The relationships of social capital, institutional support, and emotional and behavioral adjustment of vulnerable children in Addis Ababa

Abstract

Child vulnerability has been a very serious problem in Africa, including Ethiopia, despite rich stock of social capital resources and efforts from child-affiliated institutions to handle the problem. Thus, this study aimed to investigate the role of social capital variables and institutional support on vulnerable children’s emotional and behavioral adjustment (E/BA) in Ethiopia. A correlational research design was employed to investigate the relationships among variables. A total of 250 vulnerable children were randomly selected and filled adapted scales/questionnaires of the study. Independent t-test, correlation, and hierarchical multiple regression were utilized to analyze the collected data through SPSS version 20. The findings show that social capital variables and institutional support have significant negative association with vulnerable children’s E/BA. All the variables jointly explained 31.7% of the variances of E/BA, where institutional support and peer social capital were significant unique contributors. These results suggest that children with high perceived social capital resources and institutional support reported lower emotional and behavioral problems, which led to better adjustment. Finally, children’s demographic variables have insignificant correlation and predictive roles on vulnerable children’s E/BA. Social capital variables and institutional support show significant negative association with, and predict a significant proportion of emotional and behavioral adjustment problems of vulnerable children. The result suggests that any child protection practices for vulnerable children need to consider these social capital resources and strengthen them through appropriate interventions in order to improve the wellbeing of such children and ensure the sustainability of their care.

Keywords:

• institutional support
• social capital
• vulnerable children
• emotional and behavioral adjustment

1. Introduction

Child vulnerability has been a very serious problem in developing countries, especially in Sub Saharan Africa. Accordingly, the number of vulnerable children has been growing alarmingly mainly due to escalating myriad socioeconomic and political factors (Save the Children, 2015). Vulnerable children are broadly defined as children under 18 years and whose survival, care, and protection might have been endangered by several factors (MOWA & HAPCO, 2010), including poverty. In response, institutional child care, a Western-driven alternative, has been used as a prime solution to the growing number of vulnerable children, in place of traditional care mechanisms for several decades.

Institutional child care is a residential, group living arrangement for children in difficult conditions, especially orphans cared by paid staff (Engle et al., 2011). Despite its contributions in meeting some of the immediate needs of these children, their emotional and social needs are mostly overlooked in such settings (Tadesse et al., 2014). Studies conducted in the area documented that institutionalization has devastating effects on various domains of children’s development and functioning (Farooqi & Intezar, 2009; Pinheiro Mota & Matos, 2012; The St Petersburg-USA Orphanage Research Team, 2008). Besides, it has posed unintended consequences by increasing the vulnerability of orphans and excluding other vulnerable children, thereby threatening their overall future (Abebe, 2008; The Faith to Action Initiative, 2014; Foster, 2000; Tottenham, 2012).

On the other hand, most vulnerable children, including those who are institutionalized, have at least one living parent and other extended families/guardians (Browne, 2009; UNICEF, 2012) and are more prone to experience psychological problems than true orphans (Nsabimana, 2016). In addition, an overwhelming number of orphan and vulnerable children in Africa have been under the traditional social support and care mechanisms (Beegle et al., 2010, Browne, 2009; Mutiso & Mutie, 2018). On top of this, most communal societies of Africa, including Ethiopia, have an age-old indigenous social support mechanism for caring and promoting the wellbeing of marginalized and vulnerable children (Abebe & Aase, 2007; Save the Children, 2015).

Cognizant of all the above mentioned evidence, the international community has reconsidered shifting the child protection approach to family and community-based care and to capitalize on the potential resources of traditional care mechanisms to deal with the glaring problem of vulnerable children. Moreover, the need to address the root causes of vulnerability, especially poverty, through strengthening the capacity of families and communities via appropriate interventions was underscored in the shift (Nsabimana, 2016; United Nations General Assembly, 2010).

Consequently, child affiliated Non-Governmental Organizations (NGO) have been providing support to vulnerable children through building the socioeconomic capacity of families and communities, yet the practice is still at its infant stage in Ethiopia. In this regard, evidence indicated that NGOs can play a crucial role in child protection efforts by improving the socioeconomic rights of orphan and vulnerable children (OVC) (Department of Gender, Youth and Community Services, 2003:19). Besides, some intervention programs on OVC in Tanzania and Kenya were found to be associated with fewer emotional problems and higher pro-social behaviors (Nyangara et al., 2009). However, the contribution of NGO’s support, which represents institutional support in this study, to the well being of vulnerable children who are being supported within their family and community contexts was not yet studied.

Furthermore, social capital, a resource embedded in social relationships and networks, has been a potential resource that can facilitate the positive outcomes of children (Bourdieu, 1986; Coleman, 1988; Putnam, 1993). This is especially true in traditional society, like Ethiopia, that has a rich stock of social networks (Carroll, 2001). It is indicated that social capital serves as a buffer for a multitude of socioeconomic, psychological, and political problems (Dorsey & Forehand, 2003; Zou; Su & Wang, 2018). In addition, it has been investigated in relation to many social problems other than vulnerable children related issues in Ethiopian contexts (Aynie, 2010; Endris et al., 2017).

This study, therefore, aimed to investigate whether social capital and institutional support have significant contributions to emotional and behavioral adjustment among vulnerable children in Addis Ababa, Ethiopia.

2. Theoretical and conceptual framework

The dynamic relationships between the independent variables (institutional support, family social capital, peer social capital, and neighborhood social capital) and the dependent variable (emotional and behavioral adjustment) were explored from ecological systems. The ecological system theory proposes that human development occurs within multiple social contexts, where the various components of an ecological system can play independent and interactive effects on the development of children and youth (Bronfenbrenner, 1986). In his 1979 work, Bronfenbrenner also indicated that the most immediate systems, such as the microsystem (home, day care group, friends and classmates at school, hobby club members, or close relatives), and the mesosystem that represents the interactions among microsystems, exert the strongest influence on child outcomes Bronfenbrenner, 1979).

Social capital theory essentially refers to the resources embedded in social relationships, and networks that enhance individual functioning (Bourdieu, 1986; Coleman, 1988). Coleman (1990) conceptualized social capital as the social network and relationship-driven resources that promote various social outcomes. Parcel and Menaghan (1993) also underpin that social capital occurs at any level of social aggregation, and it is conceived as the feature and quality of social relationships in a range of social contexts (Coleman, 1990; Portes, 1998; Putnam, 2000). Moreover, it is an attribute of communities in the form of networks, institutionalized social relations (e.g., family and school), and informal relationships between neighbors and friends (Bourdieu, 1985; Coleman, 1988, 1990; Portes, 1998; Putnam, 1995). Furthermore, psychosocial interventions that consider multiple factors, including the child’s family environment, peer relationships, age, experiences, and family and peer group reactions promote the psychosocial recovery process of the target groups (Save the Children Federation, 2004).

In general, social capital, cultivated within a network of relationships among social contexts, and at times has an institutional structure, is a crucial resource for various domains of child wellbeing, especially for vulnerable children. Accordingly, a conceptual framework was formulated showing the relationships between the independent and dependent variables (Figure 1).

Figure 1. Hypothesized conceptual framework of the study based on literature.

Four interlinked independent variables are connected to an independent variable by single headed arrows pointing towards the latter, in order to examine their predictive roles.

Legend:

Shows the relationship between any two independent

variables

Shows the relationship between independent variables with

the dependent variable

Shows the co-variation among independent variables

3. Material and methods

3.1. Research design

This study employed a correlational design to examine the relationships between social capital and institutional support with emotional and behavioral adjustment of vulnerable children. The relationship between sociodemographic variables and the dependent variable was also explored.

3.2. Participants

A total of 250 randomly selected vulnerable children have participated in the study. In this study, vulnerable children refer to children of poor families aged 12–18 years and are currently supported by an institution in Addis Ababa. “Vulnerable children” instead of “adolescents” was used in this study because these participants are part of the generic definition given for the term “Vulnerable Children, that refers to children under 18 years and whose care, protection and wellbeing is at a higher risk” in the literature. Moreover, the focus of this study is on the effect of “social support”, rather than the “characteristics of the period”. Participants of this age range were selected believing that they are better able to fill the questionnaire than younger children and are more susceptible to experience emotional and behavioral problems.

The target institution, a charitable non-governmental organization (NGO) was selected purposefully because it has been supporting poor children who are living with their families and community context. As the ages of two participants were outside the target age range, two questionnaires were discarded. Thus, the final analysis was carried out based on the responses of 248 participants (female = 132, male = 116, mean age = 15.2 and SD = 1.79).

3.3. Data collection tools

The study investigated children’s perceptions of institutional support and social capital in family, peer, and neighborhood contexts, representing family social capital, peer social capital, and neighborhood social capital, respectively.

Family social capital was assessed in terms of three indicators: parent–adolescent relationship quality, parental monitoring, and parental support measured by the parent-adolescent relationship scale (15 items) (Burke et al., 2021), parental support scale (12 items) (Malecki et al., 2000), and parental monitoring scale (9 items) (Stattin & Kerr, 2000), respectively. The first two scales were changed from 6-point likert scale to 5-point likert scale in this study to facilitate the completion process and increae the response rate (Bouranta et al., 2009). Scholars also suggested that the psychometric quality of a scale is independent of the number of its responses (Brown et al., 1991; Matell & Jacoby, 1971). Consequently, all of these scales were answered on a 5-point Likert scale, with a reliability coefficient ranging from .72 (parent–adolescent relationship scale) to .85 (parental support scale). The reliability coefficients of the scales were in the range of 0.701 (parent–adolescent relationship scale) to 0.923 (parental support scale), and 0.892 for the total scale in the current study. The responses of participants to these scales were based on their relationships with their mothers since children have closer attachments to their mothers in Ethiopian culture, and that mothers are the legal caretakers for most of these children. To facilitate score interpretation, negatively worded items were recoded so that higher total score is an indicator of high family social capital.

Peer social capital was measured by Friendship Quality Scale (Bukowski et al., 1994), which consists of 23 items presented in five dimensions or subscales (companionship, conflict, help, security, and closeness) with only companionship and conflict subscales having four items. Responses used a 5-point Likert scale ranging from 5 (strongly agree) to 1(strongly disagree). The conflict subscale items were negatively worded and then reversely scored. Originally, the subscales have a reliability coefficient ranging from 0.71 to 0.86 of the subscales. In this study, the subscales were found to have an internal reliability ranging from 0.681 (conflict and companionship) to 0.833 (help), and the total scale has 0.873 reliability coefficient. The total sum of scores was used in this study, and higher score was interpreted as high peer social capital.

Neighborhood social capital was measured using two indicators: informal social control and functional social support. Informal social control is a 4-item scale adapted from Sampson (1992) and used to assess the neighbors’ involvement in supervising and monitoring children in their surroundings. Functional social support is a 6-item scale adapted from Fram (2003), which measures the availability of functional support (informational, financial & instrumental) in the neighborhood. All of them were rated on a 4-point Likert scale, with the former used responses ranging from 1 (very unlikely) to 4 (very likely), and the latter from 1 (not at all true) to 4 (completely true). These scales have reliability coefficients of 0.85 (informal social control) and 0.76 (functional social support) in the literature. In this study, the reliability coefficients of the scales were 0.685 and 0.777 for informal social control and functional social support, respectively. The internal reliability of the total scale was 0.826, showing high level of internal consistency. The sum of the scale scores was used in this study, and higher score indicates the presence of high neighborhood social capital in the study area.

Institutional support was measured using the adapted Orphan and Vulnerable Children’s Wellbeing Tool developed by Catholic Relief Services (CRS; 20009). The tool is designed to assess 10 domains (food and nutrition, shelter/environment, protection, family support, health, spirituality, mental health, education, economic opportunities, and community cohesion) of services provided by humanitarian institutions from the children’s perspective. However, only nine dimensions were used, excluding “spirituality” in this study. The tool was pilot tested on children and has been used in many African countries to measure children’s wellbeing. The well-being tool is a self-administered instrument that consists of a total of 33 items rated in three responses: 1(none of the time), 2 (some of the time), and 3 (all of the time). The tool has high internal consistency and reliability (Cronbach’s alpha = .85). Seven items were reverse-coded, and the total score was computed by taking the average of each domain’s average. The reliability coefficient of the total scale was found to be 0.862 in the current study, which is almost similar to the reliability index of the original scale. An average score of 23 or above is interpreted as an appropriate score, showing a reasonably good level of support.

Emotional and behavioral adjustment was assessed using Strength and Difficulty Questionnaire (SDQ) (Goodman, 1997). The scale has 25 items presented in five subscales (emotional symptoms, conduct problems, hyperactivity, peer problems, and prosocial behavior), with five items each. It is a self-report instrument rated on a 3-point Likert scale (0 = not true, 1 = somewhat true, and 2 = certainly true). The total difficulty scale has a reliability coefficient (α = 0.78) and the five subscales ranging from 0.47 (conduct problem) to 0.73 (emotional problem). In the current study, the reliability coefficient of the total difficulty scale was 0.0807 and for the subscales ranging from 0.481 (peer problem) to 0.756 (emotional problem), indicating a better level of reliability index. Total difficulty score is interpreted using the recommended cutoff values of the scale, normal (0–15), borderline (16–19), and abnormal (20–40).

3.4. Procedures

The research was first approved by the School of Psychology, and then permission was secured from the administrative body of the institution on behalf of the children by providing a letter of cooperation from the School of Psychology. The researcher also agreed to be confidential, not to induce any kind of harm, and to be mindful in using the information provided by participants for the study purpose with strict privacy. Moreover, written and verbal informed consents were obtained from parents and children, respectively.

In addition, the purpose of the research, their right to decline at any time if they feel uncomfortable during the data collection process and the implications of the research for their service provision were communicated to the selected participants. Data were collected with the help of social workers, volunteers, and research assistants. Above all, ethical clearance was obtained from the Research Ethics Committee of the School of Psychology at Addis Ababa University with Ref. No. SoP/Eth009/2023.

3.5. Method of data analysis

Independent sample t-test, Pearson product moment correlation, and hierarchical multiple regression analysis were used to analyze the collected data using the Statistical Package for Social Science (SPSS, version 20). Independent sample t-test was used to examine significant gender differences in the emotional and behavioral adjustments and other study variables. The assumption of the equality of variance was met through insignificant Levene’s test of homogeneity of variance result, indicating that the variances of the two groups are equal. The Pearson product-moment correlation was used to examine the relationship between institutional sport and social capital variables as an independent variable and emotional and behavioral adjustment, as a dependent variable. Hierarchical multiple regression was also employed to determine the joint and unique influences of the independent variables on the dependent variable by controlling for the role of sociodemographic variables.

The sociodemographic and independent variables were entered into the analysis in two blocks. Sociodemographic variables, such as sex, age, and grade level, were entered first (Model 1), and then the independent variables, including family, peer, neighborhood social capital, and institutional support, were entered as the second block (Model 2). Overall, the models were tested using the most commonly reported fit indices in the literature, such as the Relative Chi-square (CMIN/DF), Comparative Fit Index (CFI), Tucker—Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR).

The criteria of a good fitting model; CMIN/DF value < 3 (Schumacker & Lomax, 2004); Comparative Fit Index (CFI), and Tucker—Lewis Index (TLI) values > .95 (Kline, 2011), and RMSEA and SRMR values < .08 (Hu & Bentler, 1999) are used. Accordingly, the result shows that model 2 was a best fitting model with fitness test values; CMIN/DDF = 2.217, CFI = .998, TLI = 9.57, RMSEA = .070 and SRMR = .032, than model 1 with fit indices; CMIN/DF = 19.999, CFI = .954, TLI = .725, RMSEA = .276 and SRMR = .075. In general, model 2, which included both the sociodemographic and the independent variables, was found to have good fitness to the data.

4. Results

As indicated in the previous sections, the analysis was conducted on 248 participants to explore the relationship between institutional support and social capital variables with emotional and behavioral adjustment. Table 1 shows the characteristics of the respondents, including sex, age, and grade level. Accordingly, of the total sample, 132 (53%) were female, and the remaining 116 (47%) were males. Regarding the age of the respondents, 109 (44%) of the participants were in the age range of 12–14 years. The remaining 139 (56%) were between 15 and 18 years of age. Table 2 also reveals that 107 (41.5%) participants were from grades 7 to 8. The second and third largest number of participants, 71 (28.6%) and 50 (20%) were from grade 9-10 and grade 11-12, respctively. The remaining, 24 (9.7%) participants were in grades 5 and 6. This result showed that the majority of the participants were first-cycle secondary and second-cycle primary students.

Table 1. Sociodemographic characteristics (sex, age, and grade level) of the respondents (N = 248)

No.	Demographic variables	Frequency (%)
1	Sex
	Female	132 (53.2)
	Male	116 (46.8)
2	Age
	12-14 years	109 (43.9)
	15-17 years	105 (42.3)
	18-19 years	34 (13.7)
3	Grade level
	5 & 6	24 (9.7)
	7 & 8	103 (41.5)
	9 & 10	71 (28.6)
	11 & 12	50 (20.2)

Table 2. Mean difference test results by Sex. Mean and Standard deviation (N = 248)

Variables	Sex	N^1	Mean	Std.^2	T^3	Df^4	Sig.^5 (2-tailed)
Family social capital	Male	116	12.30	2.15	-3.898	246	.000
	Female	132	13.37	2.18
Neighborhood social capital	Male	116	5.84	1.31	-.869	246	.386
	Female	132	5.98	1.23
Emotional & Behavioral Adjustment	Male	116	16.83	5.36	1.62	246	.106
	Female	132	16.26	5.41
Institutional Support	Male	116	22.04	2.61	-1.880	246	.061
	Female	132	22.64	2.44
Peer social capital	Male	116	18.76	2.91	-2.921	246	.004
	Female	132	19.80	2.71

*p< .05 **p< .01 ***p< .001.

¹Total number of participant in the study.

²Standard deviation.

³Critical value of the t statistic.

⁴Degree of freedom.

⁵Significance level of non-direction test.

Table 2 presents an independent sample t-test carried out to examine whether there was significant gender difference in the mean scores of institutional support, social capital, and emotional and behavioral adjustment among vulnerable children. The result confirmed that there was a significant gender difference in the family social capital scores, t(246) = −3.89, p < .001, indicating that females have higher social capital scores (M = 13.37, SD = 2.18) than males (M = 12.30, SD = 2.15); peer social capital score, t(246) = −3.89, p < .01, with females having higher peer social capital scores (M = 19.80, SD = 2.71) than males (M = 18.76, SD = 2.91). However, there were no statistically significant gender difference in neighborhood social capital, t(246) = −.87, p > .05, emotional and behavioral adjustment, t(246) = 1.62, p > .05, and institutional support, t(246) = −1.88, p > .05, respectively. This significant gender difference implies that females have higher family social capital and more friendship networks than their counterparts, but this significant difference did not account for differences in their emotional and behavioral adjustment.

Table 3 shows the associations between the independent and dependent variables. The results show that the four independent variables have significant moderate negative relationships with emotional and behavioral adjustment problems: institutional support, r(146) = −.496, p < .01; peer social capital, r(146) - .415, p < .01; family social capital, r(146) = −.349, p < .01, and neighborhood social capital, r(146) = −.265, p < .01. These results suggested that institutional support and peer social capital have the first and the second higher associations with emotional and behavioral adjustment of vulnerable children. Family and neighborhood social capitals were also moderately related with the dependent variable. In general, the correlation result depicted that high level of institutional support and social capital in families, peers, and neighborhoods tends to be related with better adjustment of vulnerable children, by reducing emotional and behavioral problems. The table also shows that there were significant positive and moderate relationships among the independent variables of the study.

Table 3. Bivariate correlation, mean, Standard deviation, and Cronbach's alpha of the study variables (N = 248)

Variables	1	2	3	4	5
1. Emotional & behavioral Difficulties	-
2. Peer social capital	-.415**	-
3. Neighborhood social Capital	-.265**	390**	-
4. Family social capital	-.349**	.311**	.462**	-
5. Institutional support	-.496	.381**	.489**	.569**	-
Means	16.53	19.32	5.92	12.87	22.36
Standard deviations	5.39	2.84	1.27	2.22	2.53
Crownbach’s Alpha	.71	.76	.76	.79	.85

*p< .05 **p < .01 ***p < .001

Table 4 presents hierarchical multiple regression analysis that intends to examine the predictive roles of institutional support and social capital variables on emotional and behavioral adjustment, after controlling the influence of sociodemographic variables. The analysis was regressed using two models (Model 1 and Model 2) (See Section 3.5). The result shows that model 1 explained an insignificant proportion of the variance in emotional and behavioral adjustment. The overall regression model (Model 2) was statistically significant (R²=.317, F(7, 240) = 15.53, p < .001), indicating that all the variables jointly explained 31.7% of the variance in the emotional and behavioral adjustment of vulnerable children. Moreover, independent variables (institutional support and social capital variables) have a significant addition to the proportion of explained variance in emotional and behavioral adjustment (R²A = .289). In other words, 28.9% of the explained variance of emotional and behavioral adjustment of vulnerable children was accounted by the independent variables.

Table 4. Hierarchical multiple regression results for emotional and behavioral adjustment

Variables	B^†^6	95% CI^‡		SEB^‡†	^⶧	R^2‡§	R^2A^†¶
		LL^§	UL^¶
Step1
Constant	5.66***	3.897	7.432	.897		.028	.028
Sex	−.209	−.518	.100	.157	−.087
Age	.164	−.017	.345	.092	.247
Grade level	−.187	−.364	−.009	.090	−.281
Step2
(Constant)	11.86***	9.89	13.82	.99		.317***	.289***
Sex	.044	−.23	.318	.139	.019
Age	.106	−.049	.261	.079	.159
Grade Level	−.115	−.266	.036	.077	−.173
Family social capital	−.047	−.120	.027	.037	−.087
Neighborhood social capital	.057	−.064	.179	.062	.060
Peer social capital	−.112***	−.163	−.062	.026	−.267***
Institutional support	−.174***	−.239	−.108	.033	−.367***

*p< .05 **p < .01 ***p < .001

^†Unstandardized Beta coefficient

^‡Confidence Interval.

^§Lower Limit of the confidence interval.

^¶Upper limit of the confidence interval.

^‡†Standard Error of the unstandardized Coefficient.

^¶§Standardized Beta Coefficient

^‡§R Square represents the total variance explained by all variables entered in the model.

^†¶R square change shows the extra explained variance by the next variable added in the model.

Concerning the unique contributions, institutional support was found to have a strong and negative unique predictor of emotional and behavioral adjustment (ß = −.37, p < .001), suggesting that higher institutional support appears to be associated with lower emotional and behavioral adjustment scores of vulnerable children. Moreover, social capital variables, such as peer social capital, were also significant negative predictors of emotional and behavioral adjustment (ß = −.27, p < .001), indicating that vulnerable children with high peer social capital tend to experience better emotional and behavioral adjustment. In other words, peer social capital resources tend to ameliorate emotional and behavioral problems and facilitate more positive emotions and behaviors for such vulnerable children. However, family and neighborhood social capitals were found to have insignificant unique contributions in the prediction of emotional and behavioral adjustment of vulnerable children. This is unexpected result may be explained by factors, such as the characteristics of children’s age, the socioeconomic conditions of their parents, the neighborhood conditions, and others, where the details are beyond the scope of this study.

5. Discussion

This study explores the relationships among institutional support, social capital variables, and emotional and behavioral adjustment of vulnerable children. Accordingly, the results revealed non-significant gender differences in the emotional and behavioral adjustment of vulnerable children. This suggests that male and female participants of the study were not different in their emotional and behavioral adjustment. The result also shows that sociodemographic variables (i.e., age, sex, and grade level) were not significantly related to the emotional and behavioral adjustment of these children, which is inconsistent with the findings (Li et al., 2020), who found a statistically significant association of gender and grade level with negative and positive mental health of children. On the other hand, the current finding that suggests demographic variables are insignificant in the prediction of the dependent variable is supported by previous study (for instance, Popliger et al., 2009). This study underpins that demographic variables explain an insignificant proportion of the variance in emotional and behavioral functioning.

The present study also demonstrated that institutional support and social capital variables (family, peer, and neighborhood social capitals) were statistically significant but negative relationship with the emotional and behavioral adjustment of vulnerable children. These results suggest that children who reported a high level of institutional support and social capital in multiple contexts (family, peer, and neighborhood) tend to experience fewer emotional and behavioral problems, leading to better adjustment. In addition to this, the regression analysis shows that the overall regression model was significant and all the variables in the model jointly explained 31.7% of the variance in emotional and behavioral adjustment. Taken together, the results suggested that the independent variables have significant associations and played a significant role in the prediction of emotional and behavioral adjustment of vulnerable children. Moreover, the negative association of independent variables with the dependent variable is an indication that the former variables improve the emotional and behavioral adjustment of children by constraining the problem behaviors.

The present results are consistent with many previous studies (e.g., Li et al., 2020; Novak & Kawachi, 2015; Novak et al., 2016) reported that social capital is inversely related to adolescent’s psychological distress. Additionally, Hunduma et al. (2022) indicated that higher social capital tends to decrease mental health problems of school adolescents. In this study, it was indicated that peer social capital and institutional support were a significant unique predictors of emotional and behavioral adjustment of vulnerable children. Regarding the effect of peer social capital, previous findings provide consistent result (Popliger et al., 2009), suggesting that friend support accounted for a significant proportion (16%) of emotional and behavioral functioning. Likewise, higher social capital in peers was found to promote positive mental health in children and adolescents (Yugo & Davidson, 2007; Zambon et al., 2010). These two findings also added that positive peer relations are an effective handling mechanism for negative mental health. Yet another study complements the above findings by amplifying the fact that high social capital in peers leads to lower anxiety and depression levels in children. The evidence also remarked that peer social capital has a direct influence on positive mental health, where high peer social capital improves the self-esteem and self-efficacy level of children (Li et al., 2020).

Similarly, this study depicts that institutional support has a significant negative association with and a unique predictor of children’s emotional and behavioral adjustment. The result sugests that children with high perceived institutional support tends to have lower emotional and behavioral problems, which in turn, leads to better adjustment. This result is in good terms with other studies (Ayodeji et al., 2022), which assert that perceived social support significantly promotes children’s psychosocial adjustment, characterized by increased self-esteem, positive behavioral adjustment, and improved mental health. In general, social support is perceived as an essential contributor to the psychosocial adjustment of children with special needs. Additionally, Van Griensven et al. (2006) reported a significant but negative association between perceived community support and depression and post-traumatic stress disorder (PTSD) symptoms in people under difficult conditions (post-tsunami). The findings also indicate that increased perceived community support was positively related to the recovery process both at individual and family levels after disaster experience. Likewise, community support plays a protective role by diminishing the severity of PTSD and depression, as well as restoring the functioning level of individuals in their daily activities (Andermann, 2002).

However, in the current study, neighborhood social capital explained an insignificant proportion of children’s emotional and behavioral adjustment of vulnerable children. In contrast to this result, previous evidence (Erdem et al., 2015; Jonsson et al., 2020; Sund et al., 2009) suggesting that neighborhood social capital was significantly associated with and played a buffering role in the psychological well-being of adolescents. This insignificant result in the current study may be explained by the fact that these children are living in a neighborhood characterized by social isolation, social exclusion, or social marginalization. This, in turn, leads to a lack of collective efficacy and low social cohesiveness that ultimately result in low or no social capital available for vulnerable children in that neighborhood (Sampson et al., 1999).

Finally, the present study revealed unexpected result that family social capital failed to have a significant contribution in the perdition of their children’s emotional and behavioral adjustment. This result contradicts with many studies (El-Dardiry et al., 2012; Li et al., 2020; Rothon et al., 2012; Wit et al., 2011; Wright et al., 2001) that demonstrate positive and significant role of family social capital on children’s wellbeing. This inconsistent result may be explained by the influence of family characteristics, such as, family structure, parents’ work condition, and the number of children in the family. The social capital literature suggests that high social capital is more likely in families where both parent’s present, no dual-earning parent, and small number of children in the household (Coleman, 1988). Therefore, insignificant result might be due to the presence of single and dual-earning parent, as well as too many children in the family of these children, in which cases social capital could be low or absent. This may be partly true that single parenthood and/or extreme poverty are the two criteria considered to qualify for the support in the target institution. Consequently, the single parent with a low economic status might be busy doing different activities to cover the living costs of the family. As a result, these parents hardly get enough time and attention to invest in and monitor their children. All these conditions, in turn, make the presence and production of social capital hardly possible.

6. Conclusions

Overall, this study revealed that the sociodemographic variables did not have significant association with the emotional and behavioral adjustment of vulnerable children. The four independent variables (institutional support and social capital in families, neighbors, and peers) have significant but negative associations with and explained significant proportion of the variance in emotional and behavioral adjustment of vulnerable children. The results also demonstrated that institutional support and peer social capital were the only significant unique contributors to children’s emotional and behavioral adjustment. These results suggest that higher levels of institutional support and peer social capital appear to be related with lower emotional and behavioral adjustment problems. This result entails the need to consider social capital resources in any child protection efforts and enrich them by designing appropriate interventions so as to sustainably care and protect the well-being of vulnerable children.

7. Limitations of the study

The study has encountered some limitations. First, although the researcher explored the association between social capital and institutional support with the emotional and behavioral adjustment of vulnerable children, this association hardly shows causation. Second, the study includes only a few demographic variables for children, while there are multiple children- and family-related variables that might affect the psychological adjustment of these children. Third, the current study has investigated the issue by using quantitative data from children’s perspective only and with small sample size. This, in turn, may compromise the completeness and comprehensiveness of the result, as well as limit its generalizability.

8. Implications for practice and future research

Despite these limitations, the findings of this study suggest important implications for child protection practices and intervention efforts. First, the negative association between independent variables and the emotional and behavioral adjustment of vulnerable children may suggest that any child protection policies and practices need to consider the protective role of these factors and work on capacitating these socializing agents in order to protect and further the psychological wellbeing of children under difficult conditions.

Second, family social capital shows an insignificant association with the emotional and behavioral adjustment of their children in this study. This may be an indicator that parents are not properly playing their parenting role in the family probably due to economic hardship and stress. Therefore, future child protection efforts for such children should take into account the economic condition of the family and design appropriate capacity-building interventions to better address the sources of their children’s vulnerability and improve their wellbeing.

Third, the current study investigates very few individual factors using only quantitative research approach and small sample size. Therefore, future research works that integrate both quantitative and qualitative methods, larger sample size (both parents and children), and that include multiple children, family, and other social factors in their research process are needed. Moreover, these future research works need to employ more advanced data analysis techniques like (SEM) in order to fully uncover the existing dynamic relationships and better inform the child protection practices.

Open scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered. The data and materials are openly accessible at https://doi.org/10.1080/23311908.2023.2260109

Correction

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Acknowledgements

The authors thank Addis Ababa University and Dilla University for financing this study. We are also grateful to all those who supported us during the process of this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data that support the conclusion of this study will be provided upon request after the article is accepted for publication. https://drive.google.com/file/d/1YuhYdPSk0SyVIZrz-t9UZyg5XwIx_Uro/view?usp=drive_link

Additional information

Notes on contributors

Simachew Ayalew

Simachew A. Asmare is a Senior Lecturer at Dilla University, Institute of Education and Behavioral Studies, Department of Psychology. He is currently a PhD Fellow in Social Psychology Program at Addis Ababa University, College of Education and Behavioral Studies, School of Psychology.

Tamirie Andualem

Tamirie A. Adal (PhD, Addis Ababa University) is Associate Professor of Applied Developmental Psychology, School of Psychology; Editor-In-Chief of Ethiopian Journal of Behavioral Studies (EJOBS); College of Education and Behavioral Studies; Addis Ababa University. He is currently working as Academic Supervisor (Masters and PhD level students), Consultant in different Educational Projects and Peer Reviewer in Local and International Academic Journals.