1 Introduction

Subjective well-being can be defined as an individual assessment of life, which includes two elements: emotion and satisfaction (Petito & Cummins, 2000). Life satisfaction can be understood as an individual and global cognitive evaluation that people make of their own life (Diener et al., 2015). It is a field of research that has grown in recent years, being considered a key element of healthy growth in children and adolescents, as well as being linked with better physical and mental health, educational success, and good interpersonal relationships (Park et al., 2014; Yoo & Ahn, 2017; Zappulla et al., 2014). Life satisfaction has been associated with multiple psychological, behavioral, social, and intrapersonal variables, including social and personal resources, as well as individual aptitudes and contextual variables (Veenhoven, 1994).

Within the social component, an important aspect in the perception of well-being, as well as in the psychological and physiological functioning of children and adolescents, is played by their ability to perform adequately in their interpersonal relationships, mastering skills such as effective communication, assertiveness, empathy, respect, and prosocial attitudes. This is defined in the literature as integration and social competence, or simply social competence (Bisquerra & Pérez, 2007); the latter form will be used in this study. Studies have shown an association between perceived social competence and higher levels of emotional regulation, better academic performance, adequate coping strategies, and a healthier self-concept among the child and adolescent population (Arslan, 2019; Franco et al., 2017).

Among the personal resources associated with life satisfaction, resilience and self-esteem stand out. Resilience can be understood as an individual’s capacity to overcome adverse situations using psychological, social, and cultural resources, allowing individuals to maintain adequate development and well-being (Liebenberg et al., 2013; Masten & Cicchetti, 2016). Self-esteem, meanwhile, is defined as the positive or negative evaluation that individuals make about themselves (Williams, 2009).

According to the literature, both resilience and self-esteem are associated with social competence in children and youth. Studies have shown that those who are less integrated, or excluded by their peers, have lower levels of resilience (Arslan, 2019; Waldeck et al., 2015), while those who perceive greater competence at the social level are more resilient because positive social bonds increase resilience (Childs et al., 2001; Libório & Ungar, 2014). Similarly, perceived social competence has a significant effect on self-esteem (Bedard et al., 2020) because it is influenced by experiences with others (Williams, 2009); hence, it can lead to scenarios where those with fewer abilities to perform socially present lower self-esteem (Dembińska et al., 2020).

Regarding the association between resilience, self-esteem, and life satisfaction, a recent study conducted with Australian university students found that resilience acts as a differential protective factor against mental health problems such as depression, anxiety, and stress (Anyan et al., 2020). Other studies have reported similar findings in cases of exposure to stress (Shi et al., 2015), health problems (Gooding et al., 2012), and discrimination (Caqueo-Urízar et al., 2021). Likewise, high levels of self-esteem have been related to high levels of life satisfaction, partially reducing the negative effects of adverse events (Arslan, 2019).

Thus, social competence and social integration have a significant effect on self-esteem and resilience. In addition, these variables affect the perception of life satisfaction. In this sense, a study conducted in Turkey found that both resilience and self-esteem fully mediated the relationship between social exclusion and life satisfaction, mitigating the negative effects of social exclusion on adolescents’ life satisfaction (Arslan, 2019). These variables are thus key factors in any intervention programs targeting adolescent life satisfaction.

However, while research on life satisfaction has grown substantially in recent years, research in developing countries, such as Chile, are still scarce and incipient (Ungar et al., 2013). As a member of the International Survey of Children’s Well-being (ISCWeB), Chile’s results indicate that its children and adolescent populations present high levels of life satisfaction (Alfaro et al., 2020). These levels are corroborated by research on the subjective well-being of children and adolescents in the country, which also found significant relationships between life satisfaction, social competence, and the sociocultural context of the students (Varela et al., 2019; Varela et al., 2020; Varela et al., 2021).

Among demographic variables, studies show that as age increases, life satisfaction tends to decrease (Goldbeck et al., 2007; Haranin et al., 2007; Soares et al., 2019). This relationship has also been observed in Chile, either as a negative association between age and life satisfaction (Varela et al., 2019; Varela et al., 2021), or as a decrease in life satisfaction scores in the different age ranges studied in ISCWeB (Alfaro et al., 2020).

Regarding gender, the differences between boys and girls worldwide are not clear (Jacobs Foundation, 2020). In Chile, the literature suggests that girls are more likely to experience psychological issues such as low self-esteem and psychological victimization than boys, which are likely to affect their life satisfaction (Instituto Nacional de la Juventud [INJUV], 2019; Varela et al., 2021).

Although these findings are a valuable tool for addressing the subjective well-being of children and adolescents in Chile, the samples come from only two of the country’s 16 administrative regions, where daily and cultural realities are different from other areas of the country (Alfaro et al., 2020; Varela et al., 2019; Varela et al., 2020; Varela et al., 2021). Therefore, this study seeks to generate more evidence and information on variables related to adolescent well-being from different sociocultural and contextual realities of Chile. The study of this relationship is important because difficulties at the interpersonal level are a source of significant psychological distress and can help in promoting adolescent social competence, which affects other psychosocial variables relevant to students’ development, such as resilience and self-esteem. On the other hand, the strengthening of resilience and self-esteem is equally important, since they are key factors for healthy adolescent development (Zaff & Hair, 2003).

Thus, the aim of this study is to analyze the mediating role of self-esteem and resilience in the relationship between social competence and life satisfaction in adolescents from northern Chile.

2 Methods

2.1 Design

A non-experimental study with a predictive cross-sectional design was conducted, since all variables were measured at a single point in time and the purpose of the study was to explore the functional relationship by predicting a criterion variable from one or more predictors (Ato et al., 2013).

2.2 Participants

The sample consisted of 2277 students from seventh grade to senior-year students from 25 educational establishments in the city of Arica, Northern Chile. Students’ ages ranged from 12 to 18 years, with a mean of 14.4 years (SD = 1.7); 50.4% (n = 1148) were girls, while 49.6% (n = 1129) were boys. Most of the educational establishments belonged to a population of medium and/or low socioeconomic levels.

2.3 Instruments

2.3.1 Ad-Hoc Sociodemographic Scale

Including gender, age, and grade of the participants.

2.3.2 Child and Youth Resilience Measure (CYRM-12; Liebenberg et al., 2013)

Being a brief version of the original CYRM-28 scale (Ungar & Liebenberg, 2011), it is composed of 12 Likert-type items with five alternatives from 1 (not at all) to 5 (very much), seeking to assess key characteristics of resilience to adversity in children and adolescents based on the interaction of individual, relational, community, and cultural factors (e.g., “I try to finish what I start,” and “My family is by my side in difficult times”). Higher scores indicate higher levels of resilience. Llistosella et al. (2019) translated and validated a 32-item version of the scale into Spanish, reporting evidence of validity based on the internal structure of the test and satisfactory reliability coefficients (α > .8). Cronbach’s alpha for the present study was .78.

2.3.3 Child and Adolescent Assessment System (Sistema de Evaluación de Niños y Adolescentes; SENA; Fernández-Pinto et al., 2015)

An instrument developed by specialists in psychopathology and psychological assessment, its purpose is to measure a wide range of emotional and behavioral problems in youth aged 3 to 18 years. The items have a 5-level Likert scale response format, ranging from 1 (never or almost never) to 5 (always or almost always). From the complete self-report version for high school (12–18 years old), this study used only the Self-Esteem, Integration, and Social Competence scales. Higher scores indicate that participants have a set of skills that make them competent in the area of interpersonal relationships and ability to recognize their personal worth. Recently, Sánchez-Sánchez et al. (2016) found that the reliability of its subscales was above .7. The reliability of the self-esteem scale in the present study was .68, while the reliability of the social integration and social competence scale was .73.

2.3.4 The Satisfaction with Life Scale-Child

(SWLS-C; Gadermann et al., 2010) is a brief 5-item scale adapted from Diener et al.’s (1985) original version, and designed to assess life satisfaction in children in a one-dimensional way. This scale contains items that assess, by means of global cognitive judgments, the degree of satisfaction that each child presents about their own life (“I am happy with my life,” and “In most ways my life is close to the way I would want it to be”). The response options were 5-point Likert behavioral/attitudinal statements (1 = “strongly disagree” to 5 = “strongly agree”). The scores were obtained by calculating the mean of all the answers. High scores suggest greater satisfaction, while low scores suggest greater dissatisfaction with life. Álvarez et al. (2018) adapted it for use in Chile, and their version has an adequate reliability coefficient (ω > .8). The reliability of the present study was α = .87.

2.4 Procedure

This research was approved by the Scientific Ethics Committee of the Universidad de Tarapacá (26–2017).

The principals and counselors of 35 educational establishments in Arica were contacted and invited to participate in this study voluntarily. Municipal, private-subsidized, and privately paid schools were included. Finally, in the second semester of 2018, students from 29 schools in the city participated in the study. Informed consent was obtained from all parents, and subsequently, consent was also obtained from the adolescent participants. The instruments were applied on scheduled dates in a group classroom setting, with at least two trained surveyors present to answer any questions, together with a school teacher. The duration of the evaluation was approximately 45 min. Among those authorized, 97% agreed to answer the survey.

2.5 Data Analysis

Initially, to process the missing values, the missing value was replaced with the mean of the scale in those instruments with less than 3% missing data, leaving the definitive sample of 2277 cases.

The missing value pattern was examined and found to be missing completely at random. Consequently, the strategy of replacing the values with the mean was acceptable (Bar, 2017).

To characterize the sample, descriptive analyses were performed on categorical variables (numbers and percentages) and quantitative variables (means and standard deviations). Subsequently, the relationships between the study variables were evaluated using a Pearson correlation matrix.

We chose to use Structural Equation Modeling (SEM) instead of Hierarchical Linear Modeling because SEM allows us to analyze latent variables and their interactions in a more complex way, as well as to better include the measurement error. We rejected using a two-level SEM as the number of parameters would have suggested using a path analysis instead of SEM, but this would not allow the details of the items to be appreciated (Tarka, 2017).

In the assessment of the SEM, the classical criteria for the interpretation of model fit were taken as a reference, considering a Root Mean Square Error of Approximation (RMSEA) of less than .08 as acceptable, and less than .06 as good, while a Comparative Fit Index (CFI) and a Tucker-Lewis Index (TLI) greater than .9 valued as adequate, and above .95 as optimal (Brown, 2015; Hu & Bentler, 1999; McDonald & Ho, 2002). Although it has been reported, Chi-square was not considered because it is not currently considered an indispensable indicator given the problems it presents with samples larger than 200 and in cases of non-normal distribution (Brown, 2015; Hu & Bentler, 1999). A Standardized Root Mean-Square (SRMR) less than .8 is valued as adequate. A χ2 / df ratio is considered adequate when less than 3, although it is subject to the same limitation as Chi-square in large samples (Brown, 2015; Hu & Bentler, 1999). The Weighted Least Square Mean and Variance adjusted estimator was used, involving Likert scales of up to five categories.

Statistical analyses were performed using version 23 of the statistical package IBM SPSS Statistics (IBM Corporation, 2015) and MPLUS, version 8.6 (Muthén & Muthén, 2021).

3 Results

Table 1 presents the descriptive statistics of the studied variables. It can be observed that the levels of skewness and kurtosis are within the expected ranges, indicating that the variables are normally distributed (George & Mallery, 2010), which is a fundamental assumption for estimation methods with SEM, as its violation may cause problems regarding statistical conclusion validity (Andreassen et al., 2006).

Table 1 Descriptives of Studied Variables
Full size table

To estimate the correlations between social integration and social competence, resilience, self-esteem, and life satisfaction, a Pearson correlation matrix is presented (Table 2). It can be observed that social competence has a medium direct effect on resilience, self-esteem, and life satisfaction (r > .3; Cohen, 1988), and a small inverse effect on age (r < −.1; Cohen, 1988). Resilience has a medium direct effect on self-esteem and life satisfaction (r > .3; Cohen, 1988), and a small inverse effect on age (r < −.1; Cohen, 1988). Self-esteem has a large direct effect on life satisfaction (r > .5; Cohen, 1988), and a small direct effect on age and gender (r < .1; Cohen, 1988). Finally, life satisfaction presents a small direct effect on gender (r < .1; Cohen, 1988), and a small inverse effect on age (r < −.1; Cohen, 1988).

Table 2 Correlation Matrix Between Studied Variables
Full size table

Figure 1 shows the proposed model for explaining the association between the variables. It can be seen that integration and social competence do not directly affect life satisfaction or resilience. However, they do have a direct effect on self-esteem (b = .053, p < .05). Self-esteem has a direct effect on life satisfaction (b = .304, p < .001), as well as an indirect effect on life satisfaction via resilience (b = .305, p < .001), with a total effect of b = .608 (p < .001). Self-esteem also has a direct effect on resilience (b = .597, p < .001). In turn, resilience has a direct effect on life satisfaction (b = .511, p < .001).

Fig. 1
figure 1

Structural Equation Model of Social Competence on Life Satisfaction. Note 1: Soc = each item of social competence; Self = each item of self-esteem; Res: each item of resilience; Sat = each item of life satisfaction. Note 2: Standardized values are presented

Full size image

Among the demographic variables, gender had a direct effect on life satisfaction (b = .044, p < .01) and self-esteem (b = .189, p < .001, and a negative direct effect on resilience (b = −.084, p < .001). Age presents a direct effect only on self-esteem (b = −.090, p < .001) and resilience (b = −.078, p < .001).

The goodness-of-fit indicators for the model were optimal [χ2/gl = 5.263, RMSEA = 0.043 (95% CI: 0.041–0.045), CFI = 0.971, TLI = 0.968 and SRMR = 0.048].

4 Discussion

This study aimed to analyze the mediating role of self-esteem and resilience in the relationship between life satisfaction and social competence in a sample of adolescents from northern Chile.

No direct effects of social competence on life satisfaction were found. This may have its origins in what Proctor et al. (2009) postulated in a literature review, pointing out that adolescents’ perception of their ability to be socially competent would not directly influence satisfaction with life, but would lead them to have higher levels of sociability, which allows them to make adaptive attributions about their social performance, contributing to greater emotional stability and greater satisfaction with life (Proctor et al., 2009; Rigby & Huebner, 2005).

Similarly, research carried out in Turkey (Arslan, 2019), England (Waldeck et al., 2015), Brazil (Libório & Ungar, 2014) and Northern Chile (Caqueo-Urízar et al., 2021), found a significant association between social integration and competence and resilience levels; however, our study found no such relationship. A possible explanation involves the multicausal nature of resilience, which is influenced not only by social aspects, but contextual and cultural variables as well (Liebenberg et al., 2013; Masten & Barnes, 2018; Ungar, 2011). The aforementioned research involved homogeneous samples where participants shared an ethnic background (Arslan, 2019; Caqueo-Urízar et al., 2021) characterized by emphasizing community integration, reciprocity, and a collectivist cultural vision (Marcus et al., 2019; Valdivia, 2006), or involved specific groups, in which there is evidence of a significant relationship between resilience levels and social integration and competence, such as exposure to child labor (Libório & Ungar, 2014). In contrast, the sample used in this study included participants from different ethnic, social, economic, and cultural realities, generating multiple variables of their context, in addition to social integration and competence, which have a significant effect on their resilience levels.

On the other hand, our results indicate a direct and positive association between social competence and self-esteem levels, which suggests an indirect effect on life satisfaction. Emphasis has been placed on variables related to social interaction and its direct effect on self-esteem levels (Yildiz & Karadas, 2017), highlighting the relevance of the social dimension in self-concept. Thus, those who have higher levels of self-esteem tend to receive positive feedback from others through the construction of meaningful social bonds, contributing to the perception of higher subjective well-being (Saricam et al., 2016; Shek & Leung, 2017; Wagner & Gerstorf, 2018).

Among demographic variables, it was observed that age did not have a direct effect on life satisfaction levels, which differs from the findings of various national and international studies (Goldbeck et al., 2007; Haranin et al., 2007; Soares et al., 2019; Varela et al., 2019; Varela et al., 2021). However, the age of the subjects did have an effect on their levels of self-esteem and resilience, which is consistent with the data provided by INJUV (2019), suggesting that the influence on the levels of life satisfaction would be indirect in the sample studied.

In terms of gender, differences were found between boys and girls. Being male was associated with higher levels of life satisfaction and self-esteem, whereas females tended to be more resilient. These results are consistent with existing literature and could be explained by the fact that females are more exposed to adverse contextual situations, as well as more likely to suffer psychological harm that can translate into lower life satisfaction, low self-esteem, and higher levels of resilience (INJUV, 2019, Varela et al., 2021).

The results of this study indicate that resilience has a mediating role in the association between self-esteem and life satisfaction; that is, self-esteem predicts subjects’ resilience levels, while resilience predicts life satisfaction. This is consistent with studies conducted in adolescent (Wang & Kong, 2020) and university populations (Liu et al., 2014; Mak et al., 2011), where it has been postulated that those who have a positive view of themselves are more resilient, which, in turn, has a positive impact on life satisfaction.

The implications of this study are related to the importance of resilience and self-esteem as key variables in perceived life satisfaction in the adolescent population. In the case of self-esteem, it has been pointed out that higher levels mitigate the negative effects of social problems (Leary & Baumeister, 2000). It has also been postulated that those with greater resilience use adaptive coping strategies when facing adverse situations, generating a sense of self-efficacy and competence that contribute to psychological well-being (Jaureguizar et al., 2018).

Enhancing self-esteem, as well as resilience, has proven to be effective in interventions in the school context (Hodder et al., 2011) and including teachers and parents (Sharma & Agarwala, 2015), showing a statistically significant increase in the levels of both constructs and thus contributing to the psychological well-being of students. In the case of the resilience intervention, it has been shown to be effective as a protective factor, especially in reducing adolescent consumption of tobacco, alcohol, and marijuana (Hodder et al., 2011).

This study has certain limitations. First, the cross-sectional design of the study does not allow us to identify and analyze the trajectories of the relationships found, nor to establish an explanatory or cause-effect model. At the same time, while self-esteem and resilience are important variables for life satisfaction levels, they are influenced by a significant number of variables that may have equal or greater importance than those mentioned in this article. Finally, the data were obtained exclusively through student self-reports; other sources of information, such as parents or teachers, are required in future studies. Future research should also consider the weight of other variables related to the different contexts in which students develop (family, school, and peers), as well as studying the relationships found longitudinally, to gain a more complete understanding of the influence of these variables and their variations over time.

5 Conclusions

The results of this study indicate that social competence enhances life satisfaction via self-esteem, and that resilience plays a key role in the perception of life satisfaction, suggesting that interventions on social competence contribute to enhancing self-esteem and life satisfaction. In the same way, results suggest that levels of self-esteem and resilience are important sources of information for the design of intervention programs that seek to foster a positive self-concept and acquire appropriate coping strategies in adverse situations that allow students to function adequately in their daily contexts.