The Association between Family Relationship, Deviant Peer Socialization, Self-Concept and Aggressive Behavior among At-Risk Student in Selangor

A total of 400 student with disciplinary problem aged 13 to 18 years old from eight secondary schools in the state of Selangor were chosen using multi-stage cluster sampling. A set of questionnaires which are consists of social demographic background, family Relationship Characteristics scale, Peer Deviancy Scale, Self-concept scale and Aggression Scale were used to collect the data. Findings revealed that majority of respondents show the moderate level of self-concept, family relationship, and aggressive behavior and low level of deviant peer socialization. The results of t-test showed that there were significant differences in family beliefs, family structure and deviant peers socialization based on gender and age. There were significant differences in aggressive behavior based on gender, while self-concept has significant differences based on age. Findings also indicated that family relationship, selfconcept, deviant peer socialization have a significant relationship with the aggressive behavior. The multiple regression analysis showed that sex, deviant peer socialization, family structure and self-concept have significantly predicted aggressive behavior and socialization of deviant peers was the most important predictor, followed by family structure, self-concept, and sex. The results of this study have important implications for those involved in school and adolescents’ systems.


Introduction
Externalizing behavior is one of the issues that have attention from various party such as academician, politician, government and community at large. Externalizing behaviors is a heterogeneous concept encompassing physical aggression, verbal bullying, relational aggression, defiance and vandalism. The externalizing behavior problems refer to a grouping of behavior problems that are manifested in our society and one of the prominent phenomena is an aggressive behavior among adolescent (Campbell et al., 2000;Eisenberg et al., 2001). Adolescence is generally described as an age where various changes happening in their lives and families. The previous studies showed that the adolescent's is at risk when the shifting period from childhood to adolescence. It is a crucial time because it was the beginning of various forms of behavioral problems (Lippold et al., 2013;Bradshaw et al., 2013;Ttofi & Farrington, 2012). Due to these changes, the prevalence rates of problematic behavior increase during the early age of adolescents, then afterward beginning to decline and become constant, and subsequently decelerating (Lahey et al., 2000).
Internalizing behaviors are actions that are focused on the self, which may hurt oneself without lashing out at others. The symptoms of internal behavior that are commonly occurred to an individual are like depression, fear, substance misuse and alienation (Perle et al., 2013). In contrast, externalizing behaviors involve harming others which can be exhibited by aggression, violence, deviance and criminal behaviors (Jianghong, 2004).
According to Achenbach and Edelbrock (1978), the principle of externalizing problem behavior refers to a set of behavioral problems which are aggressive behavior. Previous studies have shown evidence that the prevalence of behavioral and emotional problems increases during adolescence age. Based on studies in Asian countries, there was an escalation in the frequentness of problem behaviors among adolescents (Amstadter et al., 2011;Farbstein et al., 2010;Khan et al., 2009;Leung et al., 2008;Mullick & Goodman, 2005;Srinath et al., 2005;Syed, Hussein, & Mahmud, 2007;Woo et al., 2007).
In Malaysia, the behavioral problems were also debated years ago. When discussing the issues on adolescent, there are several categories of behavioral problems often committed by these groups, including crime related to property, crime related to people, drug uses, traffic, firearms, and other social problems (e.g., runaway from homes, and out-of-wedlock pregnancy). Past studies stated that the behavioral problems include sexual activities, substance abuse, illegal racing, and throwing newborn in dumpsters Bahori & Ismail, 2018).
Furthermore, another problem which closely related to adolescents' group was juvenile offenders. According to the Department of Statistics Malaysia (2016), between 2014 and 2015 the number of juvenile offenders seems to be declined by 10.3% from 5,096 cases in 2014 to 4,569 in 2015, however, it was reported that there was 12.4% increase in repeated offenses from 2014 which involved 371 cases to 417 cases in a year later. In addition, according to the same statistical report, it was revealed that property-related crimes had the highest percentage of child offenders within two consecutive years which were 36.0 % in 2015 and 42.9 % in 2014 and followed by drug use (29.7%) and people-related crimes (13.4%) in 2015.
According to American Psychiatric Association (APA, 1994), the externalization constructs also include the oppositional defiant disorder, which involves oppositional behavior (negative, hostile, and defiant behavior) especially shown by young children to their parents and teachers. This shows that there are various concepts that were used by various scholars all around the world, but the this is the similar behavior problems that spread in our nation. Most countries were facing similar problems years by years and in Malaysia, the externalizing behavior start as early as 12 years old until late of adolescents at age 18 years old. Generally, the prevalence of offending tends to increase from late childhood, peak in the teenage years (from 15 to 19) and then decline in the early 20s. In 2014 and 2015 the repeated offence was recorded an increase of 12.4% from 371 cases (2014) to 417 cases (2015) (Children Statistics, Malaysia, 2016).
In discussing adolescent externalizing behavior, it is important to know its origin and its contributing factors. Prior research has highlighted several factors that influencing the externalizing behavior problems among adolescents. Among these factors, Vandell (2000) stated that parent and peer relationships are the two major socializing factors that influence the upbringing of children and adolescents. Such finding has been observed across different cultures, contexts and countries (Lee & Yoo, 2015;Viner et al., 2012). Previous findings showed that the family factor is very diverse and there are a lot of family dimensions that have a potential to influence the antisocial behavior. Jasmina & Watson (2009) found that supportive and functional families play an essential role in children life that prompt to numerous opportunities which stimulate them to the externalizing actions especially aggression and delinquency behaviors. In addition, youth involvement in delinquent behavior might be decreased by four times when they are having the great relationship with their parents, contrasted to their comrades who did not possess a good bonding with their parents (Stouthamer-Loeber et al., 2002;Johnson, Giordano, Longmore, & Manning, 2016). Indeed, there is definite evidence that family relationship has a crucial impact on the present and future development of children's conduct and well-being.
In addition, peer socialization also has a direct relationship with delinquent behavior. Friendship is a stage in interpersonal relationship where those concerned share mutual interests and feelings at a particular level, which meant that if the adolescent is attracted to a negative kind of friendship, it will lead to delinquent conducts and other forms of dysfunctional behaviors. Past study also mentioned the relationship between peer's influence and problem behavior. For example, study by Ismail, Ahmad, Ibrahim and Nen (2017) stated that the external factors which are peer influence are important predictor to behavior problems than internal factors such as individual and family factors. This study was supported by other studies which indicated that peer factors contribute to various misconduct activities (Snyder et al., 2012;Dishion & Tipsord, 2011;Benson & Buehler, 2012).
Besides that, individual factors, such as self-concept, are very important elements to be considered for they may also influence adolescent behavior. Several studies have mentioned that individuals who have a weak self-concept have a higher chance of being involved in behavior misconducts. Conversely, those who have a stronger self-concept will have better behaviors (Tekinarslan and Kucuker, 2015;Parker, Martin, Martinez, Marsh, and Jackson, 2010;Bokhorst, Sumter, & Westenberg, 2010). Within the context of adolescence, when peer relations become increasingly important, sociability self-concept appears to vary across individuals and this self-concept is an important construct that should be addressed.
These aggressive and non-aggressive (i.e., rule-breaking) behaviors also been shown to be more useful in predicting antisocial outcomes later in life (Burt, Donnellan, Iacono, & McGue, 2011). While there have been studies that address social problems among the adolescents, comparative analysis of adolescent's involvement in negative activities while considering adolescent's self-concept as a factor towards externalizing problems is still lacking. This study was designed to determine the relationship between self-concept, family relationship and aggressive behavior among at risk adolescents in Selangor.

Research Questions and Research Objective
This study was conducted to answer the following questions 1) What are the levels of family relationship quality, deviant peer socialization, self-concept, and externalizing behaviors (aggressive behavior) among at risk student; 2) Are there any significant differences in family relationship quality, deviant peer socialization, self-concept and aggressive behavior according to socio-demographic factors; 3) Are there any significant relationships between family relationship quality, deviant peer socialization, self-concept, and aggressive behavior and 4) Which factors uniquely predict aggressive behavior among adolescents; The Main objective of this research is to determine the relationships between self-concept, family relationship quality, deviant peer socialization and aggressive behaviors among adolescents. To answer this and based on the research questions addressed in this study, four specific objectives were formulated as follows: 1) To identify the levels of family relationship quality (cohesion, belief and structure), deviant peer socialization, self-concept, and aggressive behavior of adolescents; 2) To determine the differences between adolescents' family relationship quality (cohesion, belief and structure), deviant peer socialization, selfconcept and aggressive behavior in terms of sex and age; 3) To determine the relationships between family relationship quality (cohesion, belief and structure), deviant peer socialization, self-concept and aggressive behavior of the adolescents; and 4) To determine which factors uniquely predict aggressive behaviors among adolescents.

Significance of Study
There has been a lot of previous researchers that studied internal and external behaviors. This study focused on the investigation of the two domains that affect external behaviors (aggressive behavior's) which are environmental factors (e.g., quality of family relationship and socialization with deviant peers) and personal factors (e.g., self-concept) towards the changing behavior of adolescents, externally.
Secondly, this research believe that is important as it may assist in the collaborations of practitioners such as educators, counselors and policy makers in order to strengthen the programmed and policies for the adolescent. The outcomes of this research may be used for identifying each level of family relationship quality, deviant peer associations and externalizing behaviors of Malaysian adolescents. Thus, it would give a better picture on the current scenario, especially on the externalizing behaviors among adolescents. Then, knowledge from this research may also be used in the school and related institution, where it can provide a guideline for conducting related programs to enhance adolesenct well being.
Thirdly, findings from this study may create awareness in the family institution for example families should give more attention toward the changes in their adolescent behaviors. This research hopefully can provide information regarding the current behaviors of adolescents. Indeed, the present study attempted to increase knowledge and provide extra information on the trends and causes of behaviors in Malaysian adolescents. The researcher believes that there is a growing need to examine the coexistence of family factors, peer influence and selfconcept in influencing adolescent externalizing behaviors. Lastly, the findings from this study were anticipated to be parallel with previous investigations especially regarding the significance of family relationships and deviant peer associations of adolescents development. Thus, this study may increase parents' and teacher's awareness of the importance of creating a secure environment conducive to the prevention of students' behavioral problems. Moreover, can also offer some benefits to the adolescents themselves, being the main contributors to this research and believes that awareness within the individual about his own behavior, self-concept, and the kinds of their friends will also increase.

Methodology
This study was utilized a cross-sectional research design by using descriptive and correlational method. The study involved 400 secondary school students (53% boys, 47% girls) from eight selected secondary school in state of Selangor. The state of Selangor was chosen because of high intensity of population of secondary school and in 2014, Selangor is seeing 10% jump in teenage crime compare the corresponding period in 2013. An inclusion criteria was determine by the researcher to ensure that, the sample selected will be representative to the studied population the inclusion criteria include secondary school students in Selangor aged between 13 to 18 years old who involved in disciplinary problems in school.
The sample was selected by using multi-stage cluster sampling technique. There were three stages of sample selection: 1) selection of districts, 2) selection of school sample, and 3) selection of student as respondent. At first stage, the selection of districts in Selangor, four district have being chosen randomly out of nine districts and the four districts was Gombak, Petaling, Kuala Langat and Klang. Second stage is the selection of schools, two schools have being randomly select from each selected district, followed by selection of respondent. Number of respondent for each school is culculate proportionately.
Lastly, the third stage was the selection of student sample. The purposive sampling was conducted where the students were purposively selected among those who have previously undergone disciplinary actions. During this stage, the sample was provided and selected by the schools counselors based on the students' record on disciplinary cases in those particular schools in order to participate in this study. An inclusion criteria was determined by the researcher to assure that the chosen sample could represent the studied population. The inclusion criteria included secondary school students in Selangor aged 13 to 18 years old, who were involved in disciplinary problems in school.

Research design
As a study design, this research utilized a correlational cross-sectional survey method with quantitative approach. This study employed two types of quantitative data analysis: 1) descriptive analysis, and 2) inferential data analysis. This analysis ware used for testing the significant differences between groups and significant relationships between variables. The study also aimed to investigate the relationship between environmental factors (family relationship quality and deviant peer socialization), individual factors (self-concept) and externalizing behavior (aggressive behavior).

Instrumentation
This study was using survey method and a self-administered questionnaire consisting of various instruments was used to gather information from the respondent. The instruments were adapted by the researcher based on questions from past studies and the questionnaire was divided into five parts, which are demographic background, family relationship quality, deviant peer socialization, aggressive behavior and self-concept.
First part is socio demographic profile, this section gathered basic personal information of the respondents such as age, family income and sex. Second part, independent variable: 1) Self-Concept-Individual Protective Factors Index Questionnaire (Phillips & Springer, 1992) was used to measure the self-concept. The scale composed of 16 items (e.g. I like the way I act). The items were measured by 4-point Likert scale (1=very false, 2=somewhat false, 3=somewhat true, 4=very true) with the maximum scores indicates strong self-concept whereas the minimum score indicates as weak-self-concept. The internal consistency reliability coefficient of the scale as 0.73 for the present study; 2) Family Relationship characteristics scale (Tolan et al., 1997). There were 32 items that was analyzed which consist of three subscales: family cohesion, beliefs about family and family structure. This is a four likert scales ranging from strongly disagree to strongly agree. The scores were 1= strongly disagree, 2=disagree, 3=agree and 4=strongly agree (e.g. Kids should value a close relationship with their family and not have to be asked to spend time at home (belief about family), I am able to let others in the family know how I really feel (Cohesion) and The children make the decisions in our family (family structure). The present study found the internal consistency reliability coefficient of the scale was 0.79.
Third part, dependent variables: The dimensions of externalizing behaviors were accessed through Aggression Scale (Orpinas & Frankowski, 2001), The aggression scale composed 11 items (e.g., I teased students to make them angry) with about how many times during the last month they had committed any of that behavior. The response alternatives for all of these items ranging from never to more than 6 times (0, 1, 2, 3, 4, 5, 6+). Each variable of selfconcept and externalizing behavior was divided into three categories which were low level, moderate level and high level. The higher number of times respondent's commit on that item, the higher externalizing behavior among the respondents. The internal consistency reliability coefficient for aggressive behavior were 0.78.

Results and Discussions
Result of this study employed two types of quantitative data analysis, firstly descriptive, which is summarizes the information of the data, and secondly inferential analysis which is test the relationships between variables.

Respondents' Social Demographic Background
A total of 400 secondary school student from the state of Selangor was involved in this study. The minimum age of respondent was 13 years old and maximum age was 18 years old (min = 15.6, SD. = 1.67). Group in early to middle adolescence (13-15 years old) was 57.0%; and group in late adolescents (16-18 years old) was 43.0%. In term of sex, 53% respondents were males and 47% were females. Data also shows that majority (78.0%) were Malay, followed by Chinese (13.5%) and Indians (8.5%).
The minimum income of parents' were RM 500 and maximum were RM 9000. This study revealed that majority 33.8% of the respondent's family income were low with monthly income were less than RM 2000 (min = 3611.55, SD = 2453.54). From the study outcomes, it can be concluded that the majority of students come from low income families. In terms of parent's marital status, most (86.3%) of the respondents' parents were married and live together. The study also found that 83.8%, of the respondent lived with both of their parents.
Family Relationship Quality There are three domains to measure family relationship which are family cohesion, family belief and family structure. The data showed that family belief (mean=33.50, SD= 3.70) had the highest mean score, followed by family cohesion (mean=33.09, SD=5.01) and family structure (mean=29.85, SD=3.04). The findings showed that majority (70.0%) of the respondents had a moderate level of family cohesion and family belief (73.8 %). For family structure, the percentage of respondents with low, moderate, and high levels were 35.3%, 44.8%, and 20.0%, respectively with the majority of them were in a moderate level.

Deviant peer socialization
The findings show that the mean scores for all the items in deviant peer socialization range from 1.40 to 2.56. Item with the highest score was "friend who skipped school without an excuse" (mean=2.56, SD=1.06), followed by "friend who been in a gang fight" (mean=1.90, SD=1.11) and "friend who hit someone with the idea of really hurting that person" (mean=1.81, SD=1.08). The data showed that majority (75.3%) of the respondents was a low in score of deviant peers' socialization (mean = 16.45, SD = 6.15), Results indicated that almost all of the respondents have been associating with their deviant friends, though most of them have a low number of deviant peers' socialization.

Self-concept
The findings show that the mean scores for all the items in deviant peer socialization scale range from 2.51 to 3.24. The highest score was for items "I get along well with other people" (mean=3.24, SD=0.72), "I will always have friends" (mean=3.13, SD=0.79) and "I like the way I look" (mean=3.08, SD=0.83). About 74.0% of the respondents had a moderate level of selfconcept (mean = 45.5, SD = 5.83).

Aggressive Behaviors
Findings also demonstrated that 68.0% of the respondents exhibited moderate aggressive behavior, (mean = 28.51, SD= 10.26). Contrarily, the percentage of respondents who reported having low and high levels of aggressive behavior were 17.0% and 15.0%, respectively

b. Inferential Analysis i) Differences in family relationship quality, deviant peer socialization, self-concept, and aggressive behaviors in terms of sex and age
T-test analysis was performed to determine the differences in family relationship quality, deviant peer socialization, self-concept, and aggressive behaviors in terms of sociodemographic factors (age and sex). Table 1 showed all the result of T-test analysis Note: SD= Standard Deviation, *** = significant at 0.001, ** = significant at 0.01 * = significant at 0.05 Result in table 1 shows that there was a significant difference in family belief (t=-4.54, p<0.001) and family structure (t= -3.09, P<0.01) between male and female. The result also shows that family belief (t = -6.11, p < 0.001), and family structure (t = -4.08, p < 0.001) were shown to have significantly different according to age. In term of deviant peer socialization, there are significant difference between male and female (t = 3.68, p < 0.001) and mean score of male's students showed higher than females. Finding indicated that males socialized more with deviant peers as compared to females.
The study revealed that males and females showed no significant difference in self-concept but there was significant difference based on age (t=-2.60, p < 0.01) older student tent to have more positive self-concept. In term of externalizing behaviors, the result shows that aggressive behavior (t =3.71, p<0.001) has a significant different between male and female.
The results show mean score for males were greater than females and this indicated that male respondents had higher aggressive behaviors than female.

ii) The Correlations between Family Relationship Quality, Self-concept, Deviant Peer Socialization and Aggressive Behaviors
Pearson correlation was employed to examine the relationships between socio-demographic characteristics, self-concept, family relationship quality, deviant peer socialization and externalizing behaviors. Thus, the analysis provided the answer for the third objective of this study. There was a significant relationship between family cohesion (r = -0.16, p <0.01) and family structure (r = 1.21, p <0.001) with aggressive behavior. Meaning that, the higher family cohesion and family structure the lower aggressive behavior exhibited by the adolescent. Adolescents was influenced by both family cohesion and family structure, while family belief did not influence the aggressive behavior of adolescents. The results also indicated that there was a positive significant relationship between deviant peer socialization and aggressive behavior (r = 0.36, p < 0.01), which meant that respondents who socialized more with peers had higher levels of aggression. Thus, the current findings were consistent with those of previous studies which established the relationship between socialization of deviant peers and externalizing conducts. This study also revealed that respondent's self-concept was significantly and negatively correlated with aggressive behavior (r = -0.16, p < 0.01). The finding indicated that respondents with a weak self-concept tend to be more in aggressive behavior.

iii) Determining the Predictors of Aggressive Behavior
The fourth objective of this study was to identify whether age, sex, family relationship quality, deviant peer socialization, and self-concept significantly predicted aggressive behaviors. To address this objective, Multivariate analysis which is multiple linear regression model involving the six predictor variables was performed. Force-entry or standard multiple regression was used to analyses the data. This type of regression also known as simultaneous or direct regression (Tabachnick & Fidell, 2007). Through this method, all the predictor's variables were entered into the analysis at one time and coefficients were calculated for the entire set of predictors. By using this method, the predictor variables were given equal treatment and all the predictors were tested in the bivariate analysis before being entered into the regression model. In addition, this study also applied stepwise regression analysis in order to measure the priority of variables of the study sample.

Force-entry Multiple Regression of Aggressive Behavior
The hypotheses proposed that the externalizing behaviors (aggressive behavior) were significantly predicted by the following variables: sex, family relationship quality, deviant peer socialization, and self-concept. In force entry procedures, selected independent variables have been statistically controlled and analyzed simultaneously to produce a unique set of predictors of adolescent behavior. As depicted in Table 3, the total variance explained by the model as a whole was 17.0% [F (6,393) = 15.26, p < 0.05]. The results showed that deviant peer socialization (β = 0.32, p < .001) had the largest coefficient for this model, followed by sex (β = -0.12, p < .05), family structure (β = -0.12, p <.05), then self-concept (β = -0.10, p < .05). Thus, hypotheses were accepted. Meanwhile, family cohesion (β = -0.11, p > .05) did not emerge as a significant predictor of aggression.
The finding reported that all the variables (deviant peer socialization, sex, family structure and self-concept) except family cohesion have a significant relationship with aggressive behavior. This means that these variables (sex, family structure, deviant peer socialization, and self-concept) are unique sets of predictors for aggressive behavior among adolescents. This can give the impression that adolescents who behave aggressively have a higher association with deviant peers, lack of good family relationship and low in self-concept. Plus, males have a higher tendency to engage in aggressive behavior.

Conclusion and Implication
The findings of these study demonstrate that there were significant different in family beliefs, family structure and socialization of deviant peers based on sex and gender. Moreover, there were also significant differences in aggressive behavior based on gender, while, self-concept only significant in terms of age.
Objective of this study was to determine the relationships between self-concept, quality of family relationship, deviant peer socialization and aggressive behaviors among adolescents. This study found that there was a significant negative relationship between family cohesion and family structure with aggressive behavior. Thus, the present study concludes that, the more intimate and close relationships between family members are (either with parents or siblings), the lower the aggressive behavior. Thus, this study supports the past study in which stated that family relationship was influencing the externalizing behavior (Baumrind, Larzelere & Owens, 2010;Hamid & Nawi, 2013;Arshat et al., 2016).
The study also found a significant positive relationship between deviant peer socialization and aggressive behaviors. During this period, adolescents engage in less conversation with their families; instead, they become more dependent on their peer groups. The findings also proved that deviant peer socialisation is one of the strong predictors of externalizing problems among adolescents and also confirmed the empirical studies which deviant affiliations significantly related to both antisocial behavior (Snyder et al., 2012).
The research findings also found that deviant peer's socialization was the most important predictor in predicting aggressive behavior, followed by family structure, self-concept and sex. This finding was consistent with past research where Ismail et al. (2017) stated that peer influences are the strongest predictor for externalizing behavior, more than family or personal factors.
Self-concept also show a significant relationship with externalizing behavior among adolescent since the study showed that the higher self-concept, the most likely to have lower externalizing behavior problems. Thus, this study lend support to the selection hypothesis. As mentioned by previous study, family relationship was related to self-concept (Shi et al., 2017;Zhou, 2014;An, Jia, & Li, 2010) and in turn contribute to behavioral problems (Torregrosa, Ingles, & Garcia-Fernandez, 2011;Sullivan, 2013). Self-concept could be considered as an important factor in preventing the externalizing behavior problems. The study suggested that, people should pay attention to both behavioral problems and self-concept of the adolescents, and the externalizing behavior problems of adolescents can be improved by improving the self-concept.
The research findings had several important implications on practical and Policy Implications. The findings of the present study also provide implications for parents, psychologists, educators and policy makers. It is important to study the level of externalizing behavior and the associated factors during the school years in order to acknowledge that some factors affect the development of externalizing behavior among adolescents. Reactions to externalizing behavior vary depending on several factors such as personal characteristics, family factors, and self-concept and as well as peer factor. Additionally, the adoption of adolescent personal factors in programs or activities can indirectly benefit adolescents themselves to become successful persons besides fulfilling their time with useful things and also provide some satisfying feelings for adolescents who are joining these programs. The current study also found that the findings will be useful for educators and teachers to improve the educational system for the benefits of students who exhibit higher levels of externalizing behavior problems. Teachers, as significant figures in school, should create a learning environment that will encourage students to attend school willingly and have fun in school without feeling insecure or being forced.
One of the implications of the study for parents is that the importance of family relationship quality, which is characterized by parent-child closeness, belief in the family and family structure, is favorable for the formation of positive behavior in adolescents. Indeed, the present findings proved that the family relationship quality have a relationship with externalizing behavior. Thus, the information obtained in this study were important to increase parents' awareness about their involvement in adolescent activities and subsequently make adaptations to improve the quality of their relationship in the family. Besides family factors, deviant peer also have a strong relationship with adolescents' externalizing behavior. It is very difficult to avoid the influence of deviant friends. Therefore, families and educators need to work together to overcome externalizing behavior among adolescents especially when engaging with deviant friends such as monitoring their activities at home and at school.
The current study also found that self-concept had significant relationships with environmental factors (e.g., family relationship quality) and externalizing behaviors. This implies that social psychologists could ascertain an understanding of how self-concept can be influenced by family relationship quality. Therefore, it may encourage social psychologist to conduct further studies emphasizing the promotion of self-concept as a protective factor against externalizing behavior problems among adolescents and their family.
This study determined the relationships between environmental factors (family relationship and deviant peer socialization), individual factors (self-concept) and externalizing behaviors (aggressive behavior) among adolescents.