among other socio-demographic factors and remained a significant predictor even when personality has been taken into account (Goswami 2014). It is important to note that most research on adult and adolescent subjective well-being has primarily relied on self-reports to assess both predictors and outcomes. Thus, it is possible that the effects found in previous studies may partly reflect inflation because of shared method variance. This study avoids the problems associated with shared method variance using parent reports to assess the predictors (family environment and child personality) and self-reports to assess child life satisfaction which was the outcome studied.

To our knowledge, this is the first study to examine the relationship between life satisfaction and personality traits in primary schoolchildren. In line with expectations, the results showed that both neuroticism and extraversion were related to child life satisfaction. Similar to previous findings with adults (DeNeve and Cooper 1998; Steel et al. 2008) and adolescents (Gilman and Huebner 2003; Goswami 2014; Suldo et al. 2015; Weber and Huebner 2015), neuroticism was the strongest (negative) predictor of child life satisfaction. The association with extraversion was much smaller, and this trait did not emerge as a significant predictor. Interestingly, however, the results for the lower-order traits comprising extraversion domain showed that positive emotions correlated modestly with child life satisfaction, whereas the correlation with sociability was not significant. Although many adult studies report correlations between measures of sociability and ratings of happiness and well-being (Lucas et al. 2008), there is also evidence that sociability does not add to the prediction of life satisfaction when the depression facet of neuroticism and the positive emotions facet of extraversion are taken into account (Schimmack et al. 2004).

Conscientiousness was the strongest positive predictor of child life satisfaction, which should not perhaps be unexpected as adult studies using the five factor model of personality consistently find that conscientiousness is one of the main predictors of life satisfaction (DeNeve and Cooper 1998; Steel et al. 2008). The relationship between conscientiousness and life satisfaction in adults is mainly explained by instrumental effects of work achievement, accomplishments and goal efficacy (DeNeve and Cooper 1998; Hayes and Joseph 2003; McCrae and Costa 1991). It is, therefore, somewhat surprising that the contribution of conscientiousness to life satisfaction is very similar in primary schoolchildren. One longitudinal study found that conscientiousness predicted academic satisfaction which in turn predicted life satisfaction in 21-year-old university students (Schimmack et al. 2002). There is also evidence that longitudinal relations between the Big Five and life satisfaction are remarkably consistent from age 15–90 (Soto 2015). In two recent studies, conscientiousness was one of the strongest predictors of adolescent life satisfaction (Suldo et al. 2015; Weber and Huebner, 2015). The authors suggested that this effect was due to enjoyment of academic-oriented activities and fulfilment of norms and standards by conscientious adolescents. The findings of the present study extend prior research on the relationship between conscientiousness and subjective well-being into childhood.

Cross-cultural research has shown that parents describe their children by traits linked with conscientiousness (e.g., good concentration, reliable, and hard working or forgetful, careless, and lazy) from as early as age 3 years and that they use such descriptors much more often by age 6 years (Kohnstamm et al. 1998). It is reasonable to assume that the link between conscientiousness and life satisfaction may be mediated by parental praise. Our findings, however, do not support this assumption, showing that conscientiousness made an independent contribution to life satisfaction, whereas positive parenting involving praise and recognition for good work and good behavior (Scott et al. 2011) was not related to child life satisfaction at all. Still, conscientious children who are responsible, attentive, persistent, orderly and neat, possessing high standards and thinking before acting (Caspi and Shiner 2006), are likely to evoke more positive reactions from others than those who are irresponsible, unreliable, careless, distractible and quitting easily, regardless of whether parents use praise often or not. Because life satisfaction represents the cognitive/evaluative aspects of subjective well-being (Diener 1984; Diener et al. 2002; Hayes and Joseph 2003; Proctor et al. 2009; Gilman and Huebner 2003), it is likely that children who get more positive evaluation from others will view their overall lives more positively than those who receive more negative feedback.

It is worth noting that although agreeableness and openness were correlated with child life satisfaction, they did not make a significant contribution when all Big Five traits were taken into account. The evidence on the role of these two personality traits in subjective well-being and/or life satisfaction is inconsistent (Goswami 2014; DeNeve and Cooper 1998; Soto 2015; Suldo et al. 2015; Steel et al. 2008; Weber and Huebner 2015); it might be supposed that the impact of agreeableness and openness may differ across ages and cultures, but this remains to be investigated. It is also notable that in our study lower-order traits did not show greater predictive validity than the Big Five. This is in contrast to previous adult studies that have reported that personality facets accounted for more of the variance in happiness and life satisfaction than the Big Five (Quevedo and Abella 2011; Schimmack et al. 2004). Overall, personality accounted for 4–9% of the total variance, which is less than in most other studies (Goswami 2014; Hayes and Joseph 2003; Suldo et al. 2015; Steel et al. 2008, Weber and Huebner 2015, but see also DeNeve and Cooper 1998). These discrepancy between the results obtained with self-reports from adults and adolescents and our study that used different informants to measure child life satisfaction and personality highlights the need for further research on the role of personality and environmental factors in life satisfaction across ages and countries.

Several limitations of this study deserve comment. First, we used a brief unidimensional scale for child life satisfaction because of primary schoolchildren’s limited ability to provide information on themselves; however, the study would benefit from a more comprehensive assessment of child life satisfaction. Second, although the sample was reasonably diverse in terms of socio-demographic characteristics, such as family structure, education, occupation and income, it would be preferable to carry out further studies on a larger scale and to use more elaborated sampling methods. Third, in addition to measures of family environment, it would be useful to include measures of school environment because of the length of time children spend there. Although the findings with primary schoolchildren are limited, there is evidence that school environment is associated with life satisfaction in adolescents (Currie et al. 2012; Gilman and Huebner 2003; Rees et al. 2012). Fourth, some scales had low internal consistency, limiting our chances of finding significant results. Fifth, the cross-sectional design did not allow an exploration of the developmental pathways and causal influences; the findings should be supported by longitudinal data. Finally, it is important to remember that children with similar personalities, as well as children from similar backgrounds, can develop quite differently because of interactions between characteristics of the child and the environment (e.g., Masten 2001; Pluess and Belsky 2013). While only main effects were considered in this study, we believe that, if these first findings on the role of personality traits in primary schoolchildren’s life satisfaction can be replicated, then they might provide the basis for investigation of interactive effects of personality and environmental factors on child life satisfaction.

In conclusion, this study provided evidence for the validity and reliability of the Russian version of Huebner’s Students’ Life Satisfaction scale in a community sample of Russian primary schoolchildren. The implication is that this internationally recognized scale can be used in research and practice in Russia. Although life satisfaction is a desirable outcome in and of itself, it is also associated with a plethora of positive developmental outcomes, whereas children with low levels of life satisfaction are at increased risk of a variety of negative outcomes, including mental and physical health problems (Huebner et al. 2005). Thereby, measuring life satisfaction may prove useful in the identification and monitoring of at-risk students. The study has also examined the role of demographic and socio-economic factors and revealed that family income, parental stress and parental supervision were the most powerful predictors, together accounting for 11% of the variance in child life satisfaction. Child personality traits, as assessed by parent reports, showed modest to moderate correlations with life satisfaction, the regression analysis indicated that low neuroticism and conscientiousness were the only two factors found to be significantly and independently associated with child life satisfaction, accounting for about 9% of the variance. The practical implications of these results pertain to potential prevention and intervention programs for supporting families with children and developing parenting skills that promote child life satisfaction. Overall, the findings provide a strong basis for further investigation of the role of personality and family environment in child life satisfaction.