Maternal self-efficacy and emotional well-being in Chilean adolescent mothers: the relationship with their children’s social-emotional development

Introduction

Research into the influence of parents’ emotional well-being and self-efficacy on children’s social-emotional development has recently been subject to growing attention (Davis et al., 2020; Lawrence et al., 2020; Maria et al., 2020). However, the conditions of emotional distress experienced by parents vary depending on the circumstances, one of which is early motherhood. Infants born to adolescent mothers could be at higher risk of physical, cognitive, and social-emotional problems, given that these mothers are more prone to complications during pregnancy and emotional problems. Risk factors identified in adolescent mothers include a lower income level, a higher likelihood of school dropout and lower maternal self-efficacy (Uzun et al., 2013).

In Chile, the probability of becoming pregnant during adolescence has traditionally been higher among women with low socioeconomic status; therefore, many young pregnant females from vulnerable sectors discontinue their studies (Rodríguez-Vignoli, 2005). In recent decades, the school dropout rate among low-income youth has been higher among mothers (Olavarría & Molina, 2012). It should be noted that the number of adolescent mothers has decreased over the last decade (Instituto Nacional de Estadística, 2017). According to the National Institute of Statistics of Chile, however, 7.9% of births occurred to adolescent mothers, which is still a significant percentage of the population, and it is more likely to occur in adolescents with low education levels than in females of the same age but with higher education levels (Rodríguez-Vignoli & San Juan Bernuy, 2020). Likewise, the research carried out by Berthelon & Kruger (2017) showed that adolescent motherhood in a Chilean sample had adverse effects on educational outcomes (i.e., high school completion and enrolment in technical or university studies). Consequently, labor insertion is more difficult for these women than for those who did not become mothers during adolescence. Thus, this situation perpetuates the poverty of adolescent mothers and their children.

Although Chile has a lower rate of adolescent mothers than other Latin American countries, the characteristics of adolescents who are more likely to become pregnant at an early age are similar (for instance, low economic and educational levels) (CEPAL, 2018). However, unlike other Western countries such as Mexico and the United States, in Chile adolescent pregnancies are not related to ethnicity or religion (Huneeus et al., 2018; Jiskrova & Vazsonyi, 2019; Mejía-Benavides et al., 2019). Likewise, along with England and Ethiopia, Chile stands out in terms of the measures that the government has taken to reduce the rate of pregnancy in adolescence (Chandra-Mouli et al., 2019). Furthermore, compared to other Latin American countries, Chile is relatively prosperous, although the economic differences inside the country are greater than in other OECD countries (Banco Mundial, 2020). Therefore, adolescent mothers continue to belong to the lowest socioeconomic levels, and some are likely to have more than one child (Instituto Nacional de Estadística, 2019). A study by Luttges et al. (2021) found that Chilean adolescent mothers require support in parenting strategies, since they lack the skills needed to cope with the difficulties of motherhood (Erfina et al., 2019; Mangeli et al., 2017). Scientific evidence has shown that child-rearing conflict in adolescent mothers and economic problems, such as those experienced by Chilean adolescent mothers, are associated with parental distress and self-efficacy, which are, in turn, related to problems with children’s social-emotional development (Sanglee et al., 2019). Few studies have focused on the relationship between the variables associated with the emotional well-being and self-efficacy of adolescent mothers and the social-emotional development of infants, particularly in Chilean dyads. So, it is important to consider the children’s social-emotional development when analyzing maternal factors in adolescents, and to bear in mind the unfavorable environment in which they live, characterized by poverty, lack of support, and drop-out from school.

Materials and Methods

Results

Table 2 shows the mean, SD, minimum and maximum scores for maternal factors obtained by the EEP, HADS and PSS and the ASQ-SE raw scores for children’s social-emotional development.

With respect to the relationship between mothers’ self-efficacy scores and well-being, maternal self-efficacy was negatively related to maternal anxiety (r = −0.511, p < 0.001), depression (r = −0.504, p < 0.001) and stress (r = −0.715, p < 0.001). This result means that maternal self-efficacy was lower in adolescent mothers who demonstrated higher levels of anxiety, depression, and stress. Regarding the analysis of maternal self-efficacy, anxiety, depression and stress, and their relationship with children’s development, bivariate analyses (see Table 3) showed that the total ASQ-SE scores for children’s social-emotional development were associated with maternal stress scores (r = 0.32, p = 0.004) and maternal self-efficacy (r = −0.23, p = 0.046). Moreover, self-regulation development was related to maternal self-efficacy scores (r = −0.35, p = 0.002) and maternal stress scores (r = 0.39, p < 0.001). These results indicate that children’s social-emotional and self-regulation development were better in children whose mothers presented lower stress levels. Furthermore, self-regulation development was better in children whose mothers demonstrated higher maternal self-efficacy. With respect to the relationship between mothers’ and children’s age, and household income with children’s social-emotional development and area scores there were not a significant association between them (p > 0.05) (see Table 3).

With respect to the multiple mediation models, the only social development dimension for which there were significant results in terms of predictive capacity was children’s self- regulation. For this reason, results for these scales only are presented here, although the other mediation models are included in the Supplemental Materials (see Sections S2 to S6). Figure 1 shows the model that, firstly, specified the direct effects of maternal self-efficacy on the mediating variables: maternal anxiety (a₁), maternal depression (a₂) and maternal stress (a₃). The model also specified the effects of the mediating variables (maternal anxiety, depression, and stress) on children’s self-regulation (b₁ to b₃). All effects were calculated separately, along with the three mediated effects (terms a₁b₁, a₂b₂, and a₃b₃). In Fig. 1, the c’ path represents the direct effect of maternal self-efficacy on children’s self-regulation while controlling for the mediated effects. Note that in this mediation model, the total effect of maternal self-efficacy on children’s self-regulation (i.e., the c term) was thus estimated as the difference between estimates c’ and the total mediated or indirect effect (a₁b₁ + a₂b₂ + a₃b₃).

The direct effect between maternal self-efficacy and children’s self-regulation was nonsignificant (c′ = −0.09, p = 0.36; 95% CI [−0.29 to 0.11]). The indirect effects on children’s self-regulation mediated via either maternal anxiety (a₁b₁ = −0.01; 95% BCa–CI = [−0.11 to 0.08]), maternal depression (a₂b₂ = 0.03; 95% BCa–CI = [−0.06 to 0.12]) or maternal stress (a₁b₁ = −0.14; 95% BCa–CI = [−0.33 to 0.03]) were not significantly different from zero. However, when all these effects, both direct and indirect, were included in the model, the total effect yielded a significant result (c = −0.22; 95% CI = [−0.35 to −0.08]), which indicated multiple partial mediation. This model therefore globally yielded a significant total effect on children’s self-regulation, thus indicating that higher maternal self-efficacy results in better child self-regulation and, in turn, reduces the effect of anxiety and stress on self-regulation.

Table 4 shows all estimated arcs in the final model for maternal self-efficacy (X) on children’s self-regulation (Y), mediated by maternal anxiety, depression, and stress. This final model with multiple mediation effects accounted for 16% of the variability in children’s self-regulation (F (4, 74) = 3.64; p < 0.01).

Discussion

The current study aimed to analyze the relationship between maternal self-efficacy and emotional well-being (levels of stress, depression, and anxiety) in adolescent mothers and their association with children’s social-emotional development. Maternal well-being was considered as a mediator between maternal self-efficacy and children’s social-emotional development. Our results were similar to those of Vance et al. (2020), who found that maternal self-efficacy predicted better maternal psychological well-being and was higher in older mothers. Even though we did not include older mothers to compare maternal self-efficacy levels between different age groups, in the sample of adult Chilean mothers considered to construct the instrument used in this research (Farkas-Klein, 2008), the level of maternal self-efficacy was similar to that in our sample. This was an interesting finding because it suggests that maternal self- efficacy, even in young mothers, is closely linked to maternal stress, anxiety, and depression (Brazeau et al., 2018; Seymour et al., 2015). Thus, strengthening maternal self-efficacy and reducing mothers’ emotional distress could be key to ensuring that adolescents have a better experience of motherhood. In light of the evidence provided by the studies mentioned above, the importance of our results, as pointed out by Haslam, Pakenham & Smith (2006), lies in the fact that high maternal self-efficacy could explain the lower levels of mother’s problems of emotional well-being by promoting better self-control, self-confidence and the use of strategies to face challenges. In Chile, some studies have indicated that intervention programs for adolescents during the early stages of motherhood contribute to better maternal mental health (Aracena et al., 2009, 2011). In this regard, our findings justify the evaluation of mothers’ anxiety, depression, and stress levels, in addition to key information regarding the close relationship between maternal self- efficacy and psychological well-being. This contributes to mothers’ sense of ability to tackle behavioral problems and demanding situations involving their children (Hastings & Brown, 2002).

Moreover, our findings showed that some maternal factors were related to children’s social-emotional development. Specifically, a relationship was identified between maternal stress, maternal self-efficacy, and children’s social-emotional development. The results concerning the association between maternal stress and social-emotional development are consistent with the findings of Palmer et al. (2018), who demonstrated that children’s social-emotional development problems were related to higher levels of stress reported by healthy adolescent and adult mothers who were followed up from pregnancy until their children were 2 years old. Pesonen et al. (2008), in particular, indicated that maternal stress interferes in children’s self-regulation, probably because the stability of maternal stress is related to the child’s emotional regulation (Williford, Calkins & Keane, 2007). Likewise, as mentioned in the introduction, social-emotional development, which could include a disruptive temperament and more externalizing behavioral problems in children, is related to higher maternal stress levels (Gutteling et al., 2005). The above mentioned findings might explain the significant relationship between higher maternal stress levels and negative children’s self-regulation observed in the present study. In this regard, the sample employed in this study comprised low-income mothers, as with the sample used by Bush et al. (2017), who examined the association between maternal stress and children’s self-regulation from gestation to the postpartum period. In this latter study, mothers who perceived themselves as more stressed during pregnancy and after childbirth reported that their babies had lower levels of self-regulation.

The studies mentioned considered mothers’ perceived stress. They included questionnaires to assess children’s social-emotional development, as in our research, which measured children’s social-emotional development with the ASQ-SE through a questionnaire completed by adolescent mothers. Salomonsson & Slued (2010), who studied a sample of mothers suffering from emotional problems, observed a relationship between social-emotional development levels measured with the ASQ-SE and maternal stress levels evaluated through a questionnaire, and therefore concluded that the ASQ-SE could be used as a tool to anticipate mental health problems in mothers, rather than just child development problems. Unlike this study, our sample did not consist primarily of mothers with high maternal stress levels. Despite this, our findings contribute to the literature that explains how maternal stress could influence children’s social-emotional development in low-income young mothers, as occurs in samples of older mothers and mothers with mental health problems.

Concerning the association between maternal self-efficacy and children’s social-emotional development, our results were consistent with a study by Treat et al. (2020), who found that maternal self-efficacy was positively related to better social and emotional development in children in a sample of low-income mothers. They also suggested that mothers who receive social support have higher levels of parental self-efficacy, since this could contribute to greater self-confidence in parenting skills. This is an important issue, given that maternal self-efficacy can vary depending on the support the mother receives from family, friends and the child’s father, as well as her emotional well-being (Coleman & Karraker, 1998, 2003). However, this variable was not considered in our research, but it is relevant information that could be considered in future studies on young mothers, since an adolescent mother who receives social support may have higher levels of maternal self-efficacy. Shahry et al. (2016) demonstrated that women who had unintended pregnancies perceived that their levels of social support and self-efficacy were lower, and that low maternal self-efficacy could generate frustrating feelings related to parenting. In a sample of dyads between low-income mothers and infants, Bates et al. (2020) found an association between higher levels of maternal self-efficacy and fewer problems of self-regulation in the children. The instrument used in that study measured aspects related to the child’s ability to be calm, like the instrument used in our study. According to the authors, infant care that included parents’ feelings, behaviors, and thoughts influenced the dyadic relationship between mother and child.

Our study did not find a significant association between maternal self-efficacy and well-being and children’s adaptive functioning, or any of the other ASQ-SE areas, unlike the studies by Micklewright et al. (2012) and Stika et al. (2015). However, the results of research carried out by Voliovitch et al. (2021) were similar to ours. These authors showed that the parenting stress associated with low socioeconomic status was not related to adaptive functioning in children at risk of autism spectrum disorder. Even though the study sample comprised children at risk of autism spectrum disorder, this was not related to parental stress. The discordance between maternal stress and children’s adaptive functioning (which focuses on the bodily needs of children, such as eating and sleeping) may be due to different biological factors, such as breastfeeding, which has been shown to be associated with a reduction in infant feeding problems (Schmid et al., 2011), smoking in breastfeeding mothers, which affects infant sleep (Mennella, Yourshaw & Morgan, 2007), and parents’ knowledge of and adaptation to their children’s sleeping and feeding patterns (Rosen, 2008), variables that were not evaluated in this research. In terms of the other areas measured, social-communication, interaction and affect were not found to be related to maternal factors (maternal self-efficacy, anxiety, depression, and stress). Although we already know that maternal self-efficacy and the mother’s emotional well-being play a prominent role in children’s social-emotional development, most previous studies have related them to attachment security (Groh et al., 2017; O’Connor, Collins & Supplee, 2012). This aspect should be considered in future studies on dyads of adolescent mothers and children’s social-emotional development.

Regarding maternal emotional well-being as a mediator of maternal self-efficacy and children’s social-emotional development, our results partially differ from those of Weaver et al. (2008), who found that maternal depression mediates the relationship between parental self- efficacy and childhood behavioral problems. According to our results, children’s self-regulation was better when maternal self-efficacy was higher if the effect of maternal emotional distress decreased, since the combined total model we used for this research was significant. This means that children’s self-regulation is better when maternal self-efficacy is higher if the effect of anxiety and stress is reduced, but the inverse effect occurred with depression. The indirect effects of maternal self-efficacy, mediated by anxiety, depression, and stress, on children’s self-regulation were nonsignificant in our sample. This result could be due to the small sample size and the fact that a low number of adolescent mothers had high levels of depression. Furthermore, the low rates of maternal depression could also explain why the children in our sample did not show high scores for problems of social-emotional development—a finding that is at odds with previous reports in the literature that the children of adolescent mothers have more significant behavioral problems, lower levels of social performance and poorer well-being (Hofferth & Reid, 2002; Levine, Pollack & Comfort, 2001). Despite this, our results are similar to the findings of Van den Heuvel et al. (2015), who demonstrated that the presence of maternal anxiety could be a mediating factor between maternal mindfulness, which is associated with optimism and feelings of competence, and self-regulation problems in infants. In light of the previous studies presented here, the importance of the self-confidence of mothers as caregivers of their children has relevant implications for their emotional well-being and their children’s development. However, this aspect should be considered in future studies on adolescent mothers with higher levels of emotional distress. From this standpoint, this study may contribute to the development of theoretical models to explain the effect of specific maternal factors, which could then be tested with larger samples and longitudinal research. However, the mediation models using other ASQ-SE areas (adaptive functioning, affect, social communication, and interaction) and the total score for social-emotional development presented a low predictive capacity that could not be considered useful. These results could be explained by the fact that this study showed no significant correlation between maternal factors (i.e., maternal self-efficacy, depression, anxiety, and stress) and the rest of the areas of social-emotional development. Furthermore, most of the studies which have found an association between maternal self-efficacy and emotional well-being and children’s social-emotional development have been carried out with adult mothers rather than adolescents (Treat et al., 2020). Likewise, studies of mothers’ emotional well-being problems focused on the perinatal (Junge et al., 2017) and postnatal (Letourneau et al., 2012) periods, but did not consider whether these problems originated at these stages, and in any case in our sample the children were older. Another aspect to consider is that social support in motherhood plays a key role in maternal self-efficacy and emotional well-being, as pointed out by Haslam, Pakenham & Smith (2006), who measured how mothers perceived social support. Although in our study we asked adolescent mothers if they received support in childcare, we did not inquire about the quality of the support or how they perceived it. Their perceptions of these issues may well have influenced the maternal factors studied and may have interfered in their children’s social-emotional development.

Despite this, the results obtained confirm the importance of the emotional well-being of adolescent mothers in terms of its influence on the social-emotional development of their children, as mentioned in the introduction. Previous studies on Chilean adolescent mothers have shown that those with better emotional stability have children with a lower probability of developmental delay (Aracena et al., 2011; Olhaberry et al., 2015). However, few studies have analyzed this issue in depth in Chile. Therefore, considering the importance of the results of our study, it is necessary to explore how emotional state and self-perception influence the role played by adolescent mothers in the development of their children. Thus, our results support the need to include adolescent mothers in research related to emotional well-being (Leerlooijer et al., 2014; Marino et al., 2016; Meadows-Oliver et al., 2007), maternal self-efficacy (Ford et al., 2005) and children’s development in light of their environment (Keown, Woodward & Field, 2001). In the Chilean context in particular, adolescent mothers could be considered an at-risk population, since there is a significant association between low socioeconomic level and adolescent pregnancy.

This population has been subject to growing inequality in recent years, with a higher pregnancy rate in low-income adolescents; moreover, even though the adolescent pregnancy rate has decreased in Chile, the intergenerational transmission of poverty in adolescent mothers continues to be a problem (Lavanderos et al., 2019). It should also be noted that Farkas & Valdés (2010) carried out research into Chilean low-income mothers and found that older mothers presented lower levels of stress related to the child’s characteristics and that, in turn, increased maternal stress leads to a reduction in perceived self-efficacy regarding motherhood. Our findings suggest, as proposed by Boyce et al. (2017), that strengthening parental self-efficacy would help improve emotional well-being levels and children’s social-emotional development. From this standpoint, studies that take this approach are essential to draw conclusions on how the emotional well-being and maternal self-efficacy of Chilean adolescent mothers and their families are associated with their children’s development and behavior.

Other studies in samples of Latin American women have referred to the issue of economic and educational inequality associated with early motherhood (Marteleto & Villanueva, 2018; Rodríguez-Vignoli & Cavenaghi, 2014) and have assessed the impact of the context on their children’s social-emotional development (McDermott et al., 2021). Although one of the more prosperous countries in Latin America (Banco Mundial, 2020), Chile does not differ greatly from other countries in the continent in terms of adolescent mothers’ characteristics, and so the results presented here can perhaps be considered as a reference point for research of this nature in other countries.

The present study contributes to the literature on the importance of emotional well-being and maternal self-efficacy on the social-emotional development of children of adolescent mothers. However, the study presented several limitations that should be considered when interpreting the results. First, the study sample size was moderate, which may have limited the statistical power when testing conjecture effects. Nevertheless, as reported in a literature survey by Fritz & MacKinnon (2007), this is a common problem in psychology; 40% of the studies reported in their survey used moderate to low sample sizes (below 150 observations). In order to mitigate the problem of the sample size, the current study is based on the use of a bias-corrected bootstrap confidence interval, as this is considered a trustworthy test (Hayes & Scharkow, 2013). More research is needed to confirm our results. Additionally, the correlational design did not permit us to establish a clear causality, so it would be necessary to treat the “effect” concept used in the mediation analysis with caution. The “effect” reported here was estimated based solely on the covariation between children’s self-regulation, maternal self-efficacy and maternal emotional well-being scores. In this regard, covariation is a condition that does not necessarily imply causality. The present study used a cross-sectional design, which clearly limited our ability to draw causal inferences regarding the relationships between the variables. However, the proposed models were supported by previous research and by current theory, and all the relationships estimated between the mothers’ self-efficacy and well-being, and children’s social development have also been demonstrated by longitudinal studies (e.g., Huang et al., 2014; Lawrence et al., 2020; Van den Heuvel et al., 2015; Vance et al., 2020). Other studies of these relationships have also applied cross-sectional designs (e.g., Azmoude, Jafarnejade & Mazlom, 2015; Puspasari, Nur Rachmawati & Budiati, 2018; Tan et al., 2020). In any case, the findings of the present study would need to be replicated by a longitudinal design to establish temporal order between the exogen and endogen variables, so that the mediation effects can be adequately tested.

Another limitation to consider is that all tools used to analyze the data were questionnaires. Therefore, the adolescent mothers’ self-assessment of maternal self-efficacy and emotional well-being and their completion of questionnaires on their children’s social-emotional development may have generated biases that could have interfered with our results. Furthermore, as in the study by Krijnen, Verhoeven & van Baar (2021), child development measured with ASQ-SE was reported by the mother. It is well known that mothers’ emotional well-being problems can affect their perceptions of their children’s development or behavior (Closa-Monasterolo et al., 2017); therefore, descriptions of their children’s development may be based on their emotional state. Likewise, Chilean low-income adolescents have lower linguistic skills than those who from higher economic sectors (Agencia de Calidad de la Educación, 2018), and so there is a risk that the questions were not fully understood. From this perspective, the significant associations may be due to a functional relationship, but it is also possible that they are due to individual and subjective tendencies of the mother answering the questionnaires.

Despite the limitations of the design, according to Rothbart & Bates (2007), questionnaires completed by parents to measure child development can be advantageous because parents have observed their children in different situations for longer than external observers. Their responses may therefore be more accurate. For future research, we suggest that questionnaires on children’s social-emotional development questionnaires and behavioral observation measures be complemented.

Another aspect to consider is that the study did not take account of the parental self- efficacy levels and emotional well-being of fathers involved in childcare. Therefore, the results may have varied depending on family structure, especially where the father was present (Fagan et al., 2014; Cabrera et al., 2008). However, although the father’s presence is important, not all adolescent mothers are supported by the father of their children. In the Chilean context, several adolescent fathers indicated that they did not live with the mother of their child, nor with their child, and that the relationship was unstable with a low level of commitment, which could lead to the breakdown of the couple’s relationship and risk their closeness or even their contact and affectional bond with their children (Molina-Gutiérrez, 2011). Future studies on Chilean adolescent mothers could consider a sample in which the fathers fulfill their parenting responsibilities. The intention would be to analyze fathers’ parental self-efficacy and emotional well-being, and the adolescent mothers’ degree of satisfaction with the fathers’ involvement in the care of their children, as in a study carried out by Fagan & Lee (2010) with a sample that included, among other ethnicities, Latina adolescent mothers. Further analyses of adolescent mothers and their children are needed to determine the consistency of our findings.

Conclusions

Overall, the present study confirmed that maternal self-efficacy could explain lower levels of emotional well-being problems experienced by adolescent mothers. In turn, when the emotional well-being of mothers mediates maternal self-efficacy, this could be a determining factor for their children’s social-emotional development. However, future studies should use larger samples and perform further analyses, (for instance, of sociodemographic variables and their possible relationship with children’s social-emotional development). The above findings are necessary to explore how the emotional and self-perceived state influences the role of mothers in the development of their children, especially in adolescent mothers. In this context, maternal factors should be considered when planning interventions to prevent childhood behavioral problems. These dyads, therefore, form part of a group that should receive significant attention in Chilean national programs. These findings highlight the importance of providing adolescent mothers with social support and guidance during pregnancy onwards and intervention programs tailored to their emotional needs. A growing understanding of the maternal factors analyzed should help evaluate the research results associated with this topic and the effectiveness of intervention efforts to reduce negative family implications that could generate problems in the social-emotional development of children born to Chilean adolescent mothers.

Supplemental Information