Positive Parenting and Sociodemographic Factors Related to the Development of Chilean Children Born to Adolescent Mothers

1. Introduction

The relationship between positive parenting and child development has received significant attention in recent years [1,2,3,4]. Parenting is a wide-ranging concept involving different aspects of childrearing, including support for children's needs, parental styles, provision of food and care, teaching and developmental guidance, and affectionate communication [5] Following Roggman et al. [6], in this paper, 'parenting' refers to the characteristics of high-quality parent–child dyadic interactions that occur face-to-face and promote child development. The literature suggests that a home environment characterized by positive interactions between parents and children significantly impacts typically developing children [7,8,9].

Parenting and child development have also been studied in high-risk families, specifically in families from socially and economically vulnerable contexts [10,11,12]. Existing research has demonstrated that low economic income negatively interferes in the family context due to less stimulating home learning environments for their children [13], directly affecting the maternal role of younger mothers [14]. In fact, according to the World Health Organization [15], adolescent motherhood is more prevalent among women with lower incomes. Latin America has the second-highest adolescent birth rate globally, with a higher prevalence in lower-income countries [16].

A recent study has found that Latin adolescent mothers' characteristics related to their economic and educational level, knowledge of child development, and co-parenting conflict are associated with their children's development, with cognitive stimulation and the emotional support given to their children being significant factors [17]. Given the socioeconomic challenges faced by adolescent mothers in Latin America, [18] emphasized the importance of incorporating an examination of co-parenting relationships, particularly involving the fathers of children born to adolescent mothers, and the impact of dropping out of school on adolescent parenting experiences.

In Chile, although there has been a decrease in adolescent pregnancies over the last decade in comparison to other countries in Latin America [19], pregnancies still occur among adolescents from economically and socially vulnerable backgrounds [20]. The Biobío region, known for having some of the highest poverty rates in Chile [21], also reports a higher number of adolescent mothers who have been exposed to circumstances requiring judicial determinations for the violation of their rights compared to the national average [22].

Despite the relevance of investigating adolescent motherhood, most of the studies that have analyzed the relationship between parenthood and child development have been carried out with adult mothers [23,24,25,26] (. Meanwhile, few studies have explored the relationship between parenting in adolescent mothers, family sociodemographic factors, and child development. Thus, positive interactions between mother and child can be a real challenge for adolescent mothers, with their social and family context being particularly relevant.

1.1. Chilean adolescent mothers' sociodemographic context and children's developmental outcomes

A family context characterized by high social and economic risk significantly influences child development since it has been shown that mothers with higher income and educational levels reduce delays in children's cognitive and linguistic development [27]. In contrast, children living in poverty are at a higher risk of encountering significant developmental delays, which could be primarily due to the adverse effects of reduced cognitive skill stimulation and harmful parenting practices on children's brain development [28].

Previous research suggests that children born to adolescent mothers tend to exhibit slower cognitive, linguistic, and social development compared to those born to older mothers, possibly due to the highly vulnerable context in which they are raised [29,30,31,32]. Specifically, these children often present lower levels of language development, which may be linked to reduced verbal stimulation from their mothers [33]. Moreover, the cognitive development of such children can be negatively impacted by low socioeconomic status, as poverty can interfere with adequate childrearing and stimulation [34].

In Chile, a correlation is observed between higher economic income, educational level, and older maternal age with higher maternal competencies [35]. It is well known that adolescent mothers are more likely to drop out of school than women who are not mothers or delay motherhood [36]. In this context, the impact of adolescent motherhood in Chile translates to the interruption of their studies, a reduced likelihood of pursuing higher education, and a higher probability of future unemployment, ultimately perpetuating the cycle of poverty [37].

The consequence of being unemployed for a mother who became pregnant during adolescence is an important issue. In a study by Reynolds et al. [38], Chilean children whose mothers were more involved in the workforce during their early years demonstrated enhanced cognitive, linguistic, and socio-emotional assessment performance. This association may be attributed to mothers' employment, which is associated with a more favorable learning environment as the family may be better able to invest in stimulating learning resources and participate in educational activities that promote their children's development [39].

The employment and educational status and the high-risk and vulnerable context in which adolescent mothers are situated highlight the need for parenting support to optimize their children's development [40] . In a study conducted by Luttges et al. [41], it was found that Chilean adolescent mothers require orientation with parenting strategies, as various authors have noted that mothers who have children at a young age often lack the necessary maternal skills to face adversity [42,43]. In fact, within the context of Chilean adolescent mothers, it has been demonstrated that support in the care of her baby from the extended family of the adolescent contributes to the mother's greater satisfaction with her child [44].

Indeed, child development may be less affected if grandparents provide greater protection against the adverse effects of fatherlessness [45]. Previous literature has found that support in caring for children born to adolescent mothers prompts them to be more affectionate with their babies [46]. Therefore, it is necessary to delve into the impact on child development when adolescent mothers receive support in caring for their babies and when these mothers are affectionate during their interactions with their children.

While social support for adolescent mothers is essential, research suggests that the absence of fathers can emerge as one of the primary challenges young mothers face and can significantly impact their children's development [47]. In the Chilean context, it has been observed that adolescent fathers who do not cohabit with the mother of their children may contribute to an unstable relationship characterized by a low level of commitment, potentially resulting in relationship breakdowns and subsequent loss of contact and bonding between fathers and their children [48].

The role of fathers in the early stages of child development has significant effects on the child's neurobiological development, emotional and behavioral regulation systems, and experiences in children's mental models of themselves and others [49] (Fitzgerald & Bocknek, 2013). Specifically, research indicates that father–child contact among children of adolescent mothers is associated with improved academic performance, enhanced reading skills, and fewer behavioral problems in later years [47].

In addition, Chile offers preschool centers designed for adolescent mothers' children. They are free of charge and intend to increase mothers' support in the care of their children [50]. Their primary goal is to assist young mothers and fathers in ensuring their children receive the necessary care and stimulation, all while enabling adolescents to continue their education, because early motherhood often leads to high school dropout rates among young women. In Chile, concerted efforts have been made to enhance the quality of early childhood education to provide children from impoverished environments with the same opportunities as those from higher-income families [51]. Therefore, analyzing the effects on child development of receiving support in caring for children born to adolescent mothers at home when they do not attend preschool centers is relevant.

Although it has been shown that parenting depends on various sociodemographic variables, such as parent's age, employment, and educational level [52] (), it is crucial to investigate how these factors influence the development of children of adolescent mothers.

1.2. Positive parenting in adolescent mothers and children's developmental outcomes

Studies increasingly recognize the value of parenting quality on child developmental outcomes [53,54,55]. However, the quality of parenting in adolescent mothers has been characterized by having a level of negative involvement, which leads them to have less positive interaction and a more withdrawn behavior with their children [56] since the context in which families are situated can contribute to or affect the emotional state of mothers when they interact with their children [57].

Hence, it is crucial to underscore the significance of positive parenting in adolescent mothers and its impact on their children's developmental outcomes within a natural context. Various studies have consistently shown that positive parent–child interactions significantly affect children's language skills, cognitive abilities, and socioemotional development [7,58,59]. More specifically, positive interactions between parents and children have been associated with the expansion of vocabulary and early literacy [60,61], as they play a vital role in fostering expressive language development [62]. Additionally, positive parenting has been linked to proficient executive functioning, academic achievement, memory, and problem-solving skills [63,64]. Furthermore, positive parenting practices in Latino mothers have predicted higher levels of child social self-efficacy [65]. Consequently, parenting practices cover various behaviors in domains such as affection, warmth, responsiveness, encouragement, and cognitive stimulation or teaching, all of which are closely associated with children's developmental outcomes [6,66,67].

On the one hand, emotional warmth, which refers to the expression of positive evaluation of the child [6], is associated with fewer problem behaviors in infants [68] and higher social skills and communication abilities in early childhood [69] and predicts lower levels of child externalizing problems [70] and higher levels of executive function in preschool and post-kindergarten [71], which is associated with cognitive development.

On the other hand, responsiveness is the maternal skill to detect a child's attention, interest, or emotional state, responding appropriately to their needs. This ability promotes higher pre-academic abilities, such as in reading comprehension, applied maths problems, and receptive vocabulary [72] and predicts higher cognitive development in infants [73] related to more significant memory and, in turn, with better social development in terms of self-regulation [74].

Indeed, scientific evidence has consistently shown that socioeconomic risk among adolescent mothers can negatively impact their ability to exhibit emotional warmth and responsiveness, adversely affecting their children's cognitive development [34]. Furthermore, affection and responsiveness in adolescent mothers are also associated with infant security [75].

Encouragement, understood as promoting autonomy in children, is linked to higher independence, security, and language, cognitive, and socioemotional development [76]. However, children show higher separation anxiety levels when maternal behavior is characterized by extreme intrusiveness or is very protective [77]. Children born of adolescent mothers may present lower levels of expressive and receptive language than children of older mothers, attributed to the high intrusiveness, rather than encouragement, and low verbal stimulation provided by young mothers [33].

In this sense, cognitive stimulation or teaching related to promoting the child's participation in parents–child joint activities and conversation is associated with better cognitive function skills [78], receptive and expressive communication [54], and prosocial behavior [79]. Thus, the provision of low verbal stimulation by adolescent mothers can negatively affect their child's development.

It should be noted that the quality of interactions with children may vary depending on the child's gender. For example, a systematic review found a significant difference between sons and daughters concerning autonomy-supportive strategies, even though these differences by gender have been found to be minimal in recent years [80]. Nevertheless, a more recent systematic review showed differences in parenting and the child's gender concerning vocalization, socializing strategies, play, and the provision of different toys to their sons and daughters, which were associated with some differences in child development across the children's gender [81]).

Moreover, it has been demonstrated that mothers are more affectionate with girls [82] and when their distress levels are lower [83]. Lower maternal distress is associated with the implications of the biological father’s coresidence and when they are more involved as caregivers [84]. In this sense, it is essential to understand how the presence of the father and mother's affection can influence children's development and the possible gender differences.

Given the difficulties and parental characteristics experienced by Chilean adolescent mothers, it is imperative to delve into this issue. Such investigation can provide valuable insights that enable the implementation of intervention practices to enhance the quality of interaction between adolescent mothers and their children [85,86].

Based on the evidence reviewed above, the current study is exploratory, aiming to: a) explore the relation between family-related demographic variables and parental factors (e.g., mother who has dropped out of school, employment status, support in the care of a child, child's gender, fathers/grandparents living with their child, and children attending preschool), and language, cognitive, and personal-social development in typically developing young children of Chilean adolescent mothers; and b) to explore the relation between parenting, defined in terms of affection, responsiveness, encouragement, and teaching and cognitive, linguistic and personal-social development in typically developing children of Chilean adolescent mothers at early ages.

2. Materials and Methods

2.1. Participants

The current study was conducted in seven health centers, one hospital, four preschool learning centers, and one residential home for adolescent mothers (belonging to the National Children’s Service) in Biobío, Chile. Power analyses were run in advance to determine the required sample size for the present study. In this regard, analyses based on Fisher’s transformation for the correlation test and assuming moderate-sized correlations (r=.3, α=.05, and 1-β=.80) yielded a required sample size of 84 individuals. Additionally, we ran power analyses for linear regressions by assuming models with at least four predictors and moderate effects (k=4, f2=.15 -equivalent to R2=.18-, α=.05, and 1-β=0.80), for which the minimum required sample size was 80 individuals.

The final sample included 79 dyads of mothers and children who met the eligibility criteria (low-income mothers who became pregnant at 19 years old or younger, average household income less than or equal to 678.49 USD per month for mid-low income and 391.16 USD per month for low-income, according to the criteria of the Chilean Association of Market Researchers 2019 [87] and children of typical development. Within this group of mothers, 62% had completed high school, 29% had completed primary school, and fewer than half had not completed primary school (9%, n=7); 14% had dropped out of school due to maternity.

Adolescent mothers aged between 15 and 21 years (M = 19.1, SD = 1.7) and children aged between 10 and 24 months (M = 15.5, SD = 4.2) were visited at their homes. They provided voluntary, informed consent for their participation and use of data related to their family's sociodemographic, parenting, and child developmental characteristics when aged older than 18 years old; in the case of mothers younger than 18 years old, their legal guardians signed an informed consent. The characteristics of the children and their mothers that were relevant to the study are presented in Table 1.

2.2. Procedure

Data procedures were collected via questionnaires and observations. Firstly, the coordinators of the health centers, hospitals, preschool learning centers, and residential homes for adolescent mothers provided a database with telephone numbers and addresses to contact the families directly. Families who agreed to participate were visited at home and informed about the study. A sociodemographic questionnaire was read and explained to adolescent mothers, clarifying the possible answers. Subsequently, mothers were asked to engage in a video-recorded 10-minute play session with their infants at home, with the following instruction translated into Spanish: "Interact and play with your children as you typically do." Mother–child dyads were observed during 10 minutes of free play using toys provided in three bags that included books, toys for pretend play, and age-appropriate manipulative toys, respectively. The "Three Bag task" provides a structured space with flexibility for the mother to guide the interaction and to use the toys as they wish while playing with their children [88]. PICCOLO [6]was then used to score parent–child interactions in these 10-minute video recordings. Finally, the Ages and Stages questionnaire-3 [89]was administered to check whether the child performed the behavior indicated in the items of the areas of communication, problem-solving, and personal-social with the following instruction: "We will read each question carefully and check whether your baby does the activity, does the activity sometimes or not yet." Interviewers and mothers completed the ASQ-3 in around 10–15 minutes. According to Small et al. [90], in low-income countries in Latin America, this instrument is usually read by a trained evaluator who helps parents record and interpret their child's response when they are evaluated. Mothers completed ASQ-3, but all items were read and verified whether the child did the activity indicated in the questionnaire by a member of the research team due to difficulties in reading presented by the adolescent mothers.

2.3. Measures

An ad-hoc sociodemographic questionnaire was used to record the mother's age, educational level, employment status, and whether she was receiving support or help in the care of her child from the child's grandparents when the mother was studying or working at home and in parenting tasks such as feeding, toileting, and putting the baby down to sleep. The same questionnaire was used to record the child's age, gender, relatives who resided with them, and whether they attended a preschool learning center.

The Parenting Interactions with Children: Checklist of Observations Linked to Outcomes (PICCOLO, [6]) is an observational tool assessing interactions between parents and children aged 10 to 47 months. It includes 29 observable parental behaviors that reflect parent interaction behaviors, which are scored according to their frequency as 0 (absent, no behavior observed), 1 (barely, minor, or emerging behavior), and 2 (clear, definitive, strong, and frequent behavior). These are grouped into four domains: (a) Affection (7 items), which involves physical and verbal expression of affection, positive emotions, positive evaluation, and positive regard; (b) Responsiveness (7 items), which includes reacting sensitively to a child's cues and expression of needs or interests and reacting positively to the child's behavior; (c) Encouragement (7 items), which considers parents' support of children's efforts, exploration, independence, play, choices, creativity, and initiative; and (d) Teaching (8 items), which includes cognitive stimulation, explanations, conversation, joint attention, and shared play. The instrument generates a score for each domain between 0 to 14 (0 to 16 for the teaching domain) and a total score between 0 and 58 (adding all the items). The original PICCOLO reliability is good; the analysis of Cronbach Alpha for the total instrument was .91 (.78 for affection, .75 for responsiveness, .77 for encouragement, and .80 for teaching), and the instrument had good results for construct and predictive validity [6]. In this study, the Chilean adaptation of PICCOLO [91] was applied, and the reliability (Cronbach Alpha) for the total instrument was 88 (.76 for affection, .85 for responsiveness, .75 for encouragement, and .65 for teaching). For this sample comprised of Chilean adolescent mothers (n = 79), the reliability (Cronbach α) for the total instrument was .87 (.64 for affection, .70 for responsiveness, .69 for encouragement, and .62 for teaching). Inter-rater reliability for the current study was calculated from 20% of observations, yielding an inter-rater agreement estimate of 0.85 for total scores used in the current analyses. The training of the raters involved the following steps: the second author, whom the authors of the original PICCOLO trained, trained the first author for this study, who read about the content and purpose of the measure (during an approximately 3-h session) during a university course attended by the first author that the second author dictated. In this course, the first author scored the Spanish version of the PICCOLO when she watched four video recordings (3 h). The first author was considered to have satisfactorily completed her training when the percentage of agreement between evaluators was equal to or higher than 80%. Both authors viewed and discussed 20 of the 79 video recordings for this study and coded them to establish reliability.

The Spanish version of the Ages and Stages Questionnaire Third Edition (ASQ-3), developed by Squires et al. [89], is a caregiver-report tool to assess children's developmental progress from birth to 60 months. This questionnaire includes six items that pertain to the areas of communication (both comprehensive and expressive language), problem-solving (which includes cognitive skills in terms of learning and play), and personal-social development (related to the child's self-help abilities and social interactions). For each item, respondents select one of three possible responses coded as "10" (yes), "5" (sometimes), or "0" (not yet), based on whether the child could perform a specific task. The overall score is derived by summing these responses, with higher scores indicating higher levels of development in children. To ensure consistency and avoid variations in cutoff scores based on age ranges defined in ASQ-3, the developmental scores of children in this sample were transformed into Z scores using a standard metric. The ASQ is recognized as a valid and reliable instrument for assessing child development, with a reported Cronbach's alpha coefficient of .94 for overall reliability [92]. It should be noted that ASQ-3 was validated in Chile by Armijo et al. [93]and considered an effective screening tool. In the present sample, reliability was confirmed with a Cronbach's alpha coefficient of .73.

2.4. Data analysis

Data were analyzed in several stages. Firstly the bivariate association between parenting behaviors, socio-demographic factors (i.e., whether the mother had dropped out of school, her employment status and whether she received support or help in the care of her child, child’s gender, whether fathers resided with their child, and whether children attended preschool), and the ASQ-3 scores was quantified and tested by means of correlation tests in the case of interval scales and using parametric tests based on comparisons of means in the case of categorical predictors. In order to complement the statistical analysis, effect sizes were computed using Cohen’s d, eta-squared, and product-moment correlations. In this regard, these analyses were useful to determine initial sets of predictors to be included in the predictive models for all ASQ-3 scores. More specifically, a significance (p) lower than .05 was used to specify sets of predictors to be used for all responses employed in the current study.

Linear regression models were estimated in order to predict ASQ-3 scores using as exogenous variables all relevant variables concerning parenting behaviors as well as socio-demographic factors employed as potential predictors in the modeling procedure. In this regard, the routine was based upon a feasible solutions algorithm [94], an intensive computing method that allows researchers to find an optimal solution amongst multiple possible solutions (i.e., candidate models). These multiple solutions correspond to different subsets of predictors including second-order interaction terms. Specifically, the intensive routine was as follows: 1) In the first step of every iteration evaluated, models were specified to include those predictors that proved to be useful in the bivariate analyses along with a random subset of the remaining predictors with which second-order interactions will be assessed; 2) in the next step, the complexity of the model was trimmed taking into account its predictive capacity; 3) steps 1 and 2 were iterated k times, saving the model obtained in each iteration; 4) the model that yielded the highest number of occasions, and thus having the highest predictive capacity, was kept as the fittest one; 5) in the final step of the modeling routine, some of the terms kept in the fittest model found in step 3 were removed, depending on the goodness-of-fit index. One thousand different initial random subsets of predictors were studied for each response (k=1,000), allowing us to evaluate the models’ space when searching for an optimal solution while avoiding local solutions. Adjusted R-squared was employed as a fit index during the iterations and the complexity of the final solution obtained by means of the intensive procedure was reduced whenever possible by using Akaike’s Information Criterion (AIC).

All statistical models were assessed in order to improve their specification by adding non-linear terms (i.e., polynomials) as well as detecting possible issues concerning multicollinearity (by inspecting Variance Inflation Factors, VIFs, and considering VIF < 5 as an indicator of the absence of multicolinearity problems). Finally, raw and standardized effects were estimated for interpretation using the original metrics of the scales used and as adimensional effects, thus allowing comparisons betweeen predictors in terms of their relative importance.

All statistical analyses were carried out using the R environment (version 4.1; R Core Team, 2021). The intensive modeling routine was implemented using the rFSA package [95].

3. Results

3.1. Sociodemographic factors and children's developmental outcomes

The relationship between each factor included in the sociodemographic questionnaire (parenting domains regarding affection, responsiveness, encouragement, and teaching) and children's developmental outcomes (ASQ-3 Z scores) was analyzed. The results showed that mothers were more responsive when their children were girls (t (77) = -2.19; p = .02; d = 0.49) and when fathers resided with their children (t (77) = 2.0; p = .02; d = 0.49). Moreover, mothers’ encouragement scores were higher when mothers did not drop out of school (t (77) = -2.05; p = .02; d = 0.67) and when fathers resided with their child (t (77) = 1.86; p = .03; d = 0.45). Finally, teaching domain scores were higher when mothers were employed (t (77) = 2.73; p = .02; d = 0.62), when their children were girls (t (77) = 1.82; p = .04; d = 0.41) and when the father resided with their child (t (77) = 1.70; p = .047; d = 0.42). Specifically, teaching domain scores were higher when fathers and grandparents resided with the children [F (2,29.5) = 3.42; p = 0.038; η 2 = 0.08].

Regarding the relationship between children's development and sociodemographic factors, the results showed a relationship between mothers dropping out of school (t (77) = -2.32; p = .023 d = 0.76), employment status (t (77) = 2.07; p = .042; d =.60) and children's problem-solving scores. In this regard, children with currently employed mothers who had not dropped out of school showed significantly higher scores in problem-solving development. Additionally, there was a significant association between the fact of fathers residing with their children and scores corresponding to communication (t (77 = 2.77; p = .01; d =0.69), problem-solving (t (77) = 2.54; p = .013; d = 0.63), and personal-social (t (77) = 2.61; p = .011; d = 0.65) development. Specifically, children residing with their fathers scored significantly higher in the abovementioned dimensions.

Likewise, girls performed significantly better than boys in communication (t (77) = -2.22; p = .029; d = 0.51), problem-solving (t (77) = -2.21; p = .03; d = 0.50), and personal-social (t (77) = -2.65; p = .01; d = 0.60) scales.

Table 2. Descriptive summary of ASQ-3 Z and PICCOLO scores by sociodemographic factors.

	Communication	Problem-solving	Personal-social	Affection	Responsiveness	Encouragement	Teaching
	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)
Mother`s level of education
Dropped out of school (n= 11)	0.03±.97 (-1.86−1.19)	-0.6±0.88 (-1.88−1.23)	0.23±1.01 (-1.51−1.66)	8.82±2.23 (4.0−12.0)	9.64± 2.88 (3.0 −13.0)	7.27±2.97 (1.0 −11.0)	8.82±3.74 (2.0 –14.0)
Not dropped out of school (n= 68)	-0.01±0.96 (-2.06−2.09)	0.1±0.94 (-2.34−1.7)	-0.04±0.95 (-2.13−2.07)	10.09±2.53 (3.0−14.0)	10.09±2.50 (2.0−14.0)	9.03± 2.59 (2.0− 14.0)	8.56± 2.92 (0−14.0)
Statistical test	t= 0.12	t= -2.32*	t= 0.84	t= -1.57	t=-0.55	t=-2.05*	t= 0.26
(Effect size)	d= 0.04	d= 0.76	d= 0.28	d=0.51	d=0.18	d=0.67	d=0.09
Mother’s employment status
Employee (n=15)	-0.01±0.79 (-1.27−2-09)	0.45±0.81 (-0.83−1.38)	-0.05±0.8 (-1.51−1.13)	10.40±2,13 (7.0−13.0)	10.87± 1,685 (7.0 −13.0)	9.73±1,944 (7.0 −14.0)	10.07 ±2,086 (5.0 –13.0)
Not an employee (n=64)	0 ±0.99 (-2.06−2-06)	-0.11±0.96 (-2.34−1.7)	0.01±0.99 (-2.13−2.07)	9.80 ±2.60 (3.0−14.0)	9.83 ±2.67 (2.0−14.0)	8.56 ± 2.81 (1.0− 14.0)	8.25± 3.12 (0−14.0)
Statistical test	t=-0.04	t = 2.07 *	t=-0.23	t= 0.84	t= 1.44	t= 1.53	t= 2.73*
(Effect size)	d= 0.01	d =0.60	d= 0.07	d=0.24	d=0.41	d=0.44	d=0.62
Support in the care of child
Receiving support in the care of child (n= 62)	0.08±0.94 (-2.06−2.06)	0.07±0.96 (-1.93−1.7)	0±0.98 (-2.13−2.07)	9,94±2.40 (4.0−14.0)	10,18± 2.28 (3.0 −14.0)	8,85± 2.64 (1.0 −14.0)	8,65±2.81 (2.0 –14.0)
Not receiving support in the care of child (n= 17)	-0.3±0.97 (-1.98−2.06)	-0.27±0.89 (-2.34−1.04)	-0.01±0.87 (-1.51−1.66)	9.82±2.96 (3.0−14.0)	9.47±3.34 (2.0−13.0)	8.53± 2.98 (2.0− 12.0)	8.41± 3.81 (0−14.0)
Statistical test	t= 1.47	t= 1.31	t=0.06	t=0.16	t=1.02	t=0.44	t=0.28
(Effect size)	d= 0.41	d =0.37	d= 0.02	d=0.04	d=0.28	d=0.12	d=0.08
Child gender
Girl (n=37)	0.25±0.83 (-1.13−2.09)	0.25±0.86 (-1.93−1.7)	0.29±1.01 (-1.66−2.07)	10,05 ±2.33 (3.0−14.0)	10,68 ± 2.22 (2.0 −14.0)	9,30 ±2.33 (4.0 −14.0)	9,24±2.55 (2.0 –14.0)
Boy (n=42)	-0.22±1.01 (-2.06−1.5)	-0.22±0.99 (-2.34−1.7)	-0.26±0.83 (-2.13−1.18)	9.79±2.68 (4.0−14.0)	9.45 ±2.68 (3.0−13.0)	8.33± 2.94 (1.0− 14.0)	8.02± 3.31 (0−14.0)
Statistical test	t = -2.22 *	t = -2.21*	t = -2.65**	t=-0.47	t=-2.19*	t=-1.60	t=-1.82*
(Effect size)	d = 0.51	d = 0.50	d = 0.60	d=0.11	d=0.49	d=0.36	d=0.41
Preschool center
Had attended preschool (n=28)	-0.1±1.07 (-2.06−2.06)	0.17±0.96 (-1.93−1.38)	-0.13±0.95 (-2.13−1.91)	10.04±2.55 (5.0−14.0)	10.07± 2.36 (5.0 −13.0)	9.0±2.66 (3.0 −14.0)	8.79±2.59 (3.0 –14.0)
Had not attended preschool (n=51)	0.05±0.89 (-1.98−2.09)	-0.09±0.95 (-2.34−1.7)	0.07±0.96 (-1.7−2.07)	9.84±2.52 (3.0−14.0)	10.0±2.65 (2.0−14.0)	8.67± 2.74 (1.0− 14.0)	8.49± 2.26 (0−14.0)
Statistical test	t=-0.65	t=1.18	t=-0.91	t=0.32	t=0.12	t=0.52	t=0.41
(Effect size)	d = 0.16	d = 0.28	d = 0.22	d=0.08	d=0.03	d=0.12	d=0.10
Father
Residing with his child (n=24)	0.43±0.93 (-1.86−2.06)	0.4±0.7 (-0.83−1.7)	0.41±0.82 (-1.12−1.59)	10.13±2.68 (3.0−14.0)	10.88± 2.54 (2.0 −14.0)	9.63±2.30 (4.0 −12.0)	9.46±3.16 (2.0 –14.0)
Not residing with his child (n=55)	-0.19±0.91 (-2.06−2.09)	-0.17±1 (-2.34−1.38)	-0.18±0.96 (-2.13−2.07)	9.82±2.46 (4.0−14.0)	9.65±2.47 (3.0−14.0)	8.42± 2.79 (1.0− 14.0)	8.22± 2.91 (0−13.0)
Statistical test	t = 2.77**	t = 2.54**	t = 2.61**	t=0.50	t=2.0*	t=1.86*	t=1.70*
(Effect size)	d =0.69	d = 0.63	d = 0.65	d=0.12	d=0.49	d=0.45	d=0.42
Grandparents
Mother, father, and children (n=6)	0.13±0.93 (-1.13−1.23)	0.40±0.55 (-0.24−1.04)	0.22±0.89 (-0.91−1.51)	10.0±3.69 (3.0−14.0)	9.50±4.14 (2.0−13.0)	9.83±2.93 (4.0−12.0)	7.83±4.31 (2.0−14.0)
Mother, grandparents, and children (n=56)	-0.15±0.93 (-2.06−2.09)	-0.16±0.98 (-2.34−1.38)	-0.11±0.99 (-2.13−2.07)	9.75±2.53 (4.0−14.0)	9.77±2.05 (3.0−13.0)	8.41±2.83 (1.00−14.0)	8.18±3.0 (0−13.0)
Mother, grandparents, father, and children (n=17)	0.43±0.95 (-1.87−2.06)	0.39±0.87 (-1.53−1.70)	0.29±0.81 (-1.12−1.59)	10.41±2.03 (7.0−13.0)	11.06±1.75 (8.0−14.0)	9.65±1.90 (5.0−12.0)	10.24±2.49 (3.0−14.0)
Statistical test	F = 2.54	F = 2.85	F = 1.31	F = 0.45	F = 1.87	F = 1.90	F = 3.42
(Effect size)	η2 = 0.06	η2 = 0.07	η2 = 0.03	η2 = 0.01	η2 = 0.05	η2 = 0.05	η2 = 0.08

Notes: *p < .05 **p < .01.

Table 2. Descriptive summary of ASQ-3 Z and PICCOLO scores by sociodemographic factors.

	Communication	Problem-solving	Personal-social	Affection	Responsiveness	Encouragement	Teaching
	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)	M±SD (Min-Max)
Mother`s level of education
Dropped out of school (n= 11)	0.03±.97 (-1.86−1.19)	-0.6±0.88 (-1.88−1.23)	0.23±1.01 (-1.51−1.66)	8.82±2.23 (4.0−12.0)	9.64± 2.88 (3.0 −13.0)	7.27±2.97 (1.0 −11.0)	8.82±3.74 (2.0 –14.0)
Not dropped out of school (n= 68)	-0.01±0.96 (-2.06−2.09)	0.1±0.94 (-2.34−1.7)	-0.04±0.95 (-2.13−2.07)	10.09±2.53 (3.0−14.0)	10.09±2.50 (2.0−14.0)	9.03± 2.59 (2.0− 14.0)	8.56± 2.92 (0−14.0)
Statistical test	t= 0.12	t= -2.32*	t= 0.84	t= -1.57	t=-0.55	t=-2.05*	t= 0.26
(Effect size)	d= 0.04	d= 0.76	d= 0.28	d=0.51	d=0.18	d=0.67	d=0.09
Mother’s employment status
Employee (n=15)	-0.01±0.79 (-1.27−2-09)	0.45±0.81 (-0.83−1.38)	-0.05±0.8 (-1.51−1.13)	10.40±2,13 (7.0−13.0)	10.87± 1,685 (7.0 −13.0)	9.73±1,944 (7.0 −14.0)	10.07 ±2,086 (5.0 –13.0)
Not an employee (n=64)	0 ±0.99 (-2.06−2-06)	-0.11±0.96 (-2.34−1.7)	0.01±0.99 (-2.13−2.07)	9.80 ±2.60 (3.0−14.0)	9.83 ±2.67 (2.0−14.0)	8.56 ± 2.81 (1.0− 14.0)	8.25± 3.12 (0−14.0)
Statistical test	t=-0.04	t = 2.07 *	t=-0.23	t= 0.84	t= 1.44	t= 1.53	t= 2.73*
(Effect size)	d= 0.01	d =0.60	d= 0.07	d=0.24	d=0.41	d=0.44	d=0.62
Support in the care of child
Receiving support in the care of child (n= 62)	0.08±0.94 (-2.06−2.06)	0.07±0.96 (-1.93−1.7)	0±0.98 (-2.13−2.07)	9,94±2.40 (4.0−14.0)	10,18± 2.28 (3.0 −14.0)	8,85± 2.64 (1.0 −14.0)	8,65±2.81 (2.0 –14.0)
Not receiving support in the care of child (n= 17)	-0.3±0.97 (-1.98−2.06)	-0.27±0.89 (-2.34−1.04)	-0.01±0.87 (-1.51−1.66)	9.82±2.96 (3.0−14.0)	9.47±3.34 (2.0−13.0)	8.53± 2.98 (2.0− 12.0)	8.41± 3.81 (0−14.0)
Statistical test	t= 1.47	t= 1.31	t=0.06	t=0.16	t=1.02	t=0.44	t=0.28
(Effect size)	d= 0.41	d =0.37	d= 0.02	d=0.04	d=0.28	d=0.12	d=0.08
Child gender
Girl (n=37)	0.25±0.83 (-1.13−2.09)	0.25±0.86 (-1.93−1.7)	0.29±1.01 (-1.66−2.07)	10,05 ±2.33 (3.0−14.0)	10,68 ± 2.22 (2.0 −14.0)	9,30 ±2.33 (4.0 −14.0)	9,24±2.55 (2.0 –14.0)
Boy (n=42)	-0.22±1.01 (-2.06−1.5)	-0.22±0.99 (-2.34−1.7)	-0.26±0.83 (-2.13−1.18)	9.79±2.68 (4.0−14.0)	9.45 ±2.68 (3.0−13.0)	8.33± 2.94 (1.0− 14.0)	8.02± 3.31 (0−14.0)
Statistical test	t = -2.22 *	t = -2.21*	t = -2.65**	t=-0.47	t=-2.19*	t=-1.60	t=-1.82*
(Effect size)	d = 0.51	d = 0.50	d = 0.60	d=0.11	d=0.49	d=0.36	d=0.41
Preschool center
Had attended preschool (n=28)	-0.1±1.07 (-2.06−2.06)	0.17±0.96 (-1.93−1.38)	-0.13±0.95 (-2.13−1.91)	10.04±2.55 (5.0−14.0)	10.07± 2.36 (5.0 −13.0)	9.0±2.66 (3.0 −14.0)	8.79±2.59 (3.0 –14.0)
Had not attended preschool (n=51)	0.05±0.89 (-1.98−2.09)	-0.09±0.95 (-2.34−1.7)	0.07±0.96 (-1.7−2.07)	9.84±2.52 (3.0−14.0)	10.0±2.65 (2.0−14.0)	8.67± 2.74 (1.0− 14.0)	8.49± 2.26 (0−14.0)
Statistical test	t=-0.65	t=1.18	t=-0.91	t=0.32	t=0.12	t=0.52	t=0.41
(Effect size)	d = 0.16	d = 0.28	d = 0.22	d=0.08	d=0.03	d=0.12	d=0.10
Father
Residing with his child (n=24)	0.43±0.93 (-1.86−2.06)	0.4±0.7 (-0.83−1.7)	0.41±0.82 (-1.12−1.59)	10.13±2.68 (3.0−14.0)	10.88± 2.54 (2.0 −14.0)	9.63±2.30 (4.0 −12.0)	9.46±3.16 (2.0 –14.0)
Not residing with his child (n=55)	-0.19±0.91 (-2.06−2.09)	-0.17±1 (-2.34−1.38)	-0.18±0.96 (-2.13−2.07)	9.82±2.46 (4.0−14.0)	9.65±2.47 (3.0−14.0)	8.42± 2.79 (1.0− 14.0)	8.22± 2.91 (0−13.0)
Statistical test	t = 2.77**	t = 2.54**	t = 2.61**	t=0.50	t=2.0*	t=1.86*	t=1.70*
(Effect size)	d =0.69	d = 0.63	d = 0.65	d=0.12	d=0.49	d=0.45	d=0.42
Grandparents
Mother, father, and children (n=6)	0.13±0.93 (-1.13−1.23)	0.40±0.55 (-0.24−1.04)	0.22±0.89 (-0.91−1.51)	10.0±3.69 (3.0−14.0)	9.50±4.14 (2.0−13.0)	9.83±2.93 (4.0−12.0)	7.83±4.31 (2.0−14.0)
Mother, grandparents, and children (n=56)	-0.15±0.93 (-2.06−2.09)	-0.16±0.98 (-2.34−1.38)	-0.11±0.99 (-2.13−2.07)	9.75±2.53 (4.0−14.0)	9.77±2.05 (3.0−13.0)	8.41±2.83 (1.00−14.0)	8.18±3.0 (0−13.0)
Mother, grandparents, father, and children (n=17)	0.43±0.95 (-1.87−2.06)	0.39±0.87 (-1.53−1.70)	0.29±0.81 (-1.12−1.59)	10.41±2.03 (7.0−13.0)	11.06±1.75 (8.0−14.0)	9.65±1.90 (5.0−12.0)	10.24±2.49 (3.0−14.0)
Statistical test	F = 2.54	F = 2.85	F = 1.31	F = 0.45	F = 1.87	F = 1.90	F = 3.42
(Effect size)	η2 = 0.06	η2 = 0.07	η2 = 0.03	η2 = 0.01	η2 = 0.05	η2 = 0.05	η2 = 0.08

Notes: *p < .05 **p < .01.

3.2. Parenting domains and children's developmental outcomes

Observed average levels for the PICCOLO's domains were 9.91 (SD=2.51; Min=3, Max=14) for affection, 10.03 (SD=2.54; Min=2, Max=14) for responsiveness, 8.78 (SD= 2.70; Min=1, Max=14) for encouragement, and 8.59 (SD=3.02; Min=0, Max=14) for teaching.

Statistically significant Pearson's correlation coefficients were found between children's development and parenting. Firstly, communication development was associated with the mother's responsiveness and teaching. Likewise, problem-solving was related to the mother's responsiveness, encouragement, and teaching. Finally, personal-social development was significantly associated with the mother's teaching. Supplemental material shows a correlation plot.

Table 3. Pearson's correlations between ASQ-3 Z scores and PICCOLO domain scores.

	2	3	4	5	6	7
1 Affection	.57**	.74**	.33**	.00	.10	.06
2 Responsiveness		.62**	.60**	.34**	.26*	.10
3 Encouragement			.54**	.18	.24*	.19
4 Teaching				.27*	.35**	.24*
5 Communication					.36**	.35**
6 Problem solving						.37*
7 Personal-social

Notes.: *p < .05 **p < .01.

Table 3. Pearson's correlations between ASQ-3 Z scores and PICCOLO domain scores.

	2	3	4	5	6	7
1 Affection	.57**	.74**	.33**	.00	.10	.06
2 Responsiveness		.62**	.60**	.34**	.26*	.10
3 Encouragement			.54**	.18	.24*	.19
4 Teaching				.27*	.35**	.24*
5 Communication					.36**	.35**
6 Problem solving						.37*
7 Personal-social

Notes.: *p < .05 **p < .01.

3.3. Predictive models for infants' ASQ-3 developmental scores

The results indicate that higher levels in all developmental areas of the ASQ-3, that is, communication (β = -0.23, p = .03; 95% CI = [-0.95; -0.06]), problem-solving (β = -0.21, p = .04; 95% CI = [-0.89;-0.03]), and personal-social (β = -0.22, p = .04; 95% CI = [-0.92;-0.03]) were predicted by having the father residing with their children. In turn, being a girl predicted a higher level of communication (β = 0.23, p = .03; 95% CI = [0.04;0.85]), problem-solving (β = 0.20, p = .05; 95% CI = [0.00;0.79]), and personal-social (β = 1.24, p = .01; 95% CI = [0.17;1.01]) development.

Both for communication and problem-solving development, there were interactions with the support received by the mother for childcare. Specifically, communication scores were higher when the mothers received childcare support but only when the children were not attending preschool (unstandardized marginal B = 0.41 and 95% CI = [0.12;0.70]), and the interaction between mothers' affection and mothers receiving support in the care of their children (β = -0.71, p = .06; 95% CI = [-0.82;0.13]) predicted higher scores in problem-solving development. The interaction between the mother's affection and the child's gender (β = 1.24, p = .01; 95% CI = [0.17;1.01]) showed higher personal-social scores.

It should be noted that mothers' teaching scores (β = 0.32, p = .01; 95% CI = [0.09;0.54]) and mothers dropping out of school (β = 0.25, p = .02; 95% CI = [0.13;1.30]) were significantly associated with child problem-solving development only, but not with the other developmental areas.

Table 4 summarizes the regression model obtained based on a feasible solutions algorithm using ASQ-3's communication, problem-solving, and personal-social development scores (n = 79) as the response. Categorical factors include the category to be compared to the reference category within parentheses.

4. Discussion

In the current study, we aimed to determine the relationship between some familial sociodemographic factors (e.g., mother who had dropped out of school, employment status, support in the care of child, child's gender, fathers/grandparents living with their child, and children attending preschool) and parenting characteristics such as affection, responsiveness, encouragement, and teaching [6]of Chilean adolescent mothers, concerning communication, problem-solving, and personal-social developmental outcomes in typically developing children. To better understand our results, we will discuss how sociodemographic variables and adolescent motherhood are related to child development below.

4.1. Sociodemographic factors and children's developmental outcomes

Comparative analyses showed significant differences between child development based on certain sociodemographic variables. Our results showed that children with employed mothers and mothers who had not dropped out of school had higher problem-solving developmental scores. Regarding the association of maternal employment and child cognitive development, as mentioned in the introduction, it has been shown that the children of working Chilean mothers demonstrate more robust cognitive performance since maternal employment contributes to their parenting skills, leading to a higher level of child development [38,39].The association between employment and parenting skills was also observed in this study since it was found that employed mothers had higher scores in the teaching domain. In turn, the teaching domain was significantly associated with child cognitive development in this sample. This result contributes to the current literature that discusses the reconciliation between work and child care [96], in which it has been shown that the mother's employment does not interfere with their children's daily activities [97]. In fact, according to the findings of Huston & Rosenkrantz Aronson [98], mothers who worked for longer provided their children with higher-quality home environments and interaction. This issue will be addressed in future research that will examine the association between maternal employment and parenting as factors influencing children's developmental outcomes.

Regarding the difference in problem-solving development between children of mothers who had dropped out of school and those who had not, it is well known that socioeconomic status factors have a significant association with cognitive development, making maternal education an important variable [99]. In the case of the sample in this study, the low education that adolescent mothers usually receive could be associated with disadvantaged social circumstances that put their children's cognitive development at risk [32]. It has been found that the educational level of mothers who, regardless of income, the educational level of the mother was found to be a strong predictor of their child's cognitive development [100] (Schady, 2011) and that higher educational levels in mothers who became pregnant during adolescence predicted better achievement of children in mathematics at the start of school [101]. Therefore, the relationship between adolescent mothers' continuity of education and their children's cognitive development is interesting since mothers who do not drop out of school can make a difference in their children's cognitive development even when they belong to a low-income family. However, future studies of larger samples that include different levels of socioeconomic status are required to deepen our understanding of the influence of school dropout by adolescent mothers on their children's cognitive development.

We also found that children who resided with their fathers achieved significantly higher scores in communication, problem-solving, and personal-social development. Children living with both parents may receive more financial investment and spend more time with their fathers, which, in turn, would enhance child development [102]. Furthermore, the father's presence can help in the child's development due to fathers engaging in activities that contribute to their children's learning and providing emotional support and guidance to them; indeed, the effectiveness of parenting depends on the quantity of time parents spend with their children and the quality of the relationship they establish with them [103]. This evidence could explain why children who lived with their fathers achieved higher scores in developmental outcomes in our sample. Our results are similar to another study carried out on children of adolescent mothers, which found that those children who were in contact with their father during the first eight years of life had higher levels of social-emotional functioning and reading than those who were not in contact with their father [47].

Another aspect of our results to consider is that mothers showed higher responsiveness, encouragement, and teaching scores when fathers cohabited. These maternal factors were related to their children's development. In this context, the association between the coresidence between mothers and fathers could be a protective factor for maternal well-being [104], contributing to the quality of dyadic interaction in their children's development. More specifically, the absence of the father is an aspect that directly influences stress in socioeconomically disadvantaged Chilean mothers [105]. In adolescent mothers, the father's presence helps to reduce maternal depression [106]. In turn, a low level of maternal mental health in adolescent mothers has been related to developmental problems in their children [30]. Nevertheless, this factor was not considered in our study, but it is a relevant issue that could be considered in future studies on adolescent mothers.

Another sociodemographic variable associated with the children's developmental outcomes was the child's gender. Specifically, girls scored higher than boys in communication, problem-solving, and personal-social development. Our results are concordant with studies that have shown that girls were ahead of boys in early communicative gestures, productive vocabulary, and combining words [107]and showed stronger social interaction during the preschool stage, because the forms of social and structured play appear earlier in girls than in boys, even though these differences vary over time [108]. Also, in more recent research on children aged from 4 to 7 years, girls showed higher cognitive ability to perform mental tasks faster than boys [109]. So, it is reasonable to assume that biological components that explain functional and morphological differences in the brain between boys and girls [110] lead to a higher level of performance in girls in all areas assessed.

It should be noted that the quality of interactions between parents and their children could vary according to their gender and, in turn, be associated with their developmental outcomes. Adolescent mothers in our sample showed higher levels of responsiveness and teaching when interacting with girls than boys. Our findings are similar to previous research by Butler & Shalit-Naggar [111], who observed that mothers tend to exhibit higher levels of responsiveness towards girls than boys. Leaper [112]also reported that mothers show increased responsiveness and engage in more conversations with their daughters than their sons. It is widely recognized that explanations and interactive conversations during play significantly contribute to a child's development [113,114]. Therefore, both the biological factors that indicate that girls mature before boys and the parental characteristics of adolescent mothers when interacting with their children may explain why girls had higher scores in developmental outcomes than boys.

4.2. Parenting and child development

Parenting behaviors, as measured by the PICCOLO, have been studied in diverse ethnic groups [115,116]. Recent literature has documented the association between parenting and child development, both in families with developmental delay disabilities[26,117,118] and those with typically developing children [119,120,121]. Compared with other studies in which parenting was assessed with PICCOLO in adult mothers of typically developing children [120] and children with disabilities [26], the responsiveness domain showed the highest mean score, followed by affection, the encouragement domain and finally the teaching domain. Thus, it is possible to infer that dyadic interaction between Chilean adolescent mothers and adult mothers with their children does not vary so much regarding the frequency of parental behavior during free play nor the association between the parenting domains and their children's development.

Interestingly, the parenting domain scores in the current study sample were lower than in all previously mentioned studies. However, compared with the original study of parenting assessed using the PICCOLO in a low-income sample from the United States that included European Americans, African Americans, and Latino American families [115], in our sample, the means of the affection and encouragement domains were lower, responsiveness was similar, and teaching was higher. Also, compared with a Turkish sample [116], all domains were lower, except teaching.

Given the low economic and educational status of the mothers who participated in our study, it is curious that the mean score for the adolescent mothers' teaching domain was higher than in the other studies mentioned. Despite this, a study conducted with Chilean mothers aged from 15 to 44 years old found higher means in all domains than in our sample [35] . It would be interesting for future studies to compare these parenting domains between Chilean adolescent and adult mothers to determine whether there are significant differences. This comparison would be necessary because previous evidence has found differences in the support adult mothers receive compared to adolescent mothers [122]. In this context, adolescent mothers tend to be more depressive, feel less efficacy in their maternal role, and receive lower social support than adult mothers, which could affect their parenting [123].

The results of our study confirm the association between early parenting behaviors and the cognitive, linguistic, and social development of adolescent mothers' children and are similar to the results of research that has found a close relationship between adult mothers' parenting and child development [124,125]. Specifically, our results show that responsiveness was associated with communication and problem-solving development. This was consistent with the findings of several previous studies that have analyzed parenting interactions with children with typical development [58,61,72,73,126]. It should be noted that this finding could make a valuable contribution to the existing literature because previous research has found that maternal responsiveness in adolescent mothers acted as a mediator between socioeconomic risk and cognitive development [34].

In this study, we also found that children whose mothers displayed more encouraging behaviors during mother–child interactions showed higher scores in problem-solving development. Our results are consistent with studies that found that children of mothers who support their children's autonomy (which is associated with the encouragement domain) achieved higher levels of executive function, that is related to the children's skills to solve problems [127].

Furthermore, our findings also establish that adolescent mothers' teaching is associated with communication, problem-solving, and personal-social development. A considerable amount of research has documented the association between teaching or cognitive stimulation and children's cognitive, linguistic, and social development [125,128,129], although few studies have analyzed this in adolescent mothers. However, previous studies of adolescent mothers have shown that having a poor language-learning home environment was related to lower language development in children [130].

This view is consistent with the study by Shephard et al. [131], which established that due to factors such as living in impoverished environments and mental health problems, children born to adolescent mothers represent a population at high risk for impaired child development. However, although this study involved low-income adolescent mothers, risk variables such as maltreatment, abuse, or neglect, which are frequent factors in these vulnerable dyads [132], were not analyzed. Thus, future studies of adolescent mothers must analyze the association between their context and parenting competencies.

Given that adolescent motherhood is concentrated in low socioeconomic groups in Chile [133]and, that, in this country, there is a gap between low-income and high-income child development [134], it is relevant that our findings indicate that positive parenting in terms of responsiveness, encouragement, and teaching contributes to cognitive, linguistic and socio-individual development. The significance of these results lies in the fact that the home environment and parenting competencies significantly influence children's development [135]. From this perspective, it would be interesting for intervention programs that work with low-income adolescent mothers to strengthen parental skills to prevent cognitive and linguistic problems in children of very young mothers since children born of adolescent mothers are more prone to risk for poorer developmental outcomes [136].

In this study, the mother's affection was not significantly related to their children's cognitive, linguistic, or personal-social development. Although affection may play an essential role in children's development, it has been demonstrated that maternal warmth moderates the impact of maternal intrusiveness on child anxiety and separation anxiety [137], being more associated with closeness and confidence between mother and child. A study comparing adolescent and adult mothers interacting with their children found that adolescent mothers spent more time in negative engagement with their children than adult mothers, which was associated with insecure attachment [56]. Thus, maternal affection is more associated with secure attachment than children's cognitive and linguistic development [138]. This may explain why the adolescent mother's affection was not associated with any of the evaluated areas of child development.

4.3. Predictive models for child's development

Regarding the association of sociodemographic and parental factors with children's development, higher scores in communication development could be explained by higher maternal responsiveness, being a girl, having the father residing with their children, and the mother receiving childcare support but only when children did not attend preschool. Previous research on the quality of parent–child interactions (including both mothers and fathers) has concluded that affection is higher when parents interact with their daughters than with their sons [82,139,140]. Moreover, fathers were more attentively engaged with their daughters and used more analytical language than their sons [141]. It is well known that parents' responsiveness promotes expanded word learning in children's early years [142]. This presumption could partially explain why fathers residing with their children, the mother's responsiveness and the children's gender play an essential role in communication development.

However, to determine whether this result is due to the evidence presented, it would be necessary to analyze the quality of dyadic interaction between parents and their children. The importance of inquiring about the complement of parenting competencies of fathers and mothers is that it would explain higher scores in communication development in girls than in boys. However, this was not done in the current study. As we mentioned in the introduction, mothers who reside with the fathers of their children could suffer lower levels of distress [84], which, in turn, is associated with positive parenting [83]. Maternal health was not assessed in this research. Therefore, it is a subject that should be considered in future studies examining factors related to parenting in adolescent mothers.

Regarding the support that adolescent mothers receive in childcare, preschool centers contribute to reducing the gap between children from lower and higher-income families, promoting linguistic and mathematical development by reducing problem behaviors due to the rapid brain development they experience during the first years of life [143]. Nevertheless, in the Chilean context, it has been shown that child development in children aged from 6 months to 3 years who attend preschool centers and those who stay at home is similar. Therefore, the younger the child, the more significant family variables are for child development than going to preschool [144]. In this sense, it is reasonable for adolescent mothers who receive childcare support in their own homes to have better-performing children if they do not require preschool centers to support them with the care of their children. However, future studies need to analyze this interaction between the support that adolescent mothers receive in childcare and their children's attendance at preschool centers in later years because previous literature has indicated that children who attend educational centers from an early age have a higher academic performance in the long term [145]. Thus, it is possible that the association between attendance at a preschool center by children born of low-income adolescent mothers and children's development will not be evident immediately. However, it could be significant when children are older.

The problem-solving development score was higher for girls with higher mothers' teaching scores, mothers who had not dropped out of school, and fathers residing with their children. Scientific evidence shows that the mothers' educational level contributes to the cognitive development of their children, which would also be influenced by a cognitively stimulating home environment [135]. Our findings are consistent with recent research as mentioned above. Furthermore, when fathers engage positively in activities with their children, teaching activities may promote learning and development [146]. In future studies, it will be necessary to determine what kind of activities fathers of adolescent mothers' children perform and whether these predict better development of children's problem-solving skills.

Finally, the children's social-personal development score was higher when fathers resided with them and in daughters of those mothers with higher levels of affection. Our findings are consistent with a study that found that the fathers' coresidence with the adolescent mother's children during the first three years of life was related to lower externalization problems in children [84]. Also, the evidence has shown that maternal affection/warmth is related to stronger regulation and prosocial behavior in their children [147,148] and that it is higher in daughters than in sons [149]. Although our sample did not show a direct correlation between a mother's affection and children's development or the difference between boys and girls, our findings show that mothers show more affection towards girls when the fathers reside with them from an early age. More studies are required to analyze whether this phenomenon occurs in samples of adult and adolescent mothers and fathers.

4.4. Limitations and future directions for research

The present study supports the literature on adolescent mothers' parenting of the typically developing child. However, this research has several limitations that should be taken into account. First, the study sample size might appear to be small since it only included 79 participants. Nevertheless, post-hoc power analyses for this sample size indicate that given a predictive capacity of R2=.20 in models with five predictors and α =.05, a statistical power greater than .95 is still obtained. In this regard, the sample size did allow us to reach conclusions concerning the effects adequately enough. Nevertheless, it is not possible to generalize these results since the sampling was intentional and its representativeness might thus be affected.

Additionally, it is necessary to remember that the presence of correlations does not imply causality. Specifically, in this descriptive and cross-sectional study, the term "predict" does not mean direct causality but instead estimates child development scores based on scores of sociodemographic variables and parental variables, that is, the predictor variables. It should be noted that the fact that it was not a longitudinal study may be a limitation from the point of view that the quality of interaction may change depending on the age of the child [150], the age of the mother, and her level of education [35].

Another limitation to consider is that the tool we used to measure child development was a caregiver report rather than a behavioral observation measurement instrument, as applied in other research with typically developing children [151,152]. However, the ASQ-3 is a recommended and qualified screening tool for investigating children's development (Squires et al., 2009), and UNICEF has recommended it for developing countries [153].

In addition, some internal consistency indices associated with PICCOLO subscales were low, as also found in a study carried out by Rivero et al. [154] with a sample that included children with typical development. In this sense, although it might affect the psychometric properties of some of the tools, it is not just a particular issue of the current study. Despite this, PICCOLO has been applied to children in large samples in the United States in diverse ethnic groups [115] (Roggman et al., 2013b), Turkey[116] (Bayoğlu et al., 2013) and Spain [66]. Therefore, these studies have demonstrated this instrument's strong reliability and predictive validity for low-income families with diverse cultural backgrounds.

Finally, the study did not consider the quality of parental interaction between fathers and adolescent mothers' children. Our findings indicated that the levels of development of the children who lived with their fathers were higher in all the areas evaluated (communication, problem-solving, and personal-social). From this point of view, our results could have been more robust if we had analyzed the relationship between the parental dimensions of the fathers measured by PICCOLO and their children's development, complementing this with the parental behaviors of the adolescent mothers, as has been done in other studies with adult parents [8,155]. However, not all adolescent mothers were cohabiting with the father of their children. As mentioned in the introduction, in the Chilean context, several adolescent fathers indicated that they did not live with their children [41] . Further analyses of adolescent mothers, fathers, and their children are needed to determine the consistency of our findings.

5. Conclusions

The current study confirmed that sociodemographic and parenting factors of adolescent mothers could explain children's developmental outcomes. The above findings will contribute to determining how the presence of the father and adolescent mothers' responsiveness and teaching influence their children's communication, problem-solving, and personal-social development. It should be noted that adolescent mothers constitute a risk group in terms of social and economic factors and a low educational level. Thus, promoting adolescent mothers' knowledge about stages of child development is an issue that should be considered when planning interventions to enhance parenting and children's development in Chilean national programs. These findings highlight the importance of providing adolescent mothers with social support and guidance during motherhood to continue their studies, offering them information on entering and training in the workplace, and providing support and resources that optimize child development.