|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002068377
007 cr bnu|||uuuuu
008 100413s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003142
The influence of parental support on antisocial behavior among sixth through eleventh graders
h [electronic resource] /
by Jos Ordez.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 102 pages.
Dissertation (Ph.D.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: The primary objective of this study was to explore the influence of parental support on antisocial behavior among 1514 adolescents from Sarasota County (Florida). An integrated multilevel approach was developed considering elements of the social support paradigm and social learning theory. Using Hierarchical Linear Modeling (HLM), the results suggest that both paternal and maternal support were significant factors in the prevention of antisocial behavior. However, paternal support demonstrated to be stronger when students justified school misbehavior. At the school level, the findings suggest that the influence of parental support to reduce antisocial behavior competes with favorable definitions toward crime learned by youngsters from society and deviant peers.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Christine Sellers, Ph. D.
t USF Electronic Theses and Dissertations.
The Influence of Parental Support on Antisocial Behavior Among Sixth Through Eleventh Graders by Jos Ordez A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Criminology College of Behavioral and Community Sciences University of South Florida Major Professor: Christine Sellers, Ph.D. Thomas Mieczkowski, Ph.D. Shayne Jones, Ph.D. Michael Lynch, Ph.D. Date of Approval: September 23rd, 2009 Keywords: Parents, adolescent misbehavior, social learning, school context, multilevel modeling. Copyright 2009, Jos Ordez
Dedication To my wife Pierina and my son Jean Franco, for being the inspiration to achieve the highest goals. To my parents Robertina and Alejo, for their unconditional support.
Acknowledgements I wish to thank my Major Professor, Dr. Christine Sellers, for her support and wise advices to analyze and integr ate criminological theories. I would also like to thank Dr. Carol Bryant and the Fl orida Prevention Research Center at USF for the willingness to provi de the dataset which made possible this research effort. Finally, I would like to express my d eepest gratitude to the committee members who contributed with their expertise in the preparation of this dissertation.
i Table of Contents List of Tables iii List of Figures iv Abstract v Chapter One: Introduction 1 Chapter Two: Parental Suppor t and Antisocial Behavior. 8 Social Support Paradigm 8 Social Learning Perceptive on Parental Support 16 Parental Support and Adolescen ts Social Learning Process 21 Parental Support and Peer Influence 24 The School Context and Diffe rential Social Support 27 Differential Social Support 31 School Connectedness 33 Peer Drinking Groups 35 Chapter Three: Method 37 Conceptual Model 37 Hypotheses 39 Data Collection and Sample 39 Measures of Variables 40 Methodological Approach 46 Chapter Four: Results 50 Bivariate Analyses 50 Parental Support and Antisocial Behavior 52 Parental Supports and Social Le arning: Mediation Analyses 54 Unconditional Model 57 Differential Social Support: HLM 58 Chapter Five: Discussion 62 References 72
ii Appendices 92 Appendix A: Research Variable s in the Multilevel Model 93 Appendix B: Antisocial Behavior Scale 94 Appendix C: Parental Support Scale 95 Appendix D: Maternal Support Scale 96 Appendix E: Paternal Support Scale 97 Appendix F: Emotional Reinforcement Scale 98 Appendix G: Social Reinforcement Scale 99 Appendix H: Differential Reinforcement Scale 100 Appendix I: Neutralizi ng Definitions Scale 101 Appendix J: School Connectedness Scale 102 About the Author End Page
iii List of Tables Table 1 Correlation among Study Variables 51 Table 2 Research Models for Antisocial Behavior (Level 1) 53 Table 3 Unconditional Model: One-way ANOVA 58 Table 4 Research Models for Antisocial Behavior (Level 2) 60
iv List of Figures Figure 1 Research Model: Stage 1 37 Figure 2 Multilevel Research Model: Stage 2 38
v The Influence of Parental Support on Antisocial Behavior Among Sixth Through Eleventh Graders Jos Ordez ABSTRACT The primary objective of this study was to explore the influence of parental support on antisocial behavior among 1514 adoles cents from Sarasota County (Florida). An integrated multilevel approach was devel oped considering elements of the social support paradigm and social learning theory. Using Hier archical Linear Modeling (HLM), the results suggest that both pate rnal and maternal s upport were significant factors in the prevention of an tisocial behavior. However, pa ternal support demonstrated to be stronger when studen ts justified school misbehavio r. At the school level, the findings suggest that the influence of parent al support to reduce antisocial behavior competes with favorable definitions toward crime learned by youngsters from society and deviant peers.
1 Chapter One Introduction During the last few decades some contem porary researchers in social psychology and sociology have been intere sted in the role played by so cial support in the origin of behavioral problems among adolescents (Br onfenbrenner, 1979; Pearson & Weiner, 1985; Lin, Dean, & Ensel, 1986; Vaux, 1988). In the field of criminology, Cullen (1994) argued that notions of social support appear in the criminol ogical literature, although the research efforts have been dispersed among di fferent theoretical appr oaches. As a result, Cullen (1994) proposed to integrate these diverse findings on social support into a coherent criminological para digm to take a more comprehensive approach to the understanding of crime causation. According to Lin et al. (1986), social support represents the combination of expressive, instrumental, actua l, and perceived forms of assistance provided to an individual (Lin et al., 1986) Likewise, Vaux (1988) proposed that social support must be considered a metaconstruct integrated by th ree components: support networks resources, supportive behaviors, and s upport appraisals. Thus, Vaux (1988:29) conceived social support as "a complex transactional process i nvolving an active inte rplay between a focal person and his or her support network. Based on this theoretical background, Cullen (1994) sugge sted four major dimensions of support. The first dimension is related to the perception of support, considering that people who receive support inte rpret, appraise, and anticipate it in the context of a given social situation.
2 Second, social support is usually divided into two typologies: instrumental and expressive. Instrumental support refers to the provision of material aid, financial assistance, and information or guidance. On the other hand, expressive support is associated with the affective function of support, providing emotional feedback and social reinforcements. Third, social support occurs within an ecol ogical context that links individuals to larger social institutions such as work, school or marriage. Likewise, social support is a function of human ecology that can be descri bed as a property of groups, neighborhoods, and larger social systems. Accordingly, the co nceptualization of social support as a multilevel process allows integrating social struct ure and micro-system processes in a single theoretical approach (Sampson, 1991; Sampson 2006a; Bunge, 2006). Fourth, social support can be delivered by a formal agency or through informal relations (Vaux, 1988). Informal social suppor t may be provided within interpersonal interactions such as in the context of the parent-child re lationship. In contrast, formal social support might be supplied by schools and governmental assistance programs. Cullen (1994) has described a wide numb er of scenarios wherein social support can prevent crime. One of these settings has been the criminogenic en vironment of family life. Although, the result s of the research have demonstrat ed several family correlates of crime (i.e. Loeber, & Stouthamer-Loeber, 1986; Sampson & Laub, 1993; Rossman & Rea, 2005), Cullen (1994) indicated that scant theoretical atte ntion has been paid to how family support is involved in crime cau sation. Thus, Cullen (1994:538) offered the following proposition: The more support a family provid es, the less likely it is that a person will engage in crime.
3 In accordance with Cullen (1994), several studies have revealed that parental expressive support diminishes childrens risk of criminal involvement. For example Glueck and Glueck (1950) found that non-de linquent adolescents came from cohesive families with strong emotional ties among its members. Likewise, Alexander (1973) discovered that the families of non-delinquents were characterized by supportive communication patterns. Most recently, cont emporary research has indicated that parental support has been inversely associ ated with the development of antisocial behavior (Wright, 1995, Wright & Cullen, 2001; Perrone, Sullivan, Pratt, & Margaryan, 2004; Jones, Cauffman, & Piquero, 2007). On the other hand, the influence of pare ntal support is conceived within a developmental framework. During the earliest stage of development, infants and youth are likely to experience support from parent s more often that from any other source. Later, adolescence corresponds to a transitiona l period from parental influence to peer influence, entrance into the labor market and intimate social relationships (Wright, 1995). During this developmental process, parent s become prosocial models when they provide support to their child ren (Cullen, 1994). Consistent with social learning theory, support might be conceptualized as prosocia l modeling with sources of reinforcements for prosocial definitions. In addition, parent al support activates the social learning mechanisms favorable to the learning of pr osocial behavior, preventing the development of antisocial acts (Cullen, 1994; Akers, 1998). From the social learning pe rspective, there is research evidence that parental support fosters prosocial behavioral patterns in children and insulates them from the
4 adverse effects of deviant peers and de linquent involvement (Matsueda & Anderson, 1998; Warr, 2002, 2005; Wright & Cullen, 2001; Pe rrone et al., 2004; Jones et al., 2007). However, few studies have examined the medi ating effects of social learning mechanisms in the influence of parental support on antisocial behavior (Ardelt & Day, 2003). Cullen (1994) also emphasized that resear chers of family support must avoid what Currie (1985) called the fallacy of autonom y. This fallacy means that families cannot be studied separated from the contextual fact ors that affect it from the outside. In this manner, parental support does not occur in an isolated environment. The influence of parental support on antisocial be havior changes in different c ontexts and it is shaped by contextual sources of social support. For example, youngsters are likely to find support in school settings; adolescents may receive ad ditional support from participation in sports programs or community organizations. Considering multiple sources of contextual support, Cullen (1994) emphasizes a broader concept of Differential Social Support which is defined as the balance between the social support received for crime and the social support received for conformity. In line with social learning theory, deviant peer affiliation may represent a source of social support for crime where adolescents learn pro-delinquent definitions and technical information for success in crime (Akers, 1992; Sellers & Winfree, 1990; Sellers, Winfree, & Griffith, 1993; Sampson, 1998; Lee, Aker s & Borg, 2004; Miller, Jennings, AlvarezRivera, & Miller, 2008). On the contrary, sc hool connectedness corre sponds to a source of support for conformity where students l earn prosocial definiti ons and conventional values for success in community life (Je nkins, 1997; Welsh, Green, & Jenkins, 1999; Wilcox & Clayton, 2001; Stewart, 20 03; Wilson, 2004; Payne; 2008).
5 Founded on a multilevel approach (Cullen, 1994), the construct of differential social support suggests that the influence of parental support on an tisocial behavior may be moderated by contextual variables. Ho wever, few studies have assessed the relationship between parental support and antisocial behavior considering individual and contextual variables simulta neously (Ardelt & Day, 2003; Pe rrone et al., 2004) Both individual and contextual factors re lated to the influence of parental support play an essential role for crime prevention. Antisocial behavior among adolescents is a process that automatically involves all of the people around a youth: family, peers, schools, and communities. Contemporary rese archers in the fields of criminology and public health have taken developmental ecologi cal perspectives for the understanding of antisocial behavior, linking the role played by micro-environments and the influences received from larger social settings wher e adolescents are embedded (Bronfenbrenner, 1979; Pearson & Weiner, 1985; Lin et al., 1986 ; Vaux, 1988; Jessor, 1993; Cullen, 1994). Along these lines, adolescent misbehavior ha s been considered a serious public health problem over the last decades (Frenc h & Maclean; 2006; Mi ller, Levy, Spicer, & Taylor, 2006; Song et al, 2009). According to the Centers for Disease Control and Prevention (2008a), from 1999 to 2006, most school-associated homicides included gunshot wounds (65 %), stabbing or cutting ( 27 %) and beatings (12 %). These findings remained relatively stable in recent years a nd they were significantly higher for males and students in secondary schools (CDC, 2008a ). Likewise, the national Youth Risk Behavior Survey (YRBS), revealed that am ong ninth through twelft h graders, 35.5 % of the students were involved in a physical fight one year prior to the survey, and 18 % of adolescents had carried a weapon on at least one day during the thirty days before the
6 survey. In addition, the Youth Risk Behavior Surveillance System (YRBSS) reported that 75 % of students had ever used alcohol, 44 % indicated recent alcohol use, and 23.8 % reported having five or more dr inks in one day (CDC, 2008b). In summary, parental suppor t constitutes an important pr otective factor to prevent the development of antisocial behavior among a dolescents. At the same time, research based on the social learning approach (Ake rs & Jensen, 2006) has demonstrated that parental support promotes conventional attitudes, conforming role models, and reinforcement of conformity through parental discipline. Additionally, contextual factors related to sources of different ial social support may moderate the relationship exerted by parents on deviant behaviors. Therefore, the general purpose of the curr ent study is to assess the influence of parental support on antisoc ial behavior among sixth through eleventh graders, considering both individual and contextual factors. At the individual level, this multilevel study will explore the following question: To wh at extent is the influence of parental support on antisocial behavior among adol escents mediated by social learning mechanisms? At the contextual level, the curr ent research will analyze this question: To what extent is the influence of parental support on antisoc ial behavior among adolescents moderated by the differential social support received within educa tional institutions? In Chapter two, the social support paradigm is analyzed to understand the concept of parental support. Next, the social learni ng perspective will be used to analyze the possible role that social learning mechanisms (i.e. Emotional and social reinforcement, differential reinforcement, and neutralizi ng definitions) may play in mediating the influence of parental support on antisocial behavior. Finally, the influence of differential
7 social support in the school context will be explored, considering sources of social support for both conformity and crime. The methodological background of the study is presented in Chapter three. This chapter describes the sample, conceptual m odel, and hypotheses assessed in this study. Furthermore, the measures of research variab les and plan of analys is are explained in detail. In Chapter four, the results are describe d considering the following steps: First, bivariate analyses will determine whether or not a relationship exists between parental support and antisocial behavior. Next, if such an associati on is found, I will then examine the extent to which that rela tionship is mediated by social learning mechanisms. If the effects of parental support are not fully medi ated by the social learning variables, I will next explore the extent to which contextual support variables mode rate the relationship between parental support a nd antisocial behavior. Finally, Chapter five provides a discussion of the theoretical implications of this study as well as its implications for social intervention. The discussion will be focused on the protective role of parental support, considering social l earning mechanisms and sources of differential social support to prevent antisocial behavior.
8 Chapter Two Parental Support and Antisocial Behavior Analyzing the construct social support Cullen (1994) proposed an integrated theoretical perspective useful in criminology to organize ne w research paradigms. To accomplish this goal, Cullen (1994) presented fourteen propositions regarding the relationship between social support and crime. Founded on a wide variety of criminological theories, he emphasized the common theme of social support among these theoretical explanations to answer both micro-level and meso-level questions. Within this theoretical perspective, some propositions have been derived from so cial control, social learning, and social disorg anization perspectives. Social Support Paradigm Cullen (1994) examined the definition of social support provided by Lin et al. (1986:18) and stated that social support is a perceived or actual instrumental and/or expressive provision supplied by the commun ity, the social networks, and confiding partners. In accordance with this concept, the social support process may be objectively delivered by a social agent and subjectivel y perceived by a social actor. Also, social support could be instrumental, through the pr ovision of material goods, and expressive, through emotional assistance. Finally, social support may operate at the micro-level among individuals and at the meso-level among schools. Parental support is one of several important concepts developed within the social support paradigm. Cullen (1994) pro posed that it is less likely for a person to be involved
9 in crime if this person has received family support. Wright (1995: 29) defined the term Parental Support as parental behavior s that provide love, nurturance, empathy, acceptance, guidance, information, and mate rial resources to their children. According to Cullen, Wright, and Cham lin (1999) social support plays an essential role during childhood for crime prev ention. Likewise, research on juvenile delinquency has found that receiving parent al support is inversely related to the development of antisocial behavior (Wright 1995, Wright & Cullen, 200 1; Perrone et al., 2004; Jones et al., 2007). Moreover, some studies have recognized that lack of parental support and other family conditions have been associated with conduct problems in adolescents. Loeber and Stouthamer-Loeber (1986) pointed out that lack of parental support, parental rejection, and low levels of pa rent-child involvemen t are strong predictors of antisocial behavior and delinquency. On the other hand, researchers have found that mothers and fathers contribute in different ways to the development of social competencies and antisocial behavior during middle childhood and adolescence. In a set of developmental studies, the relationships between offspring and their mothers contrast ed with father-offspr ing relationships, and differences seem to become more important in some areas of socia lization as a function of maturational changes associated with the transition to adolescence (Collins & Russell, 1991). The extensive theoretical emphasis on differences in mother-child and fatherchild relationships as primary sources of di fferential socialization for females and males has created the need for a developmental theory of relationships to understand the
10 influence of parental support. Some research ers have attributed these differences to the amount of time that children shared with thei r parents, the quality of the experience, and contextual factors rela ted to the perception of paternal and maternal auth ority (e.g. Stice & Gonzales, 1998; Harris & Marmer, 1996; Shek, 2005). Considering this important milestone in the literature on parenting, Shek (2005) identified three main groups of studies regard ing the paternal and maternal influence on adolescent development: a) research suggesti ng that maternal influence is stronger than paternal influence (e.g. Hawkins et al., 1992; Stice & Gonzales, 1998); b) findings indicating that fathers are more influential than mothers (e.g.; Forehand & Nousiainen, 1993; Harris & Marmer, 1996); c) studies sugge sting that there is no difference between paternal and maternal influence (e.g. Carl o, Roesch, & Melby, 1998; Marshal & Chassin, 2000). The first group of studies provides evid ence about the importance of maternal support in comparison with paternal suppor t, particularly during early childhood and middle school years. According to Stice and Gonzales (1998), maternal support showed a strong influence in preventing antisocial beha vior, compared with the marginal and nonsignificant effect reported by pa ternal support. However, these findings may be related to differential exposure to mothers and fathers. The amount of time shared by children with their mothers may create significant differen ces in the quality of the parent-child relationship (Stice & Gonzales 1998; Laible & Carlo, 2004) Likewise, Kliewer, Fearnow and Miller (1996) found that maternal support was more frequently related to childrens ability to handle stressful situations compared with paternal support. In addition, middle school-age children reported to be more satisfied
11 with their relationships with their mothers th an those with their fathers, because mothers are perceived as more warm and nurturant. In this regard, Bron stein (1984) found that mothers were significantly higher than fa thers on a measure of physical nurturance, which included offering food, grooming, and showing concern for safety. Nonetheless, parenting experts have indi cated the potential negative effects of maternal employment on delinquency, suggesting that the limited amount of time shared with their adolescents is one of the caus es of juvenile delinquency. Based on this assumption, Vander Ven, Cullen, Carrozza, a nd Wright (2001) examined whether the occupational status of mothers has criminoge nic effects on their ch ildren. They found that the characteristics of maternal work have a small influence on delinquency; however, they encountered an indirect e ffect on antisocial behavior due to the lack of supervision. Likewise, researchers have found that single mothers with authoritarian or permissive parenting are risk factors for th e development of antisocial and aggressive behaviors during middle childhood (Downey, Ainsworth-Darnell, & Dufur, 1998; Underwood, Beron & Rosen, 2009). Interestingly, Ca rlo et al. (1998) determined that low levels of sociability, high levels of anger, and high levels of maternal support were associated with high levels of adolescent s aggression. These findings suggested that adolescents might perceive highly supportive moth ers as intrusive and hostile rather than as supportive (Carlo et al., 1998) These results are consistent with Hawk ins et al. (1992) who indicated that a parent-child relationship char acterized by lack of maternal involvement appears to be associated with the initiation of drug use a nd criminal activities. For example, mothers may fail to perceive drinking problems in their children because they do not fit the
12 stereotype of an adolescent drinker due to lack of involvement or reciprocal communication (Guilamo-Ramos, Jaccard, Turr isi, Johansson, & Bouris, 2006) In contrast, positive maternal involvement and control appears to discourage youths initiation into delinquency (Hawkins et al., 1992; Laible & Carlo, 2004). Conversely, the second group of studies identified by Sheck (2005) emphasized the importance of paternal in fluence on adolescent well-bei ng. Some researchers have pointed out that the role play ed by fathers in the causation of crime was largely neglected in the criminological literature (Loe ber & Stouthamer-Loeber, 1986; Forehand & Nousiainen, 1993). More recently, the study of the role of fathers in adolescent adjustment has been intensified, allowing th e development of theoretical approaches focused on fathers for crime prevention (M arsiglio, Amato, Day & Lamb, 2000; Shek, 2005). Amato (1994) found that regardless of the quality of the mother-child relationship, the closer the children were to their fathers, the happi er, more satisfied, and less distressed they reported being. Overall these findings suggest that fathers are important figures in the lives of adoles cents. Similarly, Amato and Rivera (1999) demonstrated that paternal involvement is negatively associated with the number of behavior problems shown by their children and this result holds when the level of maternal support was controlled. Furthermor e, the beneficial effects of paternal involvement were similar for biological fathers, stepfathers, Caucasian, African American, and Hispanic fathers (Amato & Rivera, 1999).
13 On the other hand, Barber and Thomas (1986) revealed that fathers differentiate their expression of physical affection and sustai ned contact on the basis of the sex of the child, with much less to a son than to a da ughter. Likewise, Siegal (1987) found that boys are directed by fathers toward the autonomy and independence necessary for instrumental behavior through positive and negative reinfo rcement and techniques of discipline and control. The father transmits to the child th e norms and expectations of the world outside the family (Siegal, 1987). In the same way, boys acquire masculine characteristics from their fathers, learning though gender identity the way they are self-perceived and the way that they perceive others (Russell & Saebel, 1997) Research findings indicate a relations hip between the absence of fathers and delinquency (Marsiglio, Amato, Day & Lam b, 2000). Loeber and Stouthamer-Loeber (1986) in their meta-analysis demonstrated the existence of a stronger association between lack of paternal involvement and de linquent behavior. According to Kaczynski, Lindahl, Malik, and Laurenceau (2006), boys inte rpret fathers hostility and withdrawal as indicating possible abandonment of the famil y, resulting in a seri ous threat to boys emotional security that leads to antisocial behaviors. Alternatively, the presence of a nurturing father has shown an influence on the externalized adjustment of a dolescents. Shek (2005) suggested that paternal influence is more important than maternal support on a dolescent substance a buse and delinquency, especially among youngsters who live under pov erty conditions (Harris & Marmer, 1996; Stein, Milburn, Zane & Rotheram-Borus, 2009). Likewise, Forehand and Nousiainen (1993) uncovered that fathers acceptance scor e was the primary predictor of adolescent functioning outside the home. Pa renting by fathers, but not by mothers, was found to be a
14 significant predictor of adol escent functioning in schools, particularly during the upper grades. The findings suggest that regardle ss of degree of involvement, fathers acceptance of and closeness to their children are critical aspects of their parenting. Along these lines, there is research evid ence on the transmission of constructive paternal parenting from one ge neration to another. Kerr, Capaldi, Pears, and Owen (2009) suggested that productive aspect s of fathers parenting, such as parental monitoring, involvement, consistent discip line, and warm parent-child relations, impact similar constructive parenting behaviors in th e subsequent generation by supporting youth achievement, self-esteem, and positive peer relations. The third group of research evidence men tioned by Shek (2005) indicates that the influence of mothers and fathers are key aspe cts in the developmen t of adolescents and that there are no differences among them. For example, Bronst ein (1984) found that Mexican mothers and fathers used similar stra tegies for discipline and control and they contributed equally to the ps ychological well-being of thei r children. Similarly, Downey, Ainsworth-Darnell, and Dufur (1998) demons trated that there were few important differences detected between children living wi th a single mother and children living with a single father, while contro lling for the number of parents in the household. On the other hand, Carlo et al. (1998) disc overed that high levels of maternal and paternal support were negatively related to anger and antisocial behavior in their children. In the same way, Stoltz, Barber, and Olsen (2005) conclude that mothers and fathers overlap their abilities to dete r antisocial behaviors: On th e one hand, mothers behavioral control is relatively more important than father s in explaining sons successive antisocial
15 behavior, and on the other hand fathers suppo rt is relatively more important than maternal support in explaining youth social initiative. Consequently, according to the research evidence both maternal and paternal support must be considered as important predictors of adolescent well-being and adjustment. However, their differential effects are related to developmental aspects of the parent-child relationship and social cond itions of the family structure (Loeber & Stouthamer-Loeber, 1986; Forehand & Nousia inen, 1993; Shek, 2005; Stoltz, Barber, & Olsen, 2005) Considering the social learning roots of the social support paradigm, the developmental approach of parent-child rela tionship is consistent with the idea that parental support depends on the reciprocal communicati on and the social learning environment that parents provide to th eir children (Kandel, 1990; Baumrind, 1991; Darling & Steinberg, 1993; Cullen, 1994). In this manner, parental support is an interactive learning process that relies on th e communication skills of the parents and the information-processing skills of the adolescents (Akers, 1998; Stice & Gonzales, 1998) As an element of the social support para digm (Cullen, 1994), pa rental support is seen by social learning theorists as an importa nt factor providing prosocial models. Thus, positive expectations from parents create the opportunity to build a law-abiding selfimage in their children where antisocial behavi or does not take place. For instance, Ardelt and Day (2002) found that parental support was associated positively with adolescents feelings of competence and re lated negatively to adolesce nts deviant behavior. As described in the next section, the social learning perspective contributes to the
16 explanation of the learning mech anisms involved in the influe nce of parental support on antisocial behavior among adolescents. Social Learning Perspective on Parental Support The origin of Akers social learning theory is directly related to Edwin Sutherland (1883-1950). Sutherland has been recognized as the most important criminologist in the twentieth century. He is best known for his D ifferential Association Theory formulated in his book Principles of Criminology in 1947. Sutherland (1947) proposed his theory in nine statements, describing the pr ocess of becoming a criminal. According to Sutherland (1947), criminal behaviors are learned within intimate personal groups in a communication proce ss. This learning process includes the acquisition of techniques for committing crim es and the development of beliefs and attitudes favorable and unfavorable to legal c odes. Sutherland (1947) called these beliefs and attitudes definitions a nd stated in his sixth proposit ion that A person becomes delinquent because of an excess of defin itions favorable to violation of law over definitions unfavorable to violation of law (p. 7). This statement is central to Sutherlands theory, because it contains the ba sic principle of Differential Association. In accordance with this principle, the pr ocess of learning criminal behaviors by association depends on the frequency, durati on, priority, and intensity of exposure to those associations. At the same time, Suther land (1947) pointed out that the process of learning criminal and anti-criminal patterns involves the same mechanisms involved in any other learning; however, it is not restricted solely to the process of imitation.
17 Despite Donald Cresseys several revisions of Principles of Criminology after Sutherlands death (Sutherland & Cressey, 1978) he preserved the nine statements of Differential Association Theory as they we re originally formul ated by Sutherland in 1947. Later, Cressey (1960) noted the pres ence of some misint erpretations of Sutherlands theory in a literature review derived from papers written in the 1950s. He called these misconceptions of th e theory, literary errors. For example, Cressey (1960) indicated that some scholars identified as a main problem for Differential Association Theory the fact that not everyone who has contact with delinquents becomes a criminal. This critique is considered a theoretical mi sinterpretation by Cressey (1960), who pointed out that the principle of di fferential association, takes in to account exposure to both criminal and non-criminal patterns. Another common misinterpr etation is the notion that associations and definitions are only learned from criminals when categorized as persons. To clarify this misconception, Cressey (1960) explained that Sutherlands theory is concerned with patterns of behaviors, no ma tter who is the bearer of su ch patterns. Cressey (1960), however, identified the most important critic ism as the lack of specificity for both criminal and noncriminal behaviors involved in the learning process. Based on this criticism, Burgess and Aker s (1966) reformulated the Differential Association Theory, incorporat ing principles of Skinners Operant Conditioning Theory. They called this reformulation Different ial Association-Reinforcement Theory. However, their purpose was not to build a diffe rent theory of criminal behavior, but to improve Sutherlands original theoretical st atements to make them testable. Afterward, Akers (1973) used the term deviant behavior instead of criminal behavior to apply
18 the theory to a wide range of deviant behaviors, such as drug use, alcohol consumption, suicide, and mental illness. Since the publication of Deviant Behavior: A Social Learning Approach in 1973, the theory has been best know n as Akers Social Learning Theory. Akers (1973) presented the reformulation of the theory (Burgess & Akers, 1966) in a seven-statement format. Those statements cl early identified some elements of operant conditioning theory, such as ope rant behavior, reinforcers, reinforcement contingencies, and differential reinforcemen t (Akers & Sellers, 2004). The basic assumption of the theory is th at the same learning process operates in different directions, produci ng both prosocial and antisocia l behavior. Akers (1973; 1998) has identified four major concepts that underlie the learning process. First, definitions represent attitudes that are associated w ith a given act. These definitions may be favorable or unfavorable to criminal behavior. Second, imitation refers to the learning of behavior through observations of sim ilar behavior in admired others. Third, differential reinforcement is the balance of anticipated rewards and punishments as consequences of behavior. Fourth, differential association represents the direct association and interaction with others who exhibit certain ty pes of behavior (interaction dimension) as well as specific patterns of social norms and values (normative dimension). Both family and peers are the most impor tant social groups providing differential associations for criminal and conforming behaviors (Jensen, 1972; Warr, 1993; Warr, 2005). These major concepts of social lear ning theory are interconnected in the behavioral sequence of the criminal behavior learning process: First, definitions favorable and unfavorable to criminal behavior learned in the past produce or inhibit the initial
19 delinquent acts. Imitation of deviant models as a learning mechanism, is very important in this step. Second, the influence of these va riables continues in th e repetitions of acts, although imitation becomes less important than it was in the first commission of the act. Third, the balance of reinforcers and punishers, established through differential reinforcement, affects the probability of repeating the criminal behavior in the future. Fourth, when definitions favorable to deviant acts are strengthened and unfavorable definitions are weakened through the differential association process the repetition of criminal behavior under similar social condi tions is more likely. Fifth, progression into more sustained patterns of deviant acts will be promoted if reinforcement, exposure to antisocial models, and pro-delinquent defini tions are not counte rbalanced by negative formal and informal sanctions and prosoc ial definitions (Ake rs, 1998) Within the social learning pr ocess, some adolescents receive social reinforcement for antisocial behavior when they obtain soci al approval from parent s and peers (Sellers & Winfree, 1990; Hwang & Akers, 2003; Wang & Jensen, 2003; Chawla, Neighbors, Logan, Lewis, & Nicole, 2008) while, other youngsters get emotional reinforcement when they perceive additional benefits from deviant behavior, such as drinking alcohol to deal with emotional strain (Eaton et al, 2004; Miller et al, 2008). As mentioned before, the learning pro cess involves the establishment of differential reinforcement toward antisocial behavior According to Akers (1992; 1998), adolescents perceive the balance of anticipat ed rewards and punishments as consequences of delinquent behavior. Depe nding on learning experiences a nd the perceived balance of reinforcers and punishers, the individual will adopt favorable or unfavorable definitions and the associated behavior through the diffe rential association process. For example,
20 research has found that deviant peer rejection is one of the most aversive consequences experienced by adolescents who do not comm it antisocial acts (Warr, 2002, French & Maclean, 2006; Chawla et al, 2008). As a resu lt, the balance of reinforcers (acceptance by peers) and punishers (rejec tion by peers) promotes antis ocial behavior through the establishment of differential reinforcement. Furthermore, awareness of legal cons equences of antisocial behavior among adolescents may diminish delinquent acts. Fo r example, Lipperman-Kreda, Paschall, and Grube (2009) showed that perceived police enforcement attenuated the effects of attitudes favorable to underage drinking. This finding suggests that adolescents perception of legal risk associated with devian t behavior affects the balance of anticipated rewards and punishments (Akers, 1998). Additionally, when neutralizing defi nitions are strengthened through the differential association process, the probability of antisocial behavior increases in a given social situation. Neutralizing definitions represent beliefs and attitudes that justify antisocial behavior among adolescents, providi ng identity or acceptance by deviant peers (Akers, Krohn, Lanza-Kaduce, & Radosev ich, 1979, Warr, 2002). In contrast, adolescents who share unfavorable definitions toward deviant behavior are more oriented to autonomy in their decisions. Also, they ar e motivated to participate in community or sport activities (Chawla et al., 2008; Eaton et al., 2004). Consistent with social l earning theory, Sellers and Wi nfree (1990) revealed that youngsters are more likely to engage in in creased school misbehavior if they choose deviant peers and express neutralizing or fa vorable definitions toward drug use. Most recently, Miller et al. (2008) obtained evid ence that students who reported favorable
21 definitions of substance use were more likely to engage in school misbehavior if they also perceived friends deviant attitudes as favorable. In sum, adolescents attitudes and per ceptions of the balance of rewards and punishments create the learning environment to promote or inhibit antisocial behavior. At the same time, research has showed that pare ntal support plays an important role in the adolescents social learning pr ocess to prevent antisocial be havior by providing prosocial models and guidance to adopt conventional va lues (Akers, 1992; White, Tice, Loeber, & StouthamerLoeber, 2002; Ardelt & Day, 2002; Felson, Teasdale, & Burchfield, 2008). Parental Support and Adolescen ts Social Learning Process According to the social support paradi gm, social support from law-abiding sources can be expressive or instrumental (Vaux, 1988; Cullen, 1994) Within the parentchild relationship context, expressive parental suppor t includes love, nurturance, empathy, acceptance, and the affirmation of one s and others self-worth and dignity. At the same time, instrumental parental support involves mate rial and financial assistance and the giving of advice, guidance, and information for the positive development of their children (Colvin, Cullen, & Vander, 2002). In accordance with social learning theor y, the presence or absence of parental support represents a differential balance that determines the nature of the parent-child socialization process. In line with this idea, Baumrind (1991) suggests the existence of four prototypes that describe how parent s reconcile the joint needs of children for nurturance and limit-setting. These four prototyp es of parenting styles are the result of different balances between demandingness and responsiveness: Authoritative,
22 authoritarian, permissive, and re jecting-neglecting. The term demandingness refers to the claims parents make on children w ith their disciplinary efforts; responsiveness refers to the ability to foster individuality. She f ound that authoritative parents, who are both demanding and responsive, were the most su ccessful in protecting their children from maladjustment (Baumrind, 1991). For this reas on, authoritative parent s are assertive and their disciplinary methods are supportive rather than puniti ve, providing expressive and instrumental parental su pport to their children. In the same way, social learning theo rists (Akers, 1998; Ardelt & Day, 2002) emphasize the importance of parental practices in the process to establish deviant or conforming behaviors by differential associ ations. For example, McCord (1991a; 1991b) found through longitudinal research that aggres sive models promote criminality and that maternal support can reduce the probability th at a son will imitate a criminal father. However, the socializing behavior of parents or guardians may be re ciprocally influenced by childrens antisocial acts due to the dynamic nature of the learning environment (Dembo, Grandon, La Voie, Schmeidler, & Bu rgos, 1986; Akers, 1998; Burdzovic, & OFarrell, 2007; Blazei, Iacono & Mcgue 2008; Farrington, Co id, & Murray 2009). In terms of social learning theory, pare ntal practices influe nce the adolescents thinking process to elaborate de finitions and maintain their differential association toward criminal conduct. The development of those cognitive schemas helps to incorporate information that defines the adolescents id entity. Recently some researchers (Berzonsky, 2004; Smits et al., 2008) have focused on the so cial-cognitive processi ng orientation used by adolescents to derivate different identi ty statuses. Berzonsky (2004) indicates that
23 identity processing style refers to self-reported differen ces in how adolescents process self-relevant information and negotiate identity issues to make decisions. According to Berzonsky (1989), there ar e three styles of processing identity relevant information: Informational, nor mative, and diffuse-avoidant. First, the informational identity processing style actively l ooks for and assesses self-relevant information. Adolescents with this cognitive style are self-reflective, conscientious, open to experience, problem-focused, and vigilant decision makers. Second, persons who use a normative processing style depend more automatica lly on the expectations of significant others. Youngsters characterized by being normative are highly structured and closed to information that might conflict with their pe rsonal beliefs and valu es. Third, adolescents who utilize a diffuseavoidant identity style procrastinate and attempt to avoid solving identity issues as long as po ssible; their behavior is dete rmined mainly by situational factors and hedonistic cues. Research on parental au thority (Berzonsky, 2004; Smits et al., 2008) has found that authoritative parenting is associated with an informational identity processing style, while authoritarian parenting is related to the devel opment of a normative identity style. On the other hand, permissive parents may foster a diffuse -avoidant identity in their children. Those findings suggest that parental authority and supervis ion are involved in the development of adolescent thinking pro cesses and the way in which they build unfavorable definitions toward deviant beha viors. According to Smits et al. (2008), diffuse-avoidant identity st yle and permissive parenting are more related to maladjustment and favorable attitudes to antiso cial behavior than ot her identity styles.
24 Cullen (1994) argued that r estrictive parenting is most effective when parents provide emotional support to their children. Children are more likely to obey parents who have provided parental support. Following th is idea, Wright and Cullen (2001) proposed the term Parental Efficacy to illustrate how parental support and control are intertwined and form an important basis of parenting skil ls to keep children out of trouble. The research findings confirm the inverse relatio nship between parental efficacy (support and control) and delinquency. Also, the results suggest that the influence of the family context on delinquency is beyond pa rental control practices and that the theory of crime that focuses only on control and not on suppor t is likely to be misspecified (Wright & Cullen, 2001; Perrone et al., 2004). In sum, for social learning theorists (Akers & Jensen, 2006; Ardelt & Day, 2002), parents become prosocial models to their children when they provide expressive or instrumental support. On one hand, effective parental support and control fosters the opportunity to learn favorable law-abiding definitions ba sed on the balance between punishments and rewards and differential associ ations. On the other hand, inconsistent and erratic supervision, authorita rian disciplinary practices, a nd lack of parental support are likely to promote deviant attit udes and definitions among adolescents. Parental Support and Peer Influence Barnes and Farrell (1992) have pointed out that parental su pport and parental control play an important role in the pr evention of delinquency. However, although the researchers have found enough evidence that th e lack of parental support influences criminal behavior, they also have discove red that deviant peers constitute a strong
25 predictor of antisocial behavior s and often interact with the effects of parenting practices (Barnes & Farrell, 1992; Hawkins et al., 1992; Farrell, Barnes, & Banerjee, 1995; Wright & Cullen, 2001; Ardelt & Da y, 2002; Perrone et al., 2004) The debate between parental and peer in fluence on antisocial behavior has been essential in the criminological literature. Ba sed on the social suppor t perspective, Wright and Cullen (2001) assessed the Group Social ization Theory proposed by Harris (1995) who states that socialization is context-specific and that outside-the-home socialization takes place in the peer groups of childhood and adolescence. Therefore, Harris (1995) concludes that parents do not have any important long-term effects on the development of their childs personality. Wright and Culle n (2001) found that contrary to Harris contention, parental efficacy appears to be capable of limiting delinquent involvement. However, the data provide partial suppor t for Harris proposition about peer group influence. Similarly, Perrone et al. (2004), confirmed these findi ngs and noted that although deviant peers show a solid effect on delinque ncy, this influence did not mediate the relationship between parental e fficacy and antisocial behavior. In accordance with social learning theory, conforming definitions prom oted by parental support can counterbalance definitions coming from peers in the process of differential association, which favor the violation of law (Liska, 1973; Warr & Sta fford, 1991; Agnew, 1991; Warr, 2002; 2005; Haynie, 2002; Rebellon, 2006). However, social learning th eorists have been criticized due to the temporal order suggested by the theoretical relationship between peer association and delinquency. Hirschi (1969) argued that peer association plays a less importa nt role in the explanation
26 of crime because delinquency occurs before the association with delinquent peers. Once an adolescent becomes delinquent, he chooses delinquent peers. As part of this discussion, some researchers have connected social lear ning theory with the opposite point of view. That is, rather than de linquency causing delinque nt peers, delinquent associations cause delinquency (Hirschi, 1969; Thornberry et al., 1994; Catalano et al., 1996). According to Akers (1996; 1998), this analysis represents a misinterpretation because the theory does not state a unidir ectional relationship among these variables. Social learning theory is able to explain th e reciprocal relationshi p of delinquent peers and delinquency because they are part of the same learning process at different stages (Matsueda, 1988; Matsueda & Anderson, 1998). Likewise, attachment to parents and peer relationships are expressions of the ordinary developmental processes that ta ke place through adolescence (Warr, 2002). During early childhood, attachment to parents pl ays an important role in childrens life. However, peer relations become essential fo r adolescents who are l ooking for a sense of identity and social approval. Consistent with the social learning explanation of a dolescent deviant behavior, early drinking and deviant peer affiliation ha s been associated with subsequent alcohol consumption and antisocial orientation (Haw kins et al., 1992; Parker, Levin, & Harford, 1996). The relevance of deviant peers for adol escents later involvement in delinquency suggests that youngsters follow the behavioral examples (prosocial or antisocial) of significant others (Haynie, 2002). In the sa me vein, Henry, Slat er, and Oetting (2005), found that the overall number of deviant friends predicted studen ts antisocial behavior in early adolescence.
27 Finally, research findings have demonstrat ed the competing influence of deviant peers and parenting practices on adolescent antisocial behavior (Henry et al., 2005; French & Maclean, 2006; Chawla et al, 2008; Wright & Cullen, 2001; Haynie, 2002; Perrone et al., 2004). Although, there is evid ence that parental support and effective supervision may counteract the effects of deviant peers in early childhood, the impact of the youth subculture may be stronge r within the adolescence period. The School Context and Differential Social Support Antisocial behavior is an expression of human devel opment that is socially disruptive and undesirable at different levels of social life. In the view of criminology, social ecology models conceptualize human re lations ordered into different levels of organization from the individual through linkage s to larger social networks (Catalano, 1979; Sampson, Raudenbush, & Earls, 1997; Wikstrom, & Sampson, 2003). According to this perspective, families are embedded in communities and social institutions that reflect the cultural values of society (Coreil, Bryant, & Henderson, 2001). In the development of social support th eory, Wright, Cullen and Miller (2001) drew the concept of family so cial capital from Coleman (1990), using the principle of social support as a link between family pro cess (parent-child rela tionship) and social structure (family social cond itions). According to Colema n (1990), social capital is defined as the set of resources that inhere in family rela tions and in community social organization and that are useful for the cogni tive or social development of a child or young person (p. 300). Later, Lin (2001) ope rationalized this conceptual definition stating that social capital is an investment in social relations by individuals through
28 which they gain access to embedded res ources to enhance expected returns of instrumental or expressive actions (p. 17). Similarly, Wright, Caspi, Moffitt, and Silva (2001) indicated that social investment operates through factors identified by social learning th eory. In accordance with the social learning perspec tive, there is research evidence that family social capital fosters prosocial behavioral patterns in chil dren and insulates them from the adverse effects of deviant peers and delinquent involveme nt associated with several family social conditions (Matsueda & Anderson, 1998; Warr, 2002, 2005; Wright & Cullen, 2001; Perrone et al., 2004; Jones et al., 2007). To advance in the understanding of the ro le played by structural (macro) and processual (micro) factors in the causa tion of crime, Akers (1998) proposed the Social Structure and Social Learning Model (SSSL), a cross-level theoretical model in which social structure influences the social psyc hological process for explaining the origin of criminal behavior and crime rates. Akers (1998) identified four main dimensions of social structure that are expected to be associated with social process and i ndividual behavior. The first two dimensions refer to social structural and socio-demogr aphics correlates, which indicate societal aspects of communities, culture, and social institutions, as well as the distribution of the population related to crime rates. The third dimension emphasizes conceptually defined features of sociological theories to explai n criminogenic conditions of societies. Finally, the fourth dimension designates the differentia l social location in primary, secondary, and reference groups.
29 Consistent with the ecological persp ective (Bronfenbre nner, 1979; Catalano, 1979), the fourth dimension of the SSSL mode l comprises the small groups and personal networks that impact directly on adolescent development th rough the interactive patterns of mesosystems (Akers, 1998). According to Bronfenbrenner (1979) the mesosystem is defined as the interconnectedness of multiple immediate settings (Microsystems) in which the developing person actually participate. Therefore, families, peers, and schools constitute a meso-system, providing a meso-level of analysis. These are the social groups to which the in dividual relates and which offer the learning environments and opportunities that promote or discourage criminal or conforming beha vior (Akers, 1998). The mesosystem includes the immediate soci al context in which social structural and sociodemographic dimensions of the SSSL model impact on individual behavior and the operation of the social lear ning variables. On the other hand, from the perspective of the individual, the mesosystem is closely linke d to the concept of differential association, which is intertwined with other social l earning variables (Akers, 1998) Depending on the nature of the peer social networks in the school context and the parental support received as the expression of the family process, the social learning environment will provide the opportunity to lear n prosocial or antisoc ial behavior (Warr, 1993; Jessor, 1993; White, Loeber, Stouthame rLoeber, & Farrington, 1999; Warr, 2002; Farrington, Coid & Murray, 2009). At the same time, the developmental changes occurring among adolescents as th ey grow older create a stage of transition from parental influence to peer influence (Wright, 1995).
30 Additionally, family structur e and racial minority groups are indicative of social conditions related to the quality of family social capital within the mesosystem. The results about the influence of family structur e on parental socialization practices revealed that adolescents in single-parent families are significantly more delinquent than their counterparts residing with two biologi cal parents (Demuth & Brown, 2004; Mack, Leiber, Featherstone, & Monserud, 2007). Li kewise, Brannigan, Gemmell, Pevalin, and Wade (2002) established that contextual and proce ssual family indicators contributed significantly to aggression in chil dren aged 4 to 11 years old. In the same vein, Matsueda and Heim er (1987) and Rebellon (2002) obtained evidence suggesting that adolescents who live in broken homes are more likely to choose deviant peers and express att itudes favorable to delinquenc y. Equally, Apel and Kaukinen (2008) confirmed that youth who reside with a single biological parent who cohabits with a non-biological partner e xhibit unusually high rates of antisocial behavior. On the other hand, racial minorities such as Hispanics and African Americans have been associated with high levels of antisocial behavior when they are compared with other ethnic groups. For example, Matsue da and Heimer (1987) demonstrated that family disruption has a larger impact on delinquency among African Americans than non-African Americans. Similar results have been found among Hi spanic adolescents. Valdez, Yin, and Kaplan (1997) verified that Hispanic youth were more likely to be arrested for aggressive crimes than either African Americans or Caucasians. At the same time, evidence indicates that violent Mexi can-American youth reside in neighborhoods characterized by high rates of underemployment, single-parent families, welfare reci pients, and teenage
31 parents. These adolescents tend to develop emotional re actions and norms that are adaptations to the social conditions of th ese communities (Valdez, Kaplan, & Codina, 2000). In addition, researchers on the Hispan ic population have proposed that acculturation may be related to high levels of drug use and antisocial behavior among Hispanic adolescents. Acculturation is the pro cess of assimilation into roles and norms of another culture. Miller et al. (2008) confirmed that acculturation, particularly when operationalized as language use, is related to greater drug use and other risky behavior in a sample of Puerto-Rican youngsters. Differential Social Support Based on the social learning perspec tive (Sutherland, 1961; Akers, 1973), Cullen (1994) conceptualized differential social support as the balance between the social support received for crime and the soci al support received for conformity. In this manner, social support coming from law-ab iding sources may foster conformity. Conversely, a variety of antisoc ial behaviors could be promoted when social support comes from illegitimate sources. To illustrate the effects of differential soci al support, Cullen (1994) indicated that antisocial parents may provide knowledge, skills and role models that promote success in delinquent activities through differential opportunity (Clo ward & Ohlin, 1960). Thus, illegitimate social support allows indi viduals to accumulate criminal capital. According to Hagan and McCarthy (1997) cr iminal capital involves the information, technical skills, social ne tworks, and resources necessary for success in criminal
32 enterprises. On the contrary, law-abiding s ources of social support allow the development of social capital which includes knowle dge, a sense of obligation, expectations, trustworthiness, information channels, norms, an d sanctions embodied in the social life of a community (Coleman, 1990). Consequently, pa rental support for conformity could not only be a protective factor against delinquency, but posit ive modeling for prosocial behavior (Sutherland, 1961; Akers, 1973). The sources of differential social suppor t (Cullen, 1994) are mainly associated with the social conditions of the youth. Fo r instance, families and schools may provide social support to prevent antisocial behavior in youngsters, while peer groups promote social support for deviance. As a result, antisocial behavior will be less likel y when social support for conformity exceeds social support for deviance (Cullen, 1994). Founded on the social learning perspectiv e, Cullen (1994: 544) emphasized that social support is likely to be effective when it is linked to conformity-inducing outcomes. As stated earlier, support from conformist sources may not only address antisocial risk factors, but also provide an opportunity for prosocial modeling. Within the school context, the link to conformity-indu cing outcomes could be represented by the protective sense of belonging provided by the influence of school connectedness among adolescents. In terms of social lear ning theory, school connectedness offers the opportunity to learn law-abiding de finitions from prosocial models. Conversely, support from deviant friends may promote antisocial behavior if these associations also expose youths to crim inal influences. In th e school context, the support received from deviant friends corre sponds to the exposure to peer drinking
33 groups. In terms of the social learning pe rspective, peer drinking groups provide the opportunity for learning prodelinquent definitions from deviant models. Social Support for Conform ity: School Connectedness Research results have demonstrated th at parental support is strengthened when meso-level variables are taken into account such as school and community membership (Sampson, Morenoff & Earls, 1999; Sampson, 2006b; Vieno, Nation, Perkins, & Santinello, 2007). According to ecological theory, adolescent s daily activities can be seen as a relevant index in their devel opmental process (Bronfenbrenner, 1979). The activities in which adolescen ts participate within and outside the schools provide opportunities for learning and practicing co mpetencies and skills (Larson & Verma, 1999) and receiving differential social supports that can buffer the effects of family conflicts (Rossman & Rea, 2005). In this regard, school conn ectedness is a source of soci al support from school personnel, which in turn, increases the sense of attachment, commitment, and involvement in the school environment. In general terms, school connectedness reflects the adolescents belief that a dults in the schools care abou t them as an individual and provide support for learning, positive adult-student relati onships, and physical and emotional safety (Thomas & Smith, 2004; Wilson, 2004). The research reveals that a high degree of school connectedness and improved academic achievement reduces delinquency rates and health-compromisi ng outcomes (Jenkins, 1997; Welsh, Green, & Jenkins, 1999; Wilcox & Clayton, 2001; Stewart, 2003; Wilson, 2004; Payne; 2008). When students feel connected to their school, they may be more likely to trust in teachers
34 about violence exposure, resul ting in better coping skills a nd decreased violent behavior. (Ozer, 2005; Brookmeyer, Fanti, & Henr ich, 2006; Eaton et al., 2004) Along the same line, school connectedness involves students participation in community and civic activities (Vieno et al., 2007). Accord ing to Youniss and Yates, (1997), adolescents involvement in volunteer se rvices or participat ion in faith-based activities puts youngsters in c ontact with people in need a nd with positive role models. Consequently, the exposure to prosocial networks and school satisfaction promote favorable attitudes for social adjustment and decrease the oppor tunities for antisocial peer affiliations and delinquent definitions, crea ting supportive contexts for adolescents wellbeing (Jang & Johnson, 2001; Vieno et al ., 2007; Kaufmann, Wyman, Forbes-Jones, & Barry, 2007; McGrath, Brennan, Dolan, & Barnett, 2009). As stated earlier, perceived school connect edness, as source of social support at the meso-level, moderates the influence of parental support on antisocial behavior (Simons, Simons, Burt, Brody, & Cutrona, 2005; Stewart, 2003; Dixon, 2008; McGrath et al., 2009). Vieno et al. (2007) found that exposure to prosocial networks and school satisfaction may enhance the effects of pare ntal support on adolescen ts deviant behavior by providing a structure for increased superv ision and developing ad aptive interaction with adult and peers. Conve rsely, exposure to deviant pe er networks across schools constitutes a source of social support that promotes antisocial behavior at the meso-level because it involves the differential social lo cation in reference groups within the SSSL model (Akers, 1998). In addition, this source of social support for deviance is frequently associated with poor parent-child relations hips (Warner & Wilcox; 1997; Brookmeyer et al., 2006; Dixon, 2008; McGr ath et al., 2009).
35 Social Support for Deviance: Peer Drinking Groups Researchers have indicated that devian t peers constitute a strong predictor of antisocial behavior that often interacts with the effects of parental practices (Barnes & Farrell, 1992; Hawkins, et al., 1992; Wright & Cullen, 2001; Ardelt & Day, 2002; Perrone et al., 2004). At the micro-level, a dolescents learn definitions favorable or unfavorable toward delinquenc y from parents, peers, and schools (Akers, 1992; Sellers & Winfree, 1990; Sellers et al., 1993; Miller et al., 2008). However, families, peers, and schools are organized in mesosystems that ar e contained within socio-cultural contexts. Thus, at the meso level of analysis, the differential association process for learning delinquent behavior is influenced by defin itions shared within social groups and communities through the communication process, providing social identity, membership, and a sense of belonging (Winfree, T., Back strom, & Mays, 1994; Mears, Ploeger, & Warr, 1998; Lanza-Kaduce & Capece, 2003; Samps on, 2006a). According to Krohn, Lanza-Kaduce, a nd Akers (1984), high levels of peer drinking groups within communities were a ssociated with high levels of underage drinking across several social groups. These resu lts indicate that schoo l context, such as pro-alcohol networks and cultural traditions, have a strong effect on adolescent alcohol consumption levels. At the same time, drinking alcohol, for adolescents, may have two main cultural meanings: 1) Symbol of adult status, imitating parental drinking; 2) Symbol of adolescent rebellion, rejecting parental authority or expectations (A kers, 1992; Sellers & Winfree, 1990). Likewise, adolescents affiliation w ith peer drinking groups may provide
36 opportunities for learning prodelinquent definitions and antisocial behaviors (Akers, 1992; White et al., 2002; Felson et al., 2008). Furthermore, as youth enter adolescence, peer affiliation becomes a much stronger influence. Ingoldsby et al. (2006) found that high levels of pa rental conflicts in early childhood may result in high er levels of conduct problems, which in turn are related to involvement with deviant peers groups across middle childhood. Additionally, this relationship is maximized across disadvant aged neighborhoods. T hus, youth with poor parental support are more lik ely to integrate peer drinki ng groups (Ingram, Patchin, Huebner, McCluskey, & Bynum, 2007; Wells & Graham, 2003) Youth subculture is shaped by develo pmental changes where adolescents are looking for autonomy and identity. In this ma nner, social approval of significant peers constitutes a powerful factor at the meso leve l that influences adol escent drug abuse and deviant behavior. For instance, McIntosh, F itch, Branton, and Nyberg (1981) encountered that those adolescents attached to conv entional peers tend to disapprove alcohol consumption. Conversely, those who were attach ed to deviant friends are more inclined to drink alcohol. According to Haynie (2002), when beha vioral patterns, such as alcohol consumption, are reinforced by members of the drinking groups, the friendship network will better be able to create confidence, establish expectations, and reinforce social norms that are favorable to deviant behavior. Most importantly, the consensus about the appropriateness of drinking beha vior provides a sense of att achment to deviant peers and detachment from parents and prosocia l networks (Krohn, 1986; Haynie; 2002).
37 Chapter Three Methods As the foregoing review of the literature has shown, the theoretical relationship of interest is that between pa rental support and antisocial behavior. However, previous research also suggests that th is relationship may be partiall y mediated by social learning mechanisms. Figure 1 depicts these possible theoretical linkages at the individual or micro level. It is anticipated that parental support will retain a significant association with antisocial behavior even after social learning variables ar e incorporated into the model. Figure 1: Research Model Stage 1 Social Learning Emotional Reinforcement Social Reinforcement Differential Reinforcement Neutralizing Definitions Parental Support Control Variables Gender Ethnicity Intact Family Perceived Supervision Micro Level: Parental Support Antisocial Behavior
38 Additionally, it has b een acknowledged that the pa rental support-antisocial behavior relationship may vary depending on the context within which this relationship operates. As a result, a second theoretical model will be estimated that assesses the degree to which social support at the sc hool level moderates (Baron & Kenny, 1986) the relationship between parental support and antisocial behavior at the individual level. Figure 2 graphically depicts this multi-level model. Figure 2: Multilevel Research Model Stage 2 Social Learning Emotional Reinforcement Social Reinforcement Differential Reinforcement Neutralizing Definitions Parental Support Control Variables Gender Ethnicity Intact Family Perceived Supervision Micro Level: Parental Support Antisocial Behavior Meso Level: School Context Differential Social Support School Connectedness Peer Drinking Groups
39 Hypotheses Based on the multilevel model (Figure 2), the current study tests the following research hypotheses: Hypothesis 1: Parental support will have a di rect negative effect on antisocial behavior among adolescents. Hypothesis 2: Social learning mechanisms will partially mediate the effects of parental support on antisocia l behavior among adolescents. Hypothesis 3: There is a significant interac tion between the school support variables and parental su pport in their influence on antisocial behavior among adolescents. Data Collection and Sample The data used in this study were coll ected from the Sarasota Demonstration Project, which is a joint effort between the Florida Prevention Research Center at USF, the Sarasota County Health Department, a nd the local community advisory board. The purpose of this project was preventing the initiation of smoking and alcohol use among middle school students in Sarasota County. As pa rt of the research strategy, the Florida Prevention Research Center developed the Youth Tobacco and Alcohol Use Survey, administered to a representative sample of sixth through eleven graders during spring 2000. The database includes 113 classes repr esenting 2,004 students. The proportions of boys and girls in the sample were approxima tely the same (51 % boys; 49 % girls). To handle missing data on variables used in the analysis, 460 cases were removed from the
40 original database. The total sample size fo r the analyses presented her was thus 1,544 students enrolled in sixteen schools. Measures of Variables The variables used in the analyses are lis ted in Appendix A. These tables contain descriptive statistics (mea n, standard deviation, and sc ore range) for dependent, independent, intervening, and control variables. All research variables were examined through factor analysis using principal components analysis as the ex traction method, and promax as the rotation method. At the same time, the procedure used to estimate the reliability was the calculation of Cronbachs alpha as an indicator of inte rnal consistency (Carmines & Zeller, 1979; Bachman & Shutt, 2003). This study assesses the influence of pare ntal support on antis ocial behavior. The dependent variable, antisocial behavior, is operationalized as the behavioral pattern characterized by aggressive behavior and dr ug use during the year prior to the survey. The students answered a ten-item scale descri bing their experience with aggression, drug use, and alcohol consumption in th e past twelve months (Appendix B). Subsequently, a scale was built by summing the answers to each item and creating an index from 0 to 10, where 10 indicates high levels of antisocial behavior (Alpha=0.79). Furthermore, the factor analysis revealed the existence of three components, aggressive behavior, drug use, a nd alcohol use. Howeve r, when the factor analysis was set to produce a single solution, the results suggested (Appendix B) that this
41 group of items is measuring the same unidi mensional latent construct (Antisocial Behavior). Concerning the independent variable, pare ntal support suggests behavioral patterns from parents that provide love, nurturance, empathy, acceptance, guidance, information, and material resources to their children. The general index of parental support was a 10item scale, based on the ability of parents to provide guidance and assistance to their children (Appendix C). In addition, given past research indicating di fferential effects for maternal and paternal support, two different indexes were also created to measure separately maternal (Appendi x D) and paternal support (Appe ndix E). The factor analysis revealed two different components correspond ing to mothers and fathers. Also, the highest score obtained by these scales indicate strong parental support. In the context of this study, parental support was defined as the adolescent perception of the parents ability for delivering guidan ce and assistance to their children by providing effective communication pattern s and helpful information about alcohol consumption (Alpha= 0.82). The additional scal es assess the adolescent perception of the mothers (Alpha= 0.78) and the fathers abil ity (Alpha=0.82) to pr ovide support to their children. In accordance with social learning th eory, parental support influences communication patterns that promote differential associa tion (Akers, 1998). However, this learning process is mainly mediated by emotional and social reinforcement, differential reinforcement, and neutralizing definitions toward deviant behavior (i.e. underage drinking). These social learning m echanisms are anticipated to serve as mediating variables that are a ffected by parental support, a nd in turn affect antisocial
42 behavior (Figure 1). Once the factor analys is was performed, four separate social learning variables were identified within the research model. Emotional Reinforcement is operationalized as adoles cents beliefs about drinking alcohol as emotional reward to handle depre ssion and anger. The factor analysis for the set of four items indicated a unique component. The highest score in th is scale suggest an unfavorable emotional reinfo rcement toward alcohol use ( Appendix F, Alpha= 0.86). Social Reinforcement is measured as adolescents beliefs about alcohol consumption as social reward to improve thei r social image among their peers. The factor analysis of a five-item scale revealed one solution. Higher scores denote unfavorable social reinforcement toward alc ohol use (Appendix G, Alpha= 0.80). Likewise, the process of Differential Reinforcement is identified as the selfperceived balance of rewards and costs unfavorable to dri nking behavior among adolescents, considering peer influence on a dolescents alcohol use, health outcomes, and legal consequences of drinking alcohol (A ppendix H, Alpha= 0.71). The lowest score in this eight-item scale points out an unfavorable differential reinforcement toward alcohol consumption. In addition, the f actor analysis revealed thr ee components using promax as the rotation method. However, the data were al so fitted to a one-factor solution indicating the estimation of one single construct. Finally, the Neutralizing Definitions are specified by adolescents beliefs that justify underage drinking as a social activity a pproved by society (Appe ndix I, Alpha= 0.87). The factor analysis produced one single component, where higher scores indicated unfavorable definitions toward deviant behavior.
43 In line with social learning theory (Akers, 1992; 1998), the balance of rewards (social and emotional) and cost s (differential reinforcement unfavorable to alcohol use) decreases the probability of deviant behavior among a dolescents. Then, neutralizing definitions are weakened and unfavorable definitions are strengthened through the differential association process. Therefore, unfavorable reinforcement and definitions toward deviance are expected to have an i nverse relationship with antisocial behavior. On the other hand, a group of control vari ables were included in the research model to regulate the well know eff ect of demographic variables on deviant behavior such as gender, and ethnicity (e.g. Marshal & Chassin, 2000; Vald ez, Kaplan, & Codina, 2000). At the same time, other control variables a ssociated with the pare nt-child relationship was considered: perceived supervision, a nd intact family (e.g. Ardelt & Day, 2002; Demuth & Brown, 2004; Hawkins et al., 1992) In relation to demographic variables, gender was estimated through a dichotomous variable (0=female, 1=male). Equally, the variable ethnicity was examined through a set of dummy variables designa ting four ethnic groups: Ca ucasian (0=non-Caucasian, 1=Caucasian), African American (0=non-Af rican American, 1=African American), Hispanic (0=non-Hispanic, 1=Hispanic), and other ethnic groups (0 =non-other; 1=other ethnic groups). Caucasians was selected as the excluded category against which all other groups are compared. Regarding the factors associated with th e parent-child relationship, intact family was also assessed using a dichotomous variable (0= Live with both parents, 1= Do not live with both parents). Finally, perceived supe rvision is an index c oded 0 to 4, indicated by answers to the following question: How often do your parents/guardians let you
44 make your own decisions about the time you must be home on weekend nights? Higher scores on this variable suggest an inconsistent supervision. In addition to the individual-level variables described above, the multi-level model also includes the concep t of support at the school level th at moderates the influence of parental support on antisocial behavior. In order to assess the meso-social dimension of differential social support, two additional vari ables at level 2 were incorporated within the multilevel model: School connectedness and p eer drinking groups (Figure 2). In line with the social support paradigm (Cullen, 1994), high levels of perception of school connectedness and low levels of peer dr inking groups across schools provide social support for prosocial behavior. Conversely, low levels of school connectedness and high levels of peer drinking groups across school s provide social support for antisocial behavior. The effects of differential social s upport at the school le vel are anticipated to moderate the influence of differential soci al support at the micro-level provided by parents and peers. As discussed in Chapter 2, school connect edness is a source of social support for prosocial behavior that invol ves school attachment, stud ent satisfaction with their academic performance, and school and community participation (e.g. Thomas & Smith, 2004; Wilson; 2004; Payne; 2008; Vieno et al., 2007). In this regard, school connectedness has been operationalized as a se t of adolescents belie fs that provide a sense of personal belonging to the school environment. This va riable is measured through a composite index comprise of the following in dicators: 1) adolescents belief that adults in the school care about them as individuals providing support for learning, positive adult-student relationship, and physical a nd emotional safety; 2) frequency of
45 adolescents participation in prosocial organizations in the schools and community, and 3) self-perceived academic performance. In order to build a composite index to estimate levels of school connectedness (Appendix J), the responses to indicators of School Attachment (Items 1-3), School and Community Participation (Items 4-7), and Self-perceived Academic Performance (Item 8) were added to create the Sc hool Connectedness Scale (Alpha = 0.68) Later, an average composite index was calculated per school. The highest value indicates strong school connectedness by e ducational institution in Sarasota County. In contrast with the effects provided by school connectedness, peer drinking groups represent the opposite si de. Consistent with social le arning (Akers 1992; 1998) and the social support paradigm (Cullen, 1994), e xposure to peer drinki ng groups provides the opportunity to learn pro-delinquent definitions th at lead to antisocial behavior. Therefore, peer drinking groups can be considered a sour ce of social support for antisocial behavior. Research has demonstrated that pro-alc ohol friendship groups among adolescent males are associated with high levels of underage drinking and often provi de the opportunity to be involved in criminal activ ities (Krohn et al., 1984; Sell ers et al., 1993; White et al, 2002; Felson et al., 2008) With the purpose of estimating the existe nce of peer drinking groups across school, the answers to the question: Does your best friend ever drink alcohol? (0= No, 1= yes) were aggregated by schools indicating the poten tial affiliation with peers who engage in alcohol use. Higher scores on this variable s uggest a high probability of being involved in peer drinking groups. Afterward, the averag e index per school provides an approximation of the potential existence of peer drinki ng groups across educational institutions in
46 Sarasota County. Thus, peer drinking groups is expected to have a positive relationship with antisocial behavior, while school connect edness demonstrates an inverse association with the dependent variable. Methodological Approach According to Tashakkari and Teddlie (1998), multilevel research refers to studies in which data from more than one level of organizations are utilized to reach a more comprehensive understanding of social behavior. In educatio nal research, earlier multilevel studies estimated school-level data through the average of student-level information. Instead of aggregating individual level vari ables, Bryk and Raudenbush (1992) proposed Hierarchical Linear Modeling (HLM), an a dvanced statistical technique for analyzing both student-level and school-level data simulta neously within a single research model. In this manner, HLM examines multiple levels embedded within each other (organizational models) and allows a determin ation of cross-level effects using a single methodology. In accordance with Gottfredson (2001), school level studies that rely only on school level data have shown the following weaknesses: 1) They fail to separate the influence of the compositional context of th e school from the effect of individual processes; 2) They assume a constant eff ect of the student characteristics across all schools; 3) They do not allow for the examina tion of how variables relate to one another within schools; 4) They do not reveal the e ffects of school characte ristics on adolescents. In contrast, multilevel studies produce estimate s of school effects on student behavior and
47 are capable of separating the influence of the demographic composition of the school from individual demographics (Gottfredson, 2001). The main advantage of HLM is that it allo ws separating the individual level from the meso-level variation, assessing the tr ue effects of both levels on the dependent variable. Raudenbush and Bryk (2002) have indicated how HLM solves the most common difficulties in multilevel research, su ch as aggregation bias, and misestimated standard errors when OLS regression is use d. First, HLM helps to solve aggregation bias by facilitating a decomposition of any obs erved relationship between variables (e.g. parental support and peer drinking groups) into separate level-1 a nd level-2 components. Second, HLM also resolves the problem of mi sestimated standard errors in multilevel data by incorporating into the statistical model a unique random effect for each school. However, Kubrin and Weitzer (2003) r ecognized limitations of HLM. Although HLM is an excellent technique for testing cr oss-level effects, it requires a specific structure of the sample and it does not al low the analysis of competing contexts simultaneously (i.e. schools and neighborhoods ) in the same level (Kubrin & Weitzer, 2003). The multilevel research model (Figure 2) in the current study assesses the influence of parental suppor t on antisocial behavior and al cohol use among sixth through eleventh graders. The hierarchical linear models (Raudenbush & Bryk, 2002) require that both individual and school context factor s be evaluated simultaneously using the following statistical equation. At level-1, an HLM equation represents th e effects within schools for antisocial behavior ( ij):
48 ij = 0j + 1jx1ij + 2jx2ij + ... QjxQij + rij Where: ij = Antisocial Behavior 0j = Intercept, mean level of antisocial behavior for school j QjxQij = regression coefficient for the eff ect of individuallevel predictors (xQij) on antisocial behavior ( ij) rij = individual-level model error term. At level-2, an additional HLM equation corresponds to the effects between schools for antisocial behavior ( qj): qj = q0 + q1W1j + q2W2j + ... qSqWSqj + uqj Where: qj = mean level of antisocial behavior for school j, indicating the distributive effects in each school. q0 = intercept, grand mean level of antisocial behavior for all schools. qSqWSqj = regression coefficient for the e ffect of meso-level predictors (WSqj) of individual level slopes and intercepts on mean level of antisocial
49 behavior ( qj), indicating the effect of school characteristic on the distribution of antisocial behavior within each school. uqj = meso-level error term, indicating the unique effect associated with school j. In sum, as indicated by the multilevel research model (Figure 2) the variables included at level-1 were the following: Pa rental support (Materna l support, Paternal support), emotional reinforcement, social re inforcement, differential reinforcement, neutralizing definitions, gender, ethnicity, intact family, and perceived supervision. Likewise, the variables school connectedness an d peer drinking groups were included at level-2 to estimate their mode rating effects over the influe nce of parental support as individual-level predictors of antisocial behavior. Finally, a set of preliminary analyses were performed to assess the normality of the dependent variable, multicollinearity among the predictors, symptoms of heteroscedasticity, and cross-level collinearity.
50 Chapter Four Results Bivariate Analyses The first step to assess the influence of parental support on antisocial behavior was to examine the bivariate relationsh ips among the independent and dependent variables (Table 1, page 51). As expect ed, parental support was significantly and negatively related to antisocial behavior Likewise, the social learning variables demonstrated the same pattern in this association. Neutralizing definitions and differential reinforcement revealed the highest correlation with antisocial behavior, while emotional and social reinforcement indicated a significant moderate relationship. On the other hand, control variables such as gender, intact family and perceived supervision showed a significant and positiv e correlation. Conversely, the category of ethnicity exhibits a weak and non-significant association. Whereas Hispanics presented a weak correlation with antisocial behavior, Af rican American and ot her ethnicities showed non-significant relationshi p to this variable. The same bivariate analysis was perfo rmed with the level-2 variables. As anticipated, peer drinking groups were sign ificantly and positivel y associated with antisocial behavior. In cont rast, school connectedness was significant and negatively related to antisocial behavior. These initial findings sugge st that the proposed concep tual model has validity. However, more rigorous analyses are require d for testing the hypothesis formulated in this study. First, multivariate analyses are necessary to demonstrate that the relationship
51 Table 1: Correlation Matrix among Study Variables Predictors 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1. Antisocial ,17** ,07* 0,01 -0,003 ,13** ,15** -,27** -, 17** -,27** -,29** -,34** -,57** -,49** -,40** ,52** 2. Gender 1 0 -0,03 0,01 -0,04 ,07** ,08** ,049 ,087 ,08** -0,02 -,06* -,08** -,19** -0,003 3. Hispanics 1 -,08** -,09** 0,03 -0,02 -0,01 -,044 -,019 0,01 -0,02 0,002 0,004 -,070** 0,037 4. African Americans 1 -,06* ,13** -0,03 -0,05 -003 -071** -0,01 -0,04 0,02 0,04 -0,01 -0,02 5. Other Ethnics Groups 1 0,01 -0,01 -,096** -,097** -,066** -0,04 0,007 0,02 0,02 -0,003 0,004 6. Intact Family 1 0,02 -,19** -,052* -,236** -, 06* -0,03 -0,03 -0,003 -,097** ,05* 7. Perceived Supervision 1 -0,003 ,026 -,025 -,06* -,09** -,14** -,09** -0,02 ,08** 8. Parental Support 1 ,79** ,87** 21** ,23** ,21** ,23** ,26** -,16** 9. Maternal Support 1 ,38** ,17** ,19** ,14** ,18** ,21** -,11** 10. Paternal Support 1 ,17** ,20** ,20** ,20** ,23** -,16** 11. Emotional Reinforcement 1 ,59** ,32** ,28** ,18** -,22** 12. Social Reinforcement 1 ,43** ,40** ,15** -,30** 13. Neutralizing Definitions 1 ,54** ,30** -,48** 14. Differential Reinforcement 1 ,31** -,37** 15. School Connectedness 1 -,21** 16. Peer Drinking 1 **Correlation is significant at the 0.01 level (2-tailed). Correlation is significant at the 0.05 level (2-tailed).
52 between parental support and antisocial be havior is not spurious. Considering the literature review on parental support and antisocial behavior (i.e. Wright & Cullen, 2001; Perrone et al., 2004; Jones et al., 2007), gende r, ethnicity, intact family, and perceived supervision were included within the multivaria te model as control variables. Second, the multivariate analyses are also required to a ssess the mediating effects of social learning variables on the relationship be tween parental support and an tisocial behavior. Third, if the mediating effects of social learning va riables are demonstrated and significant variation across schools are detected, then HL M analyses will be conducted to assess the moderating effects of differ ential social support (Sc hool Connectedness and Peer Drinking Groups). Parental Support and Antisocial Behavior The main purpose of this study is to eval uate the influence of parental support on antisocial behavior. The results of this anal ysis are presented in Model 1, Table 2 (page 53). Consistent with the hypothesis 1, these fi ndings demonstrated that parental support has a direct negative effect on antisocial behavior, even when controlling for gender, intact family, perceived supervision, and et hnicity. The negative relationship found in Model 1 indicates that higher scores on parent al support were associated with low scores in antisocial behavior. In c ontrast, gender, intact famil y, perceived supervision were significant and positively related to antisocial behavior. These findings were the same across the six models at level-1 (Table 2) and revealed that being male; belonging to disrupted families, and inconsistent superv ision were also strongly associated with
53 Table 2: Model for Antisoc ial Behavior (Level 1) MODEL 1 MODEL 2 MODEL 3 MODEL 4 MODEL 5 MODEL 6 Fixed Effects Coefficient (SE) Coefficient (SE) Coefficient (SE) Coefficient (SE) Coefficient (SE) Coefficient (SE) Intercept 2.41** (0.11) 2.41** (0.10) 2.41** (0.09) 2.42** (0.08) 2.43** (0.05) 2.43** (0.05) Gender 0.86** (0.10) 0.93** (0.09) 0.88** (0.09) 0.72** (0.08) 0.67** (0.08) 0.68** (0.08) Intact Family 0.37** (0.09) 0.36** (0.09) 0.38** (0.08) 0.43** (0.09) 0.41** (0.08) 0.38** (0.08) Perceived Supervision 0.24** (0.05) 0.21** (0.05) 0.19** (0.05) 0.16** (0.04) 0.11** (0.03) 0.11** (0.03) Hispanic 0.46* (0.19) 0.48* (0.20) 0.45** (0.18) 0.50** (0.16) 0.50** (0.15) 0.52** (0.15) African American -0.09 (0.21) -0.07 (0.22) -0.12 (0.22) 0.11 (0.16) 0.17 (0.12) 0.15 (0.12) Other -0.19 (0.22) -0.24 (0.23) -0.18 (0.21) -0.04 (0.18) 0.02 (0.16) 0.03 (0.15) Parental Support -0.10** (0.008) -0.08** (0.008) -0.07** (0.008) -0.05** (0.008) -0.04** (0.007) ---Maternal Support ----------------0.01 (0.02) Paternal Support ----------------0.07** (0.02) Social Learning Emotional Reinforcement ----0.17** (0.02) -0.09** (0.02) -0.08** (0.02) -0.06** (0.02) -0.06** (0.02) Social Reinforcement -------0.13** (0.03) -0.05** (0.02) -0.004 (0.02) -0.005 (0.02) Differential Reinforcement ----------0.21** (0.01) -0.12** (0.01) -0.12** (0.02) Neutralizing Definitions -------------0.26** (0.02) -0.26** (0.02) Random Effects Variance (St. Dev.) Variance (St Dev.) Variance (St Dev.) Variance (St Dev.) Variance (St Dev.) Variance (St Dev.) Intercept (Schools) 0.15** (0.39) 0.11** (0.34) 0.08** (0.29) 0.06** (0.24) 0.01 (0.10) 0.01 (0.10) Level-1 4.44 (2.11) 4.18 (2.04) 4.08 (2. 02) 3.51 (1.87) 3.05 (1.75) 3.04 (1.74) Indicators of Fit -2LL 6719.50 6628.97 6595.78 6365.57 6149.89 6147.73 AIC 6723.50 6632.97 6599.78 6369.57 6153.89 6151.73 BIC 6734.18 6643.65 6610.46 6380.25 6164.57 6162.41 Percent Reduction Parental Support ---20 % 30 % 50 % 60 % ---* (p <0.05) ** (p <0.01)
54 antisocial behavior. On the other hand, from the three ethnicity categories, Hispanics showed a moderate signifi cant relationship with antisoc ial behavior, while African American and other ethnicities were not signi ficantly related to th e dependent variable. Finally, the random intercept i ndicated significant variations of the effect of parental support on antisocial behavior across schools. Parental Support and Social Le arning: Mediation Analyses The mediating effects of social learning mechanisms in the influence of parental support on antisocial behavior were assessed fr om in Models 2-6 (Table 2). In Model 2, emotional reinforcement was included and show ed a significant negative effect. This means that unfavorable emotional reinforcement was associated with the lowest scores of antisocial behavior. When emotional reinforcement was included within the model, the significant influence of parental support on antisocial behavior was reduced by 20 % in comparison with Model 1. This finding indicates that greater levels of parental support increased unfavorable emotional reinforcement, which then decreased antisocial behavior. Furthermore, these results suggest that parental support had both direct and indirect effects to the depe ndent variable, considering the mediating influence of emotional reinforcement. Also, the random intercept was significant in Model 2, which demonstrated the existence of variations of those effects across schools. Subsequently, social reinforcement was included within the Model 3 (Table 2). Similar to Model 2, negative significant eff ects were found, suggesti ng that unfavorable social reinforcement toward deviance was rela ted to lower levels of antisocial behavior. The significant effect of parental suppor t on antisocial behavior was reduced by 30 %
55 when emotional and social reinforcement were included in Model 3. As a result, higher levels of parental support st rengthen unfavorable emotional and social reinforcement, diminishing levels of antisocial behavior. In addition, the mediating effects of emotional and social reinforcement indicate that parental support had both direct and indirect effect on antisocial behavior. As in Model 2, the si gnificant random interc ept obtained in Model 3 revealed that the effects of parental s upport and emotional and social reinforcement vary across schools, indicating that the in fluence of the school context played an important role in the explanat ion of those effects. Next, differential reinforcement was included within the Model 4 (Table 2). The results suggested that emotiona l, social, and differential reinforcement were significant and negatively related to antis ocial behavior. Likewise, the ne gative significant effect of parental support on antisocial behavior was reduced by 50 % wh en emotional, social, and differential reinforcement were included in Model 4. Thus, greater levels of parental support intensified unfavorable emotional, social, and di fferential reinforcement, decreasing the levels of antisocial behavior Similar to Models 2 and 3, these findings suggest that parental support also presented an indirect influence th rough the effect social learning mechanisms. The significant random interc ept indicate that influence of parental support together with emotional, social, and differential reinforcement vary depending on the school context. Finally, neutralizing definitions was inco rporated with the three reinforcement variables in Model 5 (Table 2). The findings for this model revealed that emotional reinforcement, differential reinforcement, and neutralizing definitions were significant and negatively associated with antisocial behavior. However, the effect of social
56 reinforcement became non-significant when ne utralizing definitions was added in the model. This result means that social reinfo rcement turned out to be irrelevant when students presented beliefs that did not just ify deviant behaviors. Additionally, the significant influence of parental support on antisocial behavior was reduced by 60 % when the complete set of social learning m echanisms were incorporated within Model 5. From the four social learning variables evaluated in this st udy, only three demonstrated to have mediating effects over the influence of parental support on antisocial behavior. Therefore, higher levels of parental support increased unfavorable emotional and differential reinforcement, as well as unfavor able neutralizing definitions toward alcohol use, lessening the levels of antisocial beha vior. Similar to Models 2, 3, and 4, parental support demonstrated to have both direct and indirect infl uence on antisocial behavior through these three social learning mechanis ms (emotional reinforcement, differential reinforcement, and neutralizing definitions ). However, the non-significant results obtained for the random intercept in Model 5 indicated that the school context did not play an important role in the process of learning neutralizing definitions toward alcohol use. This finding suggests that neutralizi ng definitions are learned in a process of differential association dire ctly from parents and peers at the student level. Because past research suggests that maternal and paternal support may have differential effects on antisocial behavior, Model 6 (Table 2) was estimated using separate measures of paternal and maternal support. The results indicated that paternal support exhibited a negative significant eff ect on antisocial behavior, while maternal support showed non-significant outcomes in th e full model. However, in analyses not shown both maternal and paternal support re vealed significant influence on antisocial
57 behavior when the reinforcement variables (M odel 2 to 4) were added in the model. Regarding social learning va riables and the random intercep t, the findings obtained for Model 6 demonstrated the same pattern as that in Model 5. On the other hand, to estimate the comp lexity of each resear ch model a set of indices of fit (Dedrick et al., 2009) were calculated: 1) The deviance of the model was indicated by the log likelihood (-2LL); 2) Th e Akaikes Information Criterion (AIC) was calculated using the following formula: AI C = -2LL + 2p, where p is the number of estimated parameters; 3)The Schwartz Baye sian Information Criterion (BIC) was given by the formula: BIC = -2LL + p ln (N), where N is the sample size at level 1(Table 2). According to Dedrick et al. (2009), valu es closer to zero in these indicators represent a better fit of the model. Therefore, the decreasing patterns of the indices of fit across the six models indicate th at Models 5 and 6 were the models that better explained the complex relationships among the variables. Unconditional Model With the purpose of assessing whether HLM is necessary, an an alysis of variance (ANOVA) was performed to verify the exis tence of variation between schools. The unconditional model for antisocial behavior as dependent variable with no predictors was computed by the software HLM 6, according to the following equation: Level 1: = 0 + r Level 2: 0 = 00 + u0
58 The output for the unconditional model has been summarized in Table 3. These results are significant and indi cate that Antisocial Beha vior varies across schools, showing that there was significant variati on at-level 2. In ad dition, the Intraclass Correlation Coefficient (ICC) indi cates that 4.68 % of the variat ion of antisocial behavior occurs across the sixteen schools of the samp le (Level 2); while the remaining 95.32 % of the variation relies on the student level (Level 1). These findings suggest the importance of the student level in the influence of parental suppor t on antisocial behavior. Likewise, there is evidence of possible contextual eff ects across schools that may contribute to a better understanding of this relationship. Table 3: Unconditional Model: One-way ANOVA Fixed Effects Coefficient SE Aprox. d.f. P-value Average School Mean, B00 2.42 0.14 15 0.000 Random Effects Variance Component d.f. X2 P-value Between Schools Intercept (Schools) 0.25 (4.68 %) 15 89.77 0.000 Within Schools Level-1 5.08 (95.32 %) Differential Social Support: HLM Analyses With the purpose of analyzing the modera ting effects of level-2 variables over the influence of parental suppor t on antisocial behavior, severa l steps were required to accomplish the multilevel stage of the research model (Figure 2, page 38). First, the full model was reduced excluding non-significant variables (African American, Other
59 ethnicities, and Social Reinforcement) in order to preserve the statis tical stability in the presence of level-2 variables. A small vari ation in the coefficients of the remaining variables when these three variables were excl uded indicates that th ese exclusions did not mispecify the model. Second, another important methodological consideration for performing an HLM analyses was centering the vari ables in the research model. According to Kreft and De Leeuw (1998), it is important to center the va riables to reduce cro ss-level collinearity. All predictors were centered around their grand mean, and th e slopes of gender, intact family, perceived supervision, and Hispanics were fixed. Only the intercept, maternal, and paternal support were allowed to vary. In Table 4 (page 60), gender, intact family, perceived supervision, and Hispanics were significantly and positively related to antisocial behavior. These results were the same from Model 7 to 8 (Table 4) and suggested that being male, being Hispanic belonging to disrupted families, and perceiving inconsistent supervis ion were strongly associated with antisocial behavior. In Model 7, the parental support inde x was included only with the control variables. Parental support demonstrated si gnificant and negative effects on antisocial behavior, indicating that when parental support increased an tisocial behavior decreased. On the other hand, the significant finding of the random effects obtained for the intercept indicated that there was a variation of antisocial behavior between schools. Next, emotional reinforcement, differen tial reinforcement, and neutralizing definitions were incorporated in Model 8, s uggesting the same significant and negative pattern related to antisocial behavior f ound for parental support. In addition, peer drinking groups and school connectedne ss were added as level-2 predictor.
60 Table 4: Model for Antisoc ial Behavior (Level 2) MODEL 7 MODEL 8 Fixed Effects Coefficient (St Error) Coefficient (St Error) Intercept 2.40** (0.11) 2.42** (0.04) Gender 0.84** (0.10) 0.67** (0.07) Intact Family 0.37** (0.09) 0.41** (0.09) Perceived Supervision 0.24** (0.05) 0.11** (0.03) Hispanics 0.48** (0.19) 0.48** (0.15) Parental Support -0.10**(0.01) -0.04**(0.01) Social Learning Variables Emotional Reinforcement ----0.06** (0.01) Differential Reinforcement ----0.12** (0.01) Neutralizing Definitions ----0.25** (0.02) Level-2 Variables Peer Drinking Groups ---0.59** (0.19) School Connectedness ----0.04 (0.05) Random Effects Variance (St Deviation) Variance (St Deviation) Intercept (Schools) 0.15** (0.39) 0.002 (0.04) Level-1 4.36 (2.11) 3.04 (1.74) Indicators of Fit -2LL 6718.26 6138.18 AIC 6722.26 6142.18 BIC 6732.94 6152.86 (p <0.05) ** (p <0.01)
61 The results suggest that peer drinki ng groups was significantly and positively related to antisocial behavi or within schools. In c ontrast, although the negative coefficient of school connecte dness was obtained in the expect ed direction, the result was non-significant. Furthermore, the random effects did not show variation between schools, indicating that differential social support estimated through these variables did not explain such variation in the school context. Finally, the fixed effects demonstrated that paternal support was negatively related to antisocial behavior while peer drinking groups wa s positively associated with the dependent variable. These results could s uggest that the balance of differential social support is stronger for peer drinking groups, as a source of social support for deviance, and weaker for school connectedness, as a s ource of social suppor t for conformity to conventional values. On the other hand, the presence of peer affiliation across schools and the absence of school connectedness may weaken the protective role of paternal support at the individual level.
62 Chapter Five Discussion The main goal of this study was to assess the influence of parental support on antisocial behavior among adolescents. The re sults confirm that parental support has a direct negative effect on antisocial behavior. This finding is consiste nt with Hypothesis 1 and research evidence which suggests that rece iving parental support is inversely related to the development of antisocial behavior (Wright, 1995, Wright & Cullen, 2001; Perrone et al., 2004; Jones, et al., 2007). Another relevant aim of this study was to evaluate the mediating effects of social learning mechanisms over the influence of parental support on antisocial behavior. Emotional reinforcement, differential reinfor cement, and neutralizing definitions at least partially mediated the asso ciation between parental s upport and antisocial behavior within the full model. The results indicated a 60 % in the reducti on of parental support coefficient when social learning variable s were included, which means that social learning mechanisms exerted a significant func tion in the influence of parental support. These results are in line with a wide number of findings in the crim inological literature that provide support for social learning theory (i.e. Akers et al., 1979; Sellers & Winfree, 1990; Hwang & Akers, 2003; Wang & Jensen, 2003; Miller et al ., 2008). Although the results confirmed Hypothesis 2 with respect to the mediating influence of emotional, differential reinforcement, and neutralizing de finitions, the outcomes of this study did not support the mediating effect of social reinforcement in the full model.
63 At the same time, this pattern of result s means that those adolescents who report parental support are less likely to be involved in antisocial behavior if they are exposed to unfavorable reinforcement and definitions toward deviant acts. In contrast, those adolescents who do not perceive parental support are more likely to engage in antisocial behavior if they share favorable reinfo rcement and definitions toward delinquent behavior. Consequently, the significan t effects of emotional reinforcement, differential reinforcement, and neutralizing definitions suggest that emotional support provided by parents and effective guidance are critical aspects of pa rental practice to prevent antisocial behavior (Baumrind, 1991, Wright & Cullen, 2001; Perr one et al., 2004). At the same time, this significant finding also indicate that adolescent with high levels of parental support tend to be aware of the cons equences of delinquent acts and they are less likely to justify antisocial behavior. According to Berzonsky (2004) these are c ognitive characteristics associated with the informational identity processing styles. A dolescents with this cognitive style are selfreflective, conscientious, ope n to experience, problem-focused, and vigilant decision makers. This cognitive profile may be involve d in the social learning mechanisms that promote a prosocial pathway as part of the pr otective effects of pare ntal support. In this manner, parental support is an interactiv e learning process which relies on the communication skills of the parents and th e information-processing skills of the adolescents (Akers, 1998; Stice & Gonzales, 1998). However these possible relationships may be developed in future research.
64 Likewise, social learning va riables are also related to the differential effects of maternal and paternal support. Both maternal and paternal support demonstrated to be significant predictors of antisoc ial behavior in the presence of emotional, social, and differential reinforcement. Interestingly, when neutralizing definitions are included in the model, paternal support remained significant, while maternal support became nonsignificant. This result indicates that fath ers assume a relevant parenting role when justifications of antisocial behavior in crease among adolescents, probably during the upper grades (Siegal, 1987; Forehand & Nousiainen, 1993; Harris & Marmer, 1996, Russell & Saebel, 1997; Shek, 2005). Accordi ng to and Stein, Milburn, Zane and Rotheram-Borus (2009) paternal support wa s the primary predictor of adolescent functioning outside the home. On the other hand, research has demonstrated that the influence of maternal support is highly signi ficant in the emotiona l development of the child at the beginning of adolescence (Hawki ns et al., 1992; Kliewer, Fearnow & Miller, 1996; Stice & Gonzales, 1998; Laible & Carl o, 2004). The differential outcomes found between mothers and fathers may reflect a sp ecific developmental stage where authority, discipline, and control traditi onally associated with paternal image play a protective role to prevent antisocial behavior (Siegal, 1987, Collins, & Russell, 1991; Russell, & Saebel, 1997; Amato & Rivera, 1999; Kerr, Capaldi, Pears, & Owen, 2009). The purpose of the multilevel analyses was to evaluate the meso-social dimension of differential social support (Cullen, 1994). In the context of this research, Hypothesis 3 regarding the differential e ffects between school connectedness and peer drinking groups was partially supported. The results indicate d a positive significant effect for peer drinking groups within schools, while school connectedness showed non-significant
65 influence. According to Cullen (1994), these fi ndings suggest that the source of social support for deviance is stronger than the so cial support received for conformity and conventional values. The outcomes related to differential social support revealed that exposure to peer drinking groups promote antisocial behavi or, while paternal support preserves its protective effect at the indivi dual level. These findings are in line with social learning theory, where delinquent behavior is influen ced by definitions shared within social groups and communities through the communicati on process, providing social identity, membership, and a sense of belonging (Ake rs, 1992; White et al., 2002; Felson et al., 2008). At the same time, the significant eff ects of peer affiliati on indicate a partial moderating effect over the influence of pa rental support on antisoc ial behavior. Those adolescents who belong to schools with high le vels of peer drinki ng groups and perceive poor parental support are more likely to engage in antisocia l acts. In contrast, those adolescent who belong to schools with low le vels of peer drinking groups and perceive strong parental support are less likely to be involved in antisocial behavior. According to SSSL Model (Akers, 1998), fa milies, peers, and schools constitute a mesosystem to which the adolescent relates and provides the learni ng environments that promote or discourage criminal or prosoc ial behavior. The mesosystem family-peersschool comprises the immediate social context in which social learning variables operates through the process of di fferential association. Although parental support exerts its main effects on antisocial behavior at the individual level, peer dinki ng groups as part of the school s context may moderate such effects at the school level. The meso-level of analysis indicate s that the systemic
66 influence of parental support and peer dr inking groups may reflect some meanings of deviant behavior for adolescents within th e school context: 1) Deviant behavior as symbol of adult status, imitating deviant pa rents; 2) Antisocial behavior promoted by peer groups as symbol of adoles cent rebellion, rejecting parent al authority or expectations (Akers, 1992; Sellers & Wi nfree, 1990). On the other hand, the non-significant result for school conn ectedness deserves special attention in this research. Accord ing to social learni ng theory, school connectedness offers the opportunity for learni ng law-abiding definitions from prosocial models. Also, the exposure to prosocial networ ks promotes favorable attitudes for social adjustment and decreases the opportunities for antisocial peer affiliations and delinquent definitions, creating supportive contexts for adolescents well-being at the meso-level (Jang & Johnson, 2001; Vieno, et al., 2007; Ka ufmann et al., 2007; McGrath et al., 2009). The absence of an influence for school connectedness suggests that poor prosocial networks within the school c ontext weaken the protective role exerted by parental support at the individual level. Therefore, differentia l social support is unba lanced and antisocial behaviors are more likely to be promoted by de viant peer affiliation at the school level. In accordance with Hagan and McCarthy (1997), crim inal capital is encouraged in schools where adolescents share technical informa tion and deviant definitions for success in criminal enterprises. In contrast, weakened law-abiding sources of social support do not promote effectively the deve lopment of social capita l where students have the opportunity to share norms and prosocial defi nitions to ensure the well-being in the communitys social life (Coleman, 1990).
67 The mesosystem family-peers-school also re presents the immediate social context in which the communitys social life imp acts on adolescent be havior (Akers, 1998). Social disorganization theory (Shaw & Mc Kay, 1969) as an ecological perspective on neighborhood crime, may contribute to the unders tanding of social stru ctural factors in which the mesosystem is embedded. Analyzing contextual conditi ons related to crime rate s, Cullen (1994) proposed relevant arguments about the Ecology of Social Support (Shaw & McKay, 1969). Based on social disorganization theor y, Cullen (1994) argued that communities characterized by family disruption, weak friendship networks, and low voluntary participation in the neighborhood exhibit hi gher crime rates (Bursik & Grasmick, 1993; Byrne & Sampson, 1986; Reiss & Tonr y, 1986; Sampson & Groves, 1989). Along these lines, Sutherland (1947), influe nced by sociologists of the Chicago School, introduced the concept of differentia l social organization (macro-level) to explain the process of differen tial association (micro-level) as the result of exposure to different conforming and criminal definitions Sutherland (1947) argued that instead of being socially disorganized, these groups ar e socially organized around different values and goals. Thus, delinquent cultural trad itions and criminal capital (Hagan & McCarthy, 1997) are transmitted from one genera tion to the next. Over the last decades, the subcultural model has been used within so cial disorganization th eory to explain how social disorganization leads to delinque ncy (i.e. Cloward, & Ohlin, 1960; Wolfgang, & Ferracuti, 1967; Kandel, & Davies, 1991; Felson, Liska, South, & McNulty, 1994). Recently, the hypothesis of delinquent cult ural transmission from one generation to the next has been successfully tested. Blazei, Iacono, and McGue (2008) examined the
68 transmission of antisocial behavior from fa ther to child. They found strong evidence suggesting that antisocial be havior is learned and extern alized by children who were exposed to an antisocial father during pre-adolescence and la te-adolescence. Similarly, Farrington, Coid and Murray ( 2009) demonstrated that convictions of fathers predicted convictions of sons after cont rolling for risk factors. At the same time, there was significant intergenerational tran smission of delinquent behavior among males until three successive generations. In contrast Kerr, Capaldi, Pears, and Owen (2009) found evidence on the transmission of cons tructive paternal parenting from one generation to another. These findings conf irm the hypothesis of differential social organization (Sutherland, 1947) where groups are socially organized around conforming and criminal values. Thus, structural factor s distributed according the Ecology of Social Support (Cullen, 1994) influence the mesosy stem family-peers-school in patterns of Differential Social Support, strengthening the sources of social capital and criminal capital within the youth subculture. Consistent with social disorganizat ion theory (Shaw & McKay, 1969) and the Ecology of Social Support (Cullen, 1994), this study identified several risk groups for the development of antisocial behavior, consideri ng individual and contex tual factors related to the influence of parental support: 1) Adolescent males who perceive weak parental support, 2) Adolescents who live in disrupted families, 3) Adolescents who perceive inconsistent supervision, and 4) Adolescents of Hispanic origin. A ll these groups may be considered a target audience to preven t criminal behaviors among adolescents. In this manner, the significant inverse relationship between parental support and antisocial behavior indicates important imp lications for crime prevention (Cullen, 1994).
69 At the individual level, training parents to pr ovide parental support to prevent antisocial behavior must emphasize the parents abi lity to establish emotional communication patterns and effective guidance for their ch ildren to anticipate the consequences of delinquent acts (differential reinforcement) avoiding the development of beliefs that justify criminal behavior (neutralizing definitions). In contrast, interventions addressed to adolescents must highlight the development of their information-processing skills to understand parent/guardian advice, creating pr osocial definitions as part of these interactions (Akers, 1998; Stice & Gonzales, 1998). At the school level, social intervention st rategies must be concentrated on the promotion of a social support culture within the school environment. According to Cullen (1994), the lack of social s upport, and not only coercion a nd punishment, are implicated in the causation of crime. Cullen (1994) ar gued that American soci ety is not organized, structurally or culturally, to be socially supportive. Consid ering this point of view, the promotion of social support within the school context may counterbalance the effects of criminal networks and decrease the probabi lity of being involved in delinquent acts. Social support as a cultural value creates a learning environment within the school context for prosocial pathways, strengtheni ng the protective role of parental support developed inside each family. This multilevel study presents several limitations. Perhaps the most critical shortcoming of this research was the low numb er of level-2 units ( 16 Schools). Although several criminological studies have employed similar samp le sizes at level-2, the methodological consideration for HLM analys es requires more units to ensure the accuracy of the results at the school level. On the other hand, variables at level-2 were
70 aggregated from the individual level. This procedure, common in HLM analyses, may introduce the estimation of compositional eff ects of the sample rather than actual contextual influences. In this research, th e results produced at the meso level may be compositional effects and they must be consid ered as indicators of the school context environment. To avoid aggregation bias, Gottfredson (2001) recommends using meso-level or macro-level data (i.e. Department of Educati on, Census data) instead of aggregating data from individuals as level-2 i ndicators. Also, HLM allows meas uring social structure at a different level, such as neighborhood, anal yzing multiple sources of data. Further HLM research with contextual data will be needed to ensure a higher scope of multilevel research designs. Likewise, the cross-sectional nature of the data is a restriction of this study. Cross-sectional surveys present limitations to establish the caus al order between the variables. To solve this methodological shor tcoming, future studies may use longitudinal data to evaluate reciprocal influence over time between fa ther and children, and assess mediation analysis within the res earch model (Amato & Rivera, 1999). Another important limitation of this research was related to the indicators of antisocial behavior used in th e social learning variables. Adolescent alcohol use was the main characteristics included to assess th e social learning mechanisms regarding antisocial behavior. Although underage drin king is a common cause of adolescent misbehavior, there are other categories of antisoc ial behavior that must be considered. In this way, the mediating effects of social learning mechanisms might be skewed specifically toward alcohol use as an indicator of misbehavior.
71 Some shortcomings of the original da tabase limited the methodological approach of this research. The database provided by the Florida Prevention Research Institute at USF did not provide the age of the students who participated in the HLM analyses. The age of the adolescents is a critical contro l variable that serves as a developmental indicator for examining the relationship between parental support and antisocial behavior (Sampson & Laub, 1993; Warr, 2002; Farringt on, Coid & Murray, 2009). However, the variation obtained across schools might involve developmental changes among adolescents between sixth and el eventh grade. In spite of these limitations, this resear ch has contributed to the analysis of individual and contextual factors involved in the infl uence of parental support on antisocial behavior among adolescents. The multilevel approach provides a broader understanding of this complex phenomenon in a single methodology. However, this methodological approach could be improved in future research if a mixed method strategy is incorporated to determine the cult ural meanings involved in the social learning mechanisms. Finally, the social support paradigm provided a theoretical background to understand the protective effects of parental su pport within the family context. Also, the theoretical model allowed di scovering school context factors that may also serve to prevent antisocial behavior. These findings coul d be useful to formulate effective social interventions and to evaluate social polic ies regarding crime prevention based on the highest values of social inte gration, cooperation, and altruism At the same time, further research is needed to advance our understand ing of individual and c ontextual factors that determine the efficacy of parental support to prevent antisocial behavior.
72 References Agnew, R. (1991). The interactive eff ects of peer variab les on delinquency. Criminology, 29(1) 4772. Amato, P. R. (1994). Father-child relations mother child relations and offspring psychological wellbeing in early adulthood. Journal of Marriage and the Family, 56, 10311042. Amato, P. R., & Rivera, F. (1999). Patern al involvement and childrens behavior problems. Journal of Marriage and the Family, 61 375384. Akers, R. (1973). Deviant Behavior: A Social Learning Approach. Belmont, CA: Wadsworth Publishing Company. Akers, R. (1992) Drugs, Alcohol, and Societ y: Social Structure, Process and Policy. Belmont, CA: Wadsworth. Akers, R. (1996). Is differential associati on/social learning cultu ral deviance theory? Criminology, 34(2) 229-247. Akers, R. (1998). Social learning and social structure: A general theory of crime and deviance. Boston: Northeastern University Press. Akers, R., Krohn, M., Lanza-Kaduce, L. & Radosevich, M. (1979). Social learning and deviant behavior: A specific test of a general theory. American Sociological Review, 44(4) 636-655. Akers, R. & Jensen, G. (2006). The empirical status of social lear ning theory of crime and deviance: The past, present, and future. In F. T. Cullen, J. P Wright, & K. R. Blevins, (Eds.) Taking Stock: The Status of Cr iminological Theory. Advances in
73 Criminological Theory, Vol. 15. (pp. 37-76). New Brunswick, NJ: Transaction Publishers. Akers, R. L. & Sellers, C. S. (2004). Criminological Theories: Introduction, evaluation, and application Los Angeles: Roxbury Publishing Company. Alexander, J. F. (1973) Defensive and S upportive Communications in Normal and Deviant Families. Journal of Consulting and Clinical Psychology. 40(2) 223-231. Apel, R. & Kaukinen C. (2008). On the Re lationship Between Family Structure and Antisocial Behavior: Parental Coha bitation and Blended Households. Criminology 46(1), 35-70. Ardelt, M. & Day, L. (2002). Parents, Siblings and Peers: Close Social Relationships and Adolescent Deviance. Journal of Early Adolescence 22 (3), 310-349. Bachman, R. & Shutt, R. (2003). The Practice of Research in Criminology and Criminal Justice. Thousand Oaks: Sage Publications Company. Barber, B., & Thomas, D. (1986). Dimensi ons of fathers an d mothers supportive behavior: The case for physical affection. Journal of Marriage and the Family 48 783-794. Barnes, G. & Farrell, M. (1992). Parental Support and Control as Predictors of Adolescent Drinking, Delinquency, a nd Related Problem Behaviors. Journal of Marriage and the Family, 54 (4), 763-776. Baron, R., & Kenny, D. (1986). The moderator-med iator variable dis tinction in social psychology research: Conceptual, strategi c, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182.
74 Baumrind, D. (1991). The influence of pare nting style on adolescent competence and substance abuse. Journal of Early Adolescence, 11(1), 56-95. Berzonsky, M. (1989). Identity Style: Conceptualization and Measurement Journal of Adolescent Research, 268 (4), 268-282. Berzonsky, M. (2004). Identity Style, Parent al Authority, and Identity Commitment. Journal of Youth and Adolescence, 33(3), 213220. Blazei R., Iacono W. & Mcgue M. (2008). Fa ther-Child Transmission of Antisocial Behavior: The Moderating Role of Father's Presence in the Home. Journal of American Academy of Child and Ad olescent Psychiatry, 47 (4), 406-415. Brannigan, a., Gemmell, W., Pevalin, D., &W ade, T. (2002). Self-control and social control in childhood misconduct and aggressi on: The role of family structure, hyperactivity, and hostile parenting. Canadian Journal of Criminology, 44(2), 119-142. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press. Bronstein, P. (1984). Differences in mothers and fathers behavior s toward children: A cross-cultural comparison. Developmental Psychology, 20(6), 995-1003. Brookmeyer, K., Fanti, K. & Henrich, C. ( 2006). Schools, Parents, and Youth Violence: A Multilevel, Ecological Analysis. Journal of Clinical Child & Adolescent Psychology, 35(4), 504-514. Bryk, A. & Raudenbush, S. (1992). Hierarchical linear models. Newbury Park: Sage Publications.
75 Bunge, M. (2006). A systemic perspective of crime. In P. H. Wikstrom & R. J. Sampson. (Eds.) The explanation of crime: Cont ext, mechanisms, and development (pp. 830) New York: Cambridge University Press. Burdzovic, A. & OFarrell, T. (2007). Long itudinal Associations between Fathers Heavy Drinking Patterns and Children s Psychosocial Adjustment. Journal of Abnormal Child Psychology, 35(1), 116. Burgess, R. & Akers, R. (1966). A differen tial association-reinforcement theory of criminal behavior. Social Problems 14 128-147. Bursik, R. & Grasmick, H. (1993). Neighborhoods and crime: The dimensions of effective community control New York: Lexington Books. Byrne, J. & Sampson, R. (1986). The social ecology of crime. New York: SpringerVerlag. Carlo, G., Roesch, S. C., & Melby, J. (1998). The multiplicative relations of parenting and temperament to prosocial and an tisocial behaviors in adolescence. Journal of Early Adolescence, 18(3), 266-290. Carmine, E. & Zeller, R. (1979). Reliability and Validity Assessment Sage University Paper Series on Quantitative Application in the Social Sciences, 17. Newbury Park, CA: Sage. Catalano, R. (1979). Health, behavior and the commun ity: An ecological perspective. Elmsford, NY: Pergamon. Catalano, R., Kosterman, R., Hawkins, J., Newcomb, M., & Abbott, R. (1996) Modeling the etiology of adolescent subs tance use: A test of the social development model. Journal of Drug Issues, 26(2), 429-455
76 Centers for Disease Control and Preven tion (2008a). School-associated student homicides-United States, 1992-2006. MMWR, 57(02), 33-36. Centers for Disease Control and Prevention ( 2008b) Youth Risk Behavior Surveillance United States 2007. MMWR, 57(SS-4), pp. 131 Chawla, N., Neighbors, C., Logan, D., Lewi s, M. & Nicole, F. (2008). Perceived Approval of Friends and Parents as Medi ators of the Relationship Between SelfDetermination and Drinking. Journal of Studies on Al cohol and Drugs, 70(1), 92100. Cloward, R., & Ohlin, L. (1960). Delinquency and Opportunity. New York: Free Press. Coleman, J. S. (1990). Foundations of Social Theory. Cambridge, MA: Harvard University Press. Collins, A., & Russell, S. (1991). Mother-child and father-child relations in adolescence: A developmental analysis. Developmental Review 11 99-136. Colvin, M., Cullen, F. & Vander, T. (2002). Coercion, social support, and crime: An emerging theoretical consensus. Criminology, 40 (1), 19-42. Coreil, J., Bryant, C. & Henderson, N. (2001). Social and Behavioral foundations of Public Health. Thousand Oaks: Sage Publications, Inc. Cressey, D. (1960). Epidemiology and indivi dual conduct: A case from criminology. The Pacific Sociological Review 3(2) 47-58. Cullen, F. (1994). Social support as an organizing concept fo r criminology: Presidential address to the Academy of Cr iminal Justice Sciences. Justice Quarterly, 11(4), 528-559.
77 Cullen, F., Wright, J. & Chamlin, M. (1999) Social Support and Social Reform: A Progressive Crime Control Agenda. Crime and Delinquency, 45(2) 188-207. Darling, N., & Steinberg, L. (1993). Parenting style as context: An integrative model. Psychological Bulletin, 113(3), 487-496. Dedrick, R., Ferron, J., Hess, M., Hogarty, K., Kromrey, J., Lang, T, Niles, J. & Lee, R. (2009). Multilevel modeling: A review of methodological issues and applications. Review of Educational Research, 79 (1), 69 -102 Dembo, R., Grandon, G., La Voie, L., Schmeidl er, J. & Burgos, W. (1986). Parents and Drugs Revisited: Some Further Evidence in Support of Social Learning Theory. Criminology, 24(1), 85-104. Demuth, S. & Brown, S. L. (2004) Family st ructure, family processes, and adolescent delinquency: The significance of parent al absence versus parental gender. Journal of Research in Crime and Delinquency, 41 (1) 58-81 Dixon, J. (2008). Predicting student per ceptions of school connectedness: The contributions of parent att achment and peer attachment. Dissertation Abstracts International, The Humanities and Social Sciences, 68(12), pp.5029 Downey, D. B., Ainsworth-Darnell, J. W., & Dufur, M. J. (1998). Sex of parent and childrens wellbeing in single parent households. Journal of Marriage and the Family, 60, 878893. Eaton, D., Forthofer, M., Zapata, L., Mc Cormack, K., Bryant, C., Reynolds, S. & McDermott, R. (2004). Factors Relate d to Alcohol Use Among Sixth through Tenth Graders: The Sarasota County Demonstration Project. Journal of School Health, 74(3), 95-104.
78 Farrell, M., Barnes, G. & Banerjee, S. (1995) Family cohesion as a buffer against the effects of problem-drinking fathers on ps ychological distress, deviant behavior, and heavy drinking in adolescents. Journal of Health and Social Behavior 36(4) 377-385. Farrington, D., Coid, J. & Murray, J. (2009). Family factors in th e intergenerational transmission of offending. Criminal Behaviour a nd Mental Health, 19, 109124. Felson, R., Teasdale, B. & Burchfield, K. (2008). The Influence of Being under the Influence Alcohol Effects on Adolescent Violence. Journal of Research in Crime and Delinquency, 45 (2), 119-141. Felson, R., Liska, A., South, S. & McNulty, T. (1994). The subcultu re of violence and delinquency: Individual vs. school context effects. Social Forces, 73, 155 173. Forehand, R. & Nousiainen, S. (1993). Mate rnal and Paternal Parenting: Critical Dimensions in Adolescent Functioning. Journal of Family Psychology, 7(2), 213221. French, M. & Maclean, J. (2006). Underage Alcohol use, Delinquency, and Criminal Activity. Health Economics, 15, 12611281. Gottfredson, D. (2001). Schools and Delinquency. New York: Cambridge University Press Glueck, S. & Glueck, E. (1950). Unraveling Juvenile Delinquency. Cambridge, MA. Harvard University Press. Guilamo-Ramos, V., Jaccard, J., Turrisi, R., Johansson, M. & Bouris, A. (2006). Maternal Perceptions of Alcohol us e by Adolescents who Drink Alcohol. Journal of Studies on Alcohol, 67(5), 730-737.
79 Hagan, J. & McCarthy, B. (1997). Anomie, soci al capital, and street criminology. In: N. Passas, & R. Agnew (Eds.), The Future of Anomie Theory (pp. 124-141). Boston, MA: Northeastern University Press. Harris, J. (1995) Where is th e childs environment? A gr oup socialization theory of development. Psychological Review, 102(3), 458-489. Harris., K. & Marmer J. (1996). Poverty, pa ternal involvement, and adolescent wellbeing. Journal of Family Issues, 17(5), 614-627. Hawkins, D., Catalano, R. & Miller, J. (1992) Risk and Protective Factors for Alcohol and other Drug Problems in Adolescence and Early Adulthood: Implications for Substance Abuse Prevention. Psychological Bulletin, 112 (1), 64-105. Haynie, D. (2002). Friendship networks and delinquency: The relativ e nature of peer delinquency. Journal of Quantitative Criminology 18(2), 99-134. Henry, K., Slater, M. & Oetting, E. (2005). Alcohol Use in Early Adolescence: The Effect of Changes in Risk Taking, Per ceived Harm and Friends' Alcohol Use. Journal of Studies on Alcohol, 66(2), 275-283. Hirschi, T. (1969). Causes of Delinquency Berkley, CA: University of California Press. Hwang, S. & Akers, R. (2003). Substance use by Korean Adolescen ts: A cross-cultural test of Social Learning, Soci al Bonding, and Self-Control Theories. In R. L. Akers & G. F. Jensen (eds.), Social Learning Theory and th e Explanation of Crime: A guide for the New Century. Advances in Criminological Theory, Vol. 11. (pp. 3964). New Brunswick, NJ: Transaction Publishers Ingoldsby, E., Shaw, D., Winslow, E., Schonbe rg, M., Gilliom, M. & Criss, M. (2006). Neighborhood Disadvantage, ParentC hild Conflict, Neighborhood Peer
80 Relationships, and Early Antisocial Behavior Problem Trajectories. Journal of Abnormal Child Psychology, 34(3), 303319. Ingram, J., Patchin, J., Huebner, B., McClus key, J. & Bynum, T. (2007). An Examination of Direct and Indirect Effect s among At-Risk Early Adolescents. Criminal Justice Review, 32(4), 380-400. Jang, S. & Johnson, B. (2001). Neighborhood di sorder, individual religiosity, and adolescent use of illicit drugs: A Test of Multilevel Hypotheses. Criminology, 39(1), 109-143. Jenkins, P. (1997) School deli nquency and school social bond. Journal of Research in Crime and Delinquency, 34(3) 337-367. Jensen, G. (1972). Parents, peers, and deli nquent action: A test of the differential association perspective. American Journal of Sociology, 78(3) 562-575. Jessor, R. (1993). Successful adolescent deve lopment among youth in high-risk settings. American Psychologist, 48(2), 117-126 Jones, S., Cauffman, E. & Piquero, A. ( 2007). The influence of parental support among incarcerated adolescent offenders: The moderating effects of self control. Criminal Justice and Behavior, 34(2), 229-245 Kaczynski, K., Lindahl, K., Malik, N. & La urenceau, J. (2006). Marital Conflict, Maternal and Paternal Parenting, and Child Adjustment: A Test of Mediation and Moderation. Journal of Family Psychology, 20(2), 199208. Kandel, D. (1990) Parenting Styles, Drug Use, and Children's Adjustment in Families of Young Adults. Journal of Marriage and the Family, 52 183-96.
81 Kandel, D., & Davies, M. (1991). Friendshi p networks, intimacy, and illicit drug use in young adulthood: a comparison of two competing theories. Criminology, 29(3), 441 469. Kaufmann, D., Wyman, P., Forbes-Jones, E. & Barry, J. (2007). Prosocial Involvement and Antisocial Peer Affiliations as Pred ictors of Behavior Problems in Urban Adolescents: Main Effects and Moderating Effects. Journal of Community Psychology, 35(4), 417434 Kerr, D., Capaldi, D., Pears, K. & Owen, L. (2009). A prospectiv e three generational study of fathers constructive parenting: Influences from family of origin, adolescent adjustment, and offspring temperament. Developmental Psychology, 45(5) 1257-1275. Kliever, W., Fearnow, M. & Miller, P. ( 1996). Coping socializati on in middle childhood: Tests of maternal and paternal influences. Child Development, 67 2339-2357. Kreft, I. & De Leeuw, J. (1998) Introducing Multilevel Modeling Thousand Oaks, CA: Sage. Krohn, M. (1986). The web of conformity: A ne twork approach to the explanation of delinquent behavior. Social Problem, 33, 81-93. Krohn, M., Lanza-Kaduce, L. &Akers, R. ( 1984). Community Context and Theories of Deviant Behavior: An Examination of Social Learning and Social Bonding Theories. The Sociological Quarterly, 25(3), 353-372. Kubrin, C. & Weitzer, R. (2003). New direc tions in social disorganization theory Journal of Research in Crime and Delinquency, 40(4), 374-402.
82 Laible, D. & Carlo, G. (2004). The differential relations of maternal and paternal support and control to adolescent social competence, self-worth, and sympathy. Journal of Adolescent Research, 19 (6), 759-782. Lanza-Kaduce, L. & Capece, M. (2003). Social Structure-Social Learning (SSSL) and Binge Drinking: A Specific Test of an Integrated General Theory. In R. L. Akers & G. F. Jensen (eds.), Social Learning Theory and the Explanatio n of Crime: A guide for the New Century. Advances in Criminological Theory, Vol. 11. (pp. 179-196) New Brunswick, NJ: Transaction Publishers Larson, R. & Verma, S. (1999). How Children and Adolescents Spend Time Across the World: Work, Play, and De velopmental Opportunities. Psychological Bulletin, 125(6), 703-736. Lee, G., Akers, R. & Borg, M. (2004). So cial Learning and Structural Factors in Adolescents Substance Use. Western Criminology Review, 5 (1), 17-34. Lin, N., Dean, A., & Ensel, W. (1986). Social support, life events, and depression. New York: Academic Press Lin, N. (2001) Building a network theory of so cial capital. In N. Lin, K. Cook, & R. S. Burt, (Eds.) Social capital: Theory and research. (pp. 3-29) New York: Aldine de Gruyter. Lipperman-Kreda, S., Paschall, M. & Grube, J. (2009). Perceived Local Enforcement, Personal Beliefs, and Underage Drinki ng: An Assessment of Moderating and Main Effects. Journal of Studies on Alcohol and Drugs, 70(1), 64-69. Liska, A. (1973). Causal structures unde rlying the relationship between delinquent involvement and delinquent peers. Sociology and Social Research 58(1) 23-26.
83 Loeber, R., & Stouthamer-Loeber, M. (1986). Fam ily factors as correlates and predictors of juvenile conduct problems and delinque ncy. In M. Tonry & N. Morris (Eds.), Crime and Justice: An Annual Review of Research, Vol. 7. (pp. 29-149). Chicago: University of Chicago Press. Mack, K., Leiber, M., Featherstone, R. & M onserud, M. (2007) Reassessing the familydelinquency association: Do family type, family processes, and economic factors make a difference? Journal of Criminal Justice, 35 5167. Marshal, M. & Chassin, L. (2000). Peer Influence on Adolescent Alcohol Use: The Moderating Role of Parent al Support and Discipline. Applied Developmental Science, 4 (2), 8088. Marsiglio, W., Amato, P., Day, R. D., & Lamb, M. E. (2000). Scholarship on fatherhood in the 1990s and beyond. Journal of Marriage and the Family, 62, 11731191. Matsueda, R. (1988). The current state of differential association theory. Crime and Delinquency 34(3) 277-306. Matsueda, R. & Heimer, K. (1987). Race, fam ily structure, and delinquency: A test of differential association and social control theories. American Sociological Review 52(6) 826-840. Matsueda, R. & Anderson, K. (1998). The dyna mics of delinquent peers and delinquent behavior. Criminology, 36(2) 269-308. McCord, J. (1991a). The cycle of crime and socialization practices. The Journal of Criminal Law and Criminology, 82(1) 211-228. McCord, J. (1991b). Family relationships, j uvenile delinquency, and adult criminality. Criminology, 29(3), 397-418.
84 McGrath, B., Brennan, M., Dolan, P. & Barn ett, R. (2009). Adolescent Well-being and Supporting Contexts: A Comparison of A dolescents in Ireland and Florida. Journal of Community & Applied Social Psychology, 19(4), 299-320. McIntosh, A., Fitch S., Branton, W. & Nybe rg, K. (1981). The Effect of Mainstream Religious Social Controls on A dolescent Drug Use in Rural Areas. Review of Religious Research 23(1), 54-75. Mears, D., Ploeger, M., & Warr, M. (1998) Explaining the Gender Gap in Delinquency: Peer Influence and Moral Evaluations of Behavior. Journal of Research in Crime and Delinquency, 35, 251-266. Miller, H., Jennings, W., Alvarez-Rivera, L. & Miller, M. (2008). Explaining Substance use among Puerto Rican Adolescents: A Part ial Test of Social Learning Theory. Journal of Drug Issues, 38(1), 261-283. Miller, T., Levy, D., Spicer, R., & Taylor, D. (2006). Societal Costs of Underage Drinking. Journal of Studies on Alcohol, 67(4), 519-28. Ozer, E. (2005). The Impact of Violence on Urban Adolescents: Longitudinal Effects of Perceived School Connection and Family Support. Journal of Adolescent Research 20(2), 167-192. Parker, D., Levin, B., & Harford, T. (1996). E ffects of Early Drinki ng and an Antisocial Orientation on the Alcohol use of Young Russians. Alcoholism: Clinical and Experimental Research, 20(7), 1179-1183. Payne, A. (2008). A Multilevel Analysis of the Relationships among Communal School Organization, Student Bonding, and Delinquency. Journal of Research in Crime and Delinquency 45(4), 429-455.
85 Pearson, F. & Weiner, N. (1985). Toward an integration of crim inological theories. The Journal of Criminal Law and Criminology, 76(1), 116-150 Perrone, D., Sullivan, C., Pratt, T. & Marg aryan, S. (2004). Parental efficacy, selfcontrol, and delinquency: A test of the general theory of crime on a nationally representative sample of youth. International Journal of Offender Therapy and Comparative Criminology, 48(3), 298-312. Raudenbush, S. & Bryk, A. (2002). Hierarchical linear models: Applications and data analysis methods Newbury Park: Sage Publications. Rebellon, C. (2002). Reconsidering the Br oken Homes/Delinquency Relationship and Exploring its Mediating Mechanism. Criminology, 40(1), 103-135. Rebellon, C. (2006). Do Adolescents Engage in Delinquency to Attract the Social Attention of Peers? An Extension an d Longitudinal Test of the Social Reinforcement Hypothesis. Journal of Research in Crime and Delinquency, 43(4), 387-411. Reiss, A. & Tonry, M. (1986). Communities and crime. Chicago: University of Chicago Press. Rossman, B. & Rea, J. (2005). The relation of parenting styles and inconsistencies to adaptive functioning for children in conflictual and violent families. Journal of Family Violence, 20(5), 261-277 Russell, A. & Saebel, J. (1997). Mother-son, mother-daughter, father-son, and fatherdaughter: Are they dis tinct relationships? Developmental Review, 17, 111-147 Sampson, R. (1991). Linking the Microa nd Macrolevel Dimens ions of Community Social Organization. Social Forces, 70(1), 43-64.
86 Sampson, R. (1998). Local Friendship Ties a nd Community Attachment in Mass Society: A Multilevel Systemic Model. American Sociological Review 53(5), 766-779. Sampson, R. (2006a). How does community c ontext matter? Social mechanisms and explanation of crime rates. In P. H. Wikstrom & R. J. Sampson. (Eds.) The explanation of crime: Context, mechanisms, and development (pp. 31-60) New York: Cambridge University Press. Sampson, R. (2006b). Collective Efficacy Theory: Lessons Learned and Directions for Future Inquiry. In F. T. Cullen, J. P Wright, & K. R. Blevins, (Eds.) Taking Stock: The Status of Crim inological Theory. Advances in Criminological Theory, Vol. 15. (pp. 37-76) New Brunswick, NJ: Transaction Publishers Sampson, R. & Groves, B. ( 1989). Community Structure and Crime: Testing Social Disorganization Theory. American Journal of Sociology, 94(4), 774-802. Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life Cambridge: Harvard University Press. Sampson, R., Raudenbush, S. & Earls, F. ( 1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277 918-924 Sampson, R. Morenoff, J. & Earls, F. (1999). Spatial dynamics of collective efficacy for children. American Sociologi cal Review, 64, 633-660. Siegal, M. (1987). Are sons and daughters treated more differently by fathers and mothers? Developmental Review, 7, 183-209. Sellers, C. & Winfree L. (1990). Differential associations and definitions: A panel study of youthful drinking behavior. International Journal of the Addictions, 25(7) 755-771.
87 Sellers, C., Winfree, L., & Griffiths, C. (1993). Legal Attitudes, Permissive Norm Qualities, and Substance Use: A Comparis on of American Indian and NonIndian Youth. Journal of Drug Issues 23(3), 493-513. Shaw, C. & McKay, H. (1969). Juvenile delinquency and urban areas. Rev. Ed. Chicago: University of Chicago Press. Shek, D. (2005). Paternal and Maternal In fluences on the Psyc hological Well-Being, Substance Abuse, and Delinquency of Chinese Adolescents Experiencing Economic Disadvantage. Journal of Clinical Psychology, 61(3), 219234. Simons, R. L., Simons, L. G., Burt, C. H., Brody, G. H., & Cutrona, C. (2005) Collective efficacy, authoritative parent ing and delinquency: A long itudinal test of a model integrating community and family level processes. Criminology, 43(4) 989 1029. Smits, I., Soenens, B., Luyckx, K., Duriez, B., Berzonsky, M. & Goossens L. (2008). Perceived parenting dimensions and identity styles: Exploring the socialization of adolescents processing of identity-relevant information. Journal of Adolescence, 31, 151164. Song, E., Reboussin, B., Long, K., Kaltenbach, L., Wagoner, K., & Wolfson, M. (2009) Selected Community Characteri stics and Underage Drinking. Substance use and Misuse, 44, 179-194. Stein, J., Milburn, N., Zane, J. & Rotheram -Borus, M. (2009). Paternal and maternal influences on problem behavior s among homeless and runaway youth. American Journal of Orthopsychiatry, 79 (1), 39-50
88 Stewart, E. (2003). School social bonds, sc hool climate, and school misbehavior: A multilevel analysis. Justice Quarterly, 20(3), 575-604. Stice, E. & Gonzales, N. (1998). Adolescen t temperament moderates the relation of parenting to antisocial beha vior and substance use. Journal of Adolescent Research, 13(1), 5-31. Stolz, H., Barber, B. & Olsen, J. (2005). To ward disentangling fathering and mothering: An assessment of relative importance. Journal of Marriage and Family, 67, 10761092 Sutherland, E. (1947). Principles of Criminology Philadelphia: J.B. Lippincott. Sutherland, E. (1961). White collar crime New York. Holt, Rinehart and Winston Sutherland, E. & Cressey, D. (1978) Criminology. Tenth Edition. Philadelphia: J. B. Lippincott Tashakkari, A. & Teddlie, C. (1998). Mixed Methodology: Comb ining qualitative and quantitative approaches. Thousand Oaks, CA: Sage Publications. Thornberry, T. Lizotte, A. Krohn, M. Farnworth, M. & Jang S. (1994). Delinquent peers, beliefs, and delinquent behavior: A longitudinal test of interactional theory. Criminology, 32(1), 47-84 Thomas, S. & Smith, H. (2004). School Connectedness, Anger Behaviors, and Relationships of Violent a nd Nonviolent American Youth. Perspectives in Psychiatric Care, 40(4), 135-148. Underwood, M., Beron, K. & Rosen, L. (2009). Continuity and change in social and physical aggression from middle ch ildhood through early adolescence. Aggressive Behavior, 35, 357-375.
89 Valdez, A., Yin, Z. & Kaplan C. (1997) A Comparison of Alcohol, Drugs, and Aggressive Crime among Mexican-Ameri can, Black, and White Male Arrestees in Texas. The American Journal of Dr ug and Alcohol Abuse, 23(2), 249-265. Valdez, A., Kaplan C. & Codina E. ( 2000). Psychopathy among Mexican American Gang Members: A Comparative Study. International Journal of Offender Therapy and Comparative Criminology, 44(1), 46-58. Vander Ven, T., Cullen, F., Carrozza, M., & Wri ght, J. (2001). Home alone: The impact of maternal employment on delinquency. Social Problems, 48(2), 236-257 Vaux, A. (1988) Social support: Theory, re search, and intervention. New York: Praeger Vieno, A., Nation, M., Perkins, D. & Santine llo, M. (2007). Civic Participation and the Development of Adolescent Behavior Problems. Journal of Community Psychology, 35(6), 761777. Wang, S. & Jensen, G. (2003). Explaining de linquency in Taiwan: A test of social learning theory. In R. L. Akers, & G. F. Jensen (Eds.). Social Learning Theory and the Explanation of Crime. Advances in Criminological Theory, Vol. 11. (pp. 151-177). New Brunswick, NJ: Transaction Publishers. Warner, B. & Wilcox, P. (1997). Local Social Ties in a Community and Crime Model: Questioning the Systemic Nature of Informal Social Control. Social Problems, 44(4), 520-536. Warr, M., & Stafford, M. (1991). The influence of delinquent peers: What they think or what they do? Criminology 29 851-866. Warr, M. (1993). Parents, Peers, and Delinquency. Social Forces, 72(1), 247-264
90 Warr, M. (2002). Companions in Crime: The Social Aspects of Criminal Conduct Cambridge University Press. Warr, M. (2005) Making delinquent friends: adult supervision and children's affiliations. Criminology, 43(1), 77-105. Wells, S. & Graham, K. (2003), Aggression i nvolving alcohol: relationship to drinking patterns and social context. Addiction, 98, 3342. Welsh, W., Greene, J. & Jenkins P. (1999). School disorder: The influence of individual, institutional, and community factors. Criminology, 37(1), 73-115. White, H., Loeber, R., StouthamerLoeber, M & Farrington, D. (1999). Developmental associations between substance use and violence. Development and Psychopathology, 11, 785803. White, H., Tice, P., Loeber, R. & Stouthamer Loeber M (2002). Illegal Acts Committed By Adolescents under the Infl uence of Alcohol and Drugs. Journal of Research in Crime and Delinquency 39 (2), 131-152. Wikstrom, P. & Sampson, R. (2003) Social in fluence of community influences on crime and pathways in criminality. In B. L. Benjamin, T. E. Moffitt, & A. Caspi (Eds.) Causes of conduct disorder and Juvenile Delinquency. (pp. 118-148) New York: The Guilford Press. Wilcox, P. & Clayton, R. (2001). A multile vel analysis of school-based weapon possession. Justice Quarterly, 18(3), 509-541. Wilson, D. (2004). The Interface of School Climate and School Connectedness and Relationships with Aggression and Victimization. Journal of School Health 74(7), 293-299.
91 Winfree, T., Backstrom, T. & Mays, L. (1994). Delinquency and Youth Gangs A New Twist on a General Theory of Crime and Delinquency. Youth & Society, 26(2), 147-177. Wolfgang, M., & Ferracuti, F. (1967). The Subculture of Violence: Towards an Integrated Theory in Criminology. London: Tavistock Publications. Wright, J. (1995). Parental Support and Juvenile Delinque ncy: A test of Social Support Theory Unpublished Ph. D. dissertation, University of Cincinnati. Wright, B., Caspi, A., Moffitt, T. and Silva, P. (2001). The effects of social ties on crime vary by criminal propensity: A life -course model of interdependency. Criminology 39(2) 321-352. Wright, J., & Cullen, F. (2001). Parental e fficacy and delinquent behavior: Do control and support matter? Criminology, 39(3), 677-706. Wright, J., Cullen, F. & Miller, J. (2001) Family social capital and delinquent involvement. Journal of Criminal Justice, 29(1), 1-9. Youniss, J., Yates, M., & Su, Y. (1997). So cial integration: Co mmunity service and marijuana use in high school seniors. Journal of Adolescent Research, 12, 245 262.
93 Appendix A: Research Variables in the Multilevel Model Descriptive Statistics Variables Item Metric Mean S.D Index Antisocial Behavior 0= Low; 1= High 2.43 2.31 0-10 Parental Support 0 = Weak; 4= Strong 31.97 6.05 6-40 Maternal Support 0 = Weak 4= Strong 16.63 3.25 4-20 Paternal Support 0 = Weak; 4= Strong 15.34 4.02 4-20 Emotional Reinforcement 1= Favorable; 4= Unfavorable 11.86 3.07 4-16 Social Reinforcement 1= Favorable ; 4= Unfavorable 15.31 3.17 5-20 Neutralizing Definitions 1= Favorable; 4= Unfavorable 10.42 3.38 4-16 Differential Reinforcement 1= Unfavor able; 4= Favorable 24.66 4.13 8-32 Peer Drinking Groups 0= Low 1= High 0.43 0.50 0-1 School Connectedness 1= Weak; 4= Strong 22.64 4.82 8-37 Gender 0 = Female ; 1 = Male --0-1 Intact Family 0 = Intact Family; 1= Disrupted Family --0-1 Perceived Supervision 0= Consistent ; 4= Inconsistent 2,12 1,27 0-4 Ethnicity African American 0= Non-African American; 1= African American --0-1 Hispanic 0= Non-Hispanic; 1= Hispanic --0-1 Other Ethnic Groups 0= Non-Othe r; 1= Other Ethnic Groups --0-1 Caucasian (Excluded) 0= Non-Caucasian; 1= Caucasian --0-1
94 Appendix B Antisocial Behavior Scale Statement Factor Loading 1 In the past 12 months, were you in a serious physical fight? 0.46 2 In the past 12 months, did you shoplift or steal something? 0.58 3 In the past 12 months, did you da mage someone elses property? 0.59 4 In the past 12 months, did you carry a weapon for personal protection? 0.49 5 In the past 12 months, did you use marijuana (i.e., weed, pot)? 0.74 6 In the past 12 months, did you use methyl butane (Black Butterfly)? 0.41 7 In the past 12 months, did you us e other drugs (i.e., ecstasy, acid, cocaine, heroin, LSD, shrooms, inhalants)? 0.66 8 Have you ever had a drink of al cohol (more than a few sips)? 0.57 9 In the past 30 days, have you had any alcohol to drink? 0.67 10 In the last year, have you had five or more drinks of alcohol in one day? 0.72 Scale Reliability: Alpha= 0.79 Scale Metric: 0= Low levels of antisocial behavior 10= High levels of antisocial behavior
95 Appendix C Parental Support Scale Statement Factor Loading 1 Overall, I am satisfied with my re lationship with my mother (or female guardian). 0.70 2 I am satisfied with the way my mother (or female guardian) and I communicate with each other. 0.68 3 How close do you feel to your mother (or female guardian)? 0.67 4 How much do you think your mother (or female guardian) cares about you? 0.58 5 Overall, I am satisfied with my relationship with my father (or male guardian). 0.72 6 I am satisfied with the way my father (or male guardian) and I communicate with each other. 0.70 7 How close do you feel to your father (or male guardian)? 0.73 8 How much do you think your father (or male guardian) cares about you? 0.65 9 When my parents/guardians give me advice about drinking alcohol, I usually listen to them. 0.31 10 My parents/guardians have told me how they feel about me drinking alcohol. 0.41 Scale Reliability: Alpha= 0.82 Scale Metric: 6= Weak parental support 40= Strong parental support.
96 Appendix D Maternal Support Scale Statement Factor Loading 1 Overall, I am satisfied with my re lationship with my mother (or female guardian). 0.87 2 I am satisfied with the way my mother (or female guardian) and I communicate with each other. 0.86 3 How close do you feel to your mother (or female guardian)? 0.85 4 How much do you think your mother (or female guardian) cares about you? 0.74 5 My parents/guardians have told me how they feel about me drinking alcohol. 0.41 Scale Reliability: Alpha= 0.78 Scale Metric: 3= Weak maternal support 20= Strong maternal support.
97 Appendix E Paternal Support Scale Statement Factor Loading 1 Overall, I am satisfied with my relationship with my father (or male guardian). 0.88 2 I am satisfied with the way my father (or male guardian) and I communicate with each other. 0.87 3 How close do you feel to your father (or male guardian)? 0.87 4 How much do you think your father (or male guardian) cares about you? 0.77 5 When my parents/guardians give me advice about drinking alcohol, I usually listen to them. 0.42 Scale Reliability: Alpha= 0.82 Scale Metric: 3= Weak paternal support 20= Strong paternal support.
98 Appendix F Emotional Reinforcement Scale Statement Factor Loading 1 I think drinking alcohol helps kids my age deal with being sad or depressed. 0.87 2 I think drinking alcohol helps kids my age feel better when they are upset. 0.86 3 I think drinking alcohol helps kids my age deal with anger. 0.82 4 I think drinking alcohol helps kids my age deal with their problems. 0.81 Scale Reliability: Alpha= 0.86 Scale Metric: 4= Favorable emotional reinforcement. 16= Unfavorable emotional reinforcement.
99 Appendix G Social Reinforcement Scale Statement Factor Loading 1 I think drinking alcohol helps kids my age look cooler. 0.80 2 I think drinking alcohol helps kids my age to be more popular 0.83 3 I think kids who drink alcohol look more mature (grown-up) than kids who dont drink alcohol. 0.69 4 I think drinking alcohol helps kids my age fit in. 0.79 5 I think drinking alcohol helps kids my age feel comfortable at parties. 0.68 Scale Reliability: Alpha= 0.81 Scale Metric: 5= Favorable social reinforcement. 20= Unfavorable social reinforcement.
100 Appendix H Differential Reinforcement Scale Statement Factor Loading 1 If my best friend offered me alc ohol, I would be able to say no. 0,71 2 If someone more popular than me offered me alcohol, I would be able to say no. 0,70 3 If an older brother/sist er offered me alcohol, I w ould be able to say no. 0,72 4 I think drinking alcohol can cau se serious health problems 0,51 5 I think it is easy to get addicted to alcohol. 0,41 6 I think that when kids my age dri nk alcohol, they are more likely to get in an accident. 0,40 7 Kids who take alcohol to school will get caught. 0,39 8 I think that when kids my ag e drink alcohol, they usually get punished. 0,38 Scale Reliability: Alpha= 0.71 Scale Metric: 8= Unfavorable differential reinforcement. 32= Favorable differential reinforcement.
101 Appendix I Neutralizing Definitions Scale Statement Factor Loading 1 It is OK for kids to drink alcohol as long as they dont drink and drive. 0.89 2 It is OK for kids my age to dri nk alcohol if their parents/guardians approve. 0.86 3 It is OK for kids my age to dr ink alcohol during special occasions, such as holidays, weddings and family reunions. 0.83 4 It is OK for kids my age to drink alcohol. 0.83 Scale Reliability: Alpha= 0.87 Scale Metric: 4= Favorable neutrali zing definitions. 16= Unfavorable neutralizing definitions.
102 Appendix J School Connectedness Scale Statement Factor Loading 1 I am happy to be at my school. 0,74 2 I feel like I am part of my school. 0,70 3 The teachers at my school treat students fairly. 0,52 4 How often do you participat e in volunteer activities 0,58 5 How often do you participate in clubs or community groups (Girl Scouts, 4-H) 0,48 6 How often do you participate in school-sponsored activities (band, drama, clubs) 0,52 7 How often do you participate in religious club or activity 0,57 8 How would you describe the grades that you usually get in schools? 0,45 Scale Reliability: Alpha= 0.68 Scale Metric: 8= Weak school connectedness; 37= Strong school connectedness
0 About the Author Jos Ordez received a Bachelor Degree in Psychology in 1990, and a Specialization in Research Methodology in 1993 from the University Rafael Urdaneta (Venezuela). He also obtained a Master of Public Health (2004) and a Graduate Certificate in Social Market ing (2009) from The Univer sity of South Florida. Since 1994, he has been faculty member in the School of Criminology at the University of Los Andes (Venezuela) wher e he has taught underg raduate courses in Social Psychology and Criminology. During 1997, he joined the Criminological Research Center at the University of Los Andes to develop his research interest in social psychology of criminal behavior criminological theories, and interventions strategies in crime prevention. Currently, he is Associ ate Professor (2007) and recently he was designated Research Coordi nator (2008) in the School of Criminology at ULA (Venezuela).