Understanding the Antecedents of Adverse Peer Relationships Among Early Adolescents in the United States: An Ecological Systems Analysis

Youth & Society 2017, Vol. 49(8) 999 –1022

© The Author(s) 2015 Reprints and permissions:

sagepub.com/journalsPermissions.nav DOI: 10.1177/0044118X15569215

journals.sagepub.com/home/yas

Article

Understanding the Antecedents of Adverse Peer Relationships Among Early Adolescents in the United States: An Ecological Systems Analysis

Jun Sung Hong1, Dorothy L. Espelage2, and Paul R. Sterzing3

Abstract This study examines ecological level correlates of adverse peer relationships among early adolescents (ages 12-14). Data analysis was conducted using the National Longitudinal Survey of Youth (NLSY). The sample was drawn from the mother–child data set, which included youth who in 2002 or 2004 were living with their mothers and enrolled in school. Eligible participants responded to at least 1 of the 13 items from the survey and their mothers responded to at least 1 of the 2 items measuring adverse peer relationships at Times 1 (2002/2004) and 2 (2004/2006). Multivariate hierarchical logistic regression was estimated. The presence of a learning disorder and adverse peer relationships at Time 1 (socio-demographics), perceptions of school environment (microsystem), and area of residence and perceptions of safety

1Wayne State University, Detroit, MI, USA 2University of Illinois, Urbana–Champaign, IL, USA 3University of California, Berkeley, CA, USA

Corresponding Author: Jun Sung Hong, School of Social Work, Wayne State University, 4756 Cass Avenue, Detroit, MI 48202, USA. Email: [email protected]

569215YASXXX10.1177/0044118X15569215Youth & SocietyHong et al. research-article2015

1000 Youth & Society 49(8)

(exosystem) were all significantly associated with adverse peer relationships at Time 2. Assessing and targeting these ecological levels hold the potential to decrease adverse peer relationships among early adolescents.

Keywords bullying, early adolescence, ecological framework, adverse peer relationships

Peer relationships are important in early adolescents’ social and emotional development. Early adolescents spend a great deal of time with their peers during this period and learn important social and communication skills (Murphy, 2002; Sidorowicz & Hair, 2009). Adverse peer relationships (e.g., bullying, peer hostility, mutual disagreements) impact the acquisi- tion of social and communication skills (Laursen, 1993), as these experi- ences are a common form of social exchange (Laursen, Hartup, & Koplas, 1996) that arise when people with incompatible goals use a variety of strat- egies to influence one another (Malloy & McMurray, 1996). Adverse peer relationships are conceptualized as mutual disagreement or hostility between peers or among peer groups (Noakes & Rinaldi, 2006) that may or may not escalate into physical aggression (Sidorowicz & Hair, 2009). The management of adverse peer relationships during adolescence is criti- cal to the development and functioning of interpersonal relationships (Laursen et al., 1996).

This is an emerging area of study and has considered relatively few indi- vidual characteristics. Psychoanalytic, socio-biological, and cognitive-devel- opmental theories suggest differences in individual maturation account for variations in adverse peer relationships. These theories have minimized the influences of other types of relationships (e.g., familial) and contexts (e.g., school, neighborhood) on the development of adverse peer relationships (Laursen & Collins, 1994). Therefore, our study investigates adolescent adverse peer relationships within individual, relational, and environmental contexts.

Most research so far has focused on aggression such as bullying (Hong & Espelage, 2012). Less attention has been paid to non-aggressive forms of adverse peer relationships such as conflicts or difficulty getting along with peers. Even fewer studies have used a national sample to examine multiple, contextual factors associated with both aggressive and non-aggressive types of bullying. Understanding the confluence of factors associated with adverse peer relationships is important, because these affiliations establish a template for future links that can hinder access to a primary source of social support

Hong et al. 1001

during mid to late adolescence (Mikami, Szwedo, Allen, Evans, & Hare, 2010). Longitudinal research suggests continuity in patterns of interpersonal relationships over time and across various social contexts (Stocker & Richmond, 2007).

Our study draws upon Bronfenbrenner’s (1977) ecological framework. Bronfenbrenner posited that individuals are a part of four interrelated, nested system levels: micro- (immediate social environment), meso- (interaction between and among microsystems), exo- (settings not directly affecting the individual but influencing the microsystem), and macro- (broader societal and cultural influences on the other systems) levels. This process can be best understood as examining the socio-demographic, family, peer, school, and neighborhood factors that impact adverse peer relationships. This article investigates those factors as affecting early adolescents. Furthermore, it examines micro- and exosystems antecedents of adverse peer relationships. The literature reviewed comes primarily from bullying research.

Socio-Demographic Characteristics

Age Adverse peer relationships among adolescents differ by age. Children experi- ence important behavioral and interpersonal changes during early adoles- cence, as they transfer from elementary to middle school (Veronneau & Dishion, 2010). Among children in elementary school, these relationships are often the result of aggressive forms of behavior, such as physical bullying (e.g., pushing, hitting; Alexander & McConnell, 1993). In contrast, early ado- lescents entering middle school exhibit further relational behaviors (e.g., social exclusion, clique formation, gossip and rumor spreading) that can lead to non-aggressive forms of adverse peer relationships (Pellegrini & Bartini, 2000; Ray & Cohen, 2000; Smith, Madsen, & Moody, 1999).

Race/Ethnicity Researchers have also become interested in exploring peer relationships among adolescents from race/ethnic minorities. However, it is notably sparse, and focuses mostly on peer victimization, with inconsistent findings. Juvonen, Graham, and Shuster (2003) found that Latino sixth graders were less likely to experience peer victimization than were White, Black, or Asian youth. However, other researchers reported greater frequency among Latinos (Nansel et al., 2001) or Blacks (Storch, Brassard, & Masia-Warner, 2003) than among youth of other racial/ethnic groups.

1002 Youth & Society 49(8)

Gender Research examining gender differences in experiences of bullying and peer conflicts has also produced inconsistent findings. One study found that boys have more peer conflicts than girls (P. Miller, Danaher, & Forbes, 1986). Similar findings were observed in Black’s (2000) study, which reported that boys argue more frequently than girls. However, others found that peer dis- agreements and relational conflicts were more frequent among girls, while conflicts related to status/dominance were more frequent among boys (Laursen, 1993; Noakes & Rinaldi, 2006).

Learning Disorder A small number of studies have considered other socio-demographic charac- teristics, including the effects of learning disorders (C. A. Rose & Espelage, 2012). These studies have reported an association between learning disorders and adverse peer relationships, particularly bullying. Kaukiainen et al. (2002) explored the associations between bullying, learning skills, social intelli- gence, and self-concept among fifth graders. Their findings were consistent with the notion that bullying was prevalent among children with learning disorders. Youth with a learning disorder may have difficulty interpreting verbal and nonverbal communications, and possess poor social skills, ham- pering their efforts to succeed academically and socially. These children can also have impulsive behavioral tendencies (C. A. Rose, Monda-Amaya, & Espelage, 2011; Whitney, Smith, & Thompson, 1994), which may predispose them to peer conflicts (Kaukiainen et al., 2002).

Poverty Status Poverty can also lead to negative developmental outcomes (G. J. Duncan & Brooks-Gunn, 1997). For example, positive behavior development appears to be compromised for children living in poverty (Eamon, 2001a, 2001b; Eamon & Zuehl, 2001), contributing to adverse peer relationships (Civita, Pagani, Vitaro, & Tremblay, 2007). Poverty has a corrosive effect on social resources in the community (e.g., supportive family and peers), which, in turn, affects interpersonal relations. Poverty creates social disorganization and reduces social controls over conflicts (e.g., lack of effective sanctions; Kawachi & Kennedy, 2002). Thus, some youth living in poverty experience associated stressors (e.g., maternal stress, displacement) and are less likely to receive empathy and nurturance from their mothers, while these factors tend to miti- gate adverse peer relationships (Curtner-Smith et al., 2006).

Hong et al. 1003

Microsystem Antecedents Bronfenbrenner purported that “[i]t is within the immediate environment of the microsystem that proximal processes (e.g., caregiver-child and peer-peer relationships) operate to produce and sustain development” (Bronfenbrenner, 1994, p. 39). The microsystem is a pattern of activities, social roles, and inter- personal relations experienced by the individual in a direct setting (e.g., home, school; Bronfenbrenner, 1977). A review of the literature suggests par- enting, peer influence, and school relationships are microsystem-level ante- cedents of adverse peer relationships for early adolescents.

Parenting Developmental psychologists have established that parenting is significant in the development of youth’s peer relationships outside the home (Harris, 2000). The association of authoritarian parenting practices with bullying and aggressive peer interactions has been established in numerous studies (Baldry & Farrington, 1998; Curtner-Smith et al., 2006). In a sample of mothers and children attending a Head Start program, Curtner-Smith et al. (2006) found that authoritative parenting was correlated with relational bullying. Baldry and Farrington (1998) also reported a significant association between author- itative parenting and bullying and victimization from a sample of 238 stu- dents (ages 11-14) in a middle school in Rome. Bullying also appears to be more frequent among youth whose parents are characterized as physically harsh, rejecting, or uninvolved (R. D. Duncan, 2004; Holt & Espelage, 2007; Spriggs, Iannotti, Nansel, & Haynie, 2007).

Negative Peer Influence The association between peer influence and the development and mainte- nance of behavior problems has been well-established empirically (Werner & Crick, 2004). Peer influence can also affect the quality of peer relations and socialization among children. Research also indicates that peer group influ- ences play a significant role in promoting or inhibiting bullying, particularly among early adolescents (Espelage, Bosworth, & Simon, 2000; Espelage, Holt, & Henkel, 2003). From a sample of 558 students (Grades 6-8) in a large middle school located in a midwestern metropolis, Espelage et al. (2000) reported that involvement with delinquent peers was predictive of bullying. Peers influence and socialize one another into adverse peer relationships and conflicts through modeling and reinforcement, and through coercing youth into engaging in these behaviors (Dishion, Patterson, & Griesler, 1994).

1004 Youth & Society 49(8)

School Relationships Negative peer influences frequently occur in school settings, which might influence their school relationships outside of their peer group (e.g., relation- ship with teachers). School relationships include communication patterns, role relationships and perceptions, and patterns of influence (Tobin & Sprague, 2000; Welsh, Stokes, & Greene, 2000). Although negative peer influences can increase the risk of misbehavior and conflicts, positive rela- tionships with teachers perceived as caring and with students feeling at ease in making friends can also reduce adverse peer relationships (Silver, Measelle, Armstrong, & Essex, 2005). This is not surprising, given that teachers can have a significant impact on the quality of students’ peer relationships (Lee, 2010). However, there appears to be a dearth of empirical evidence to support this claim (Rodkin & Hodges, 2003). Lee’s (2010) findings from a sample of 1,238 students (ages 13-17) in six South Korean middle schools suggest that students who perceived that teachers cared were less likely to be bullies. In contrast, Fekkes, Pijpers, and Verloove-Vanhorick (2005) found from a sam- ple of 2,766 children (ages 9-11) in 32 Dutch elementary schools that teach- ers’ involvement aggravated bullying behaviors among these children.

Close friendships are emotional resources which may lead to better adjust- ment and development (Hartup, 1992). Research has identified friendships as mitigating factors for bullying and negative peer relationships (Bollmer, Milich, Harris, & Maras, 2005; Bukowski, Hoza, & Boivin, 1994). Friendships serve many functions, such as informing individuals of their values, promot- ing the acquisition of social skills, and providing a protective buffer against relationship problems (Bukowski et al., 1994). Youth without friends misbe- have and have adverse peer relationships. Using a sample of 99 children (ages 10-13), Bollmer et al. (2005) reported that friendships attenuated the link between externalizing and bullying behaviors.

Exosystem Antecedents Microsystem-level influences and interactions, such as parenting, peers, and schools are nested within an exosystem. Neighborhood is an exosystem-level factor that might place youth at risk of exhibiting behavior problems and adverse peer relationships (Low & Espelage, 2014; Tolan, Gorman-Smith, & Henry, 2003). For example, lack of neighborhood resources, particularly in urban areas, can directly affect the quality of adolescents’ peer relationships (Bronfenbrenner, 1977; Low & Espelage, 2014). Moreover, individual, fam- ily, and school risk and protective factors can co-occur or are affected by conditions in the neighborhood, which can increase or decrease bullying and

Hong et al. 1005

problematic peer relationships (Salzinger, Feldman, Stockhammer, & Hood, 2002). For example, Low and Espelage (2014) found that exposure to neigh- borhood violence increased rates of bullying. Social disorganization theorists have also hypothesized that adolescents residing in dangerous urban neigh- borhoods are exposed to antisocial behaviors and criminal activities that can model and reinforce aggressive reactions and increase adverse peer relation- ships (Sampson, 2012).

Study Aims This study addresses an important gap in the literature by examining factors associated with aggressive and non-aggressive forms of adverse peer rela- tionships. Utilizing a longitudinal, nationally representative data set of early adolescents, the current study examines the influence of socio-demographic characteristics (poverty status, learning disorders, prior adverse peer relation- ships), microsystem factors (parenting, school climate, peer influence), and exosystem factors (area of residence, neighborhood safety) on the likelihood of having adverse peer relationships. The socio-demographic variables were treated as control variables, except for poverty status, the presence of a learn- ing disorder, and adverse peer relationships at Time 1 (T1). The study posits the following: (a) poverty status, presence of a learning disorder, and adverse peer relationships at T1 is associated with an increase in adverse peer rela- tionships at Time 2 (T2); (b) negative peer influence and residing in an urban, central city area are associated with an increase in adverse peer relationships at T2; and (c) positive parenting, positive school climate, and neighborhood safety will be associated with a decrease in adverse peer relationships at T2.

Method

Data and Sample The study utilizes data from the National Longitudinal Survey of Youth (NLSY), which are publicly available, de-identified data sets. The initial sur- vey of 12,686 individuals who were 14 to 22 years was conducted in 1979. The NLSY79 contains information about education, training, employment, and family experiences of the respondents (Center for Human Resource Research, 2004). The original NLSY oversampled Blacks, Hispanics, and Whites who are economically disadvantaged. From 1979 through 1994, respondents were interviewed annually and interviewed biennially thereafter (1996-present). In 1986 and beyond, assessments of the NLSY female respondents and their children were conducted. The assessments measure the

1006 Youth & Society 49(8)

children’s cognitive ability, temperament, motor and social development, behaviors, competence, and home environment (Center for Human Resource Research, 2004). Starting in 1986, youth between the ages of 10 and 15 were interviewed using a self-administered survey, which collected information on factors related to parenting, peers, school, and neighborhoods.

The sample for the current study was drawn from the mother–child data set of NLSY who completed surveys in 2002 or 2004 (T1) and 2004 or 2006 (T2). The sample included youth who were 10 to 12 years of age, resided with their mothers, were enrolled in regular school, responded to at least 1 of the 13 items from the self-administered survey germane to this study, and had mothers who responded to at least 1 of the 2 items that measured adverse peer relationships at both T1 and T2.

Socio-demographic information was only collected on biological mothers and their households. Youth living with other caregivers were excluded. The sample contained siblings, which could lead to biased estimates if techniques are not used to handle clustered data. Although the NLSY data allow for iden- tifying siblings in the same family, it is not possible to identify whether the youth were attending the same school. To address this potential bias, one youth from each family who met the selection criteria was randomly selected for inclusion. The selection criteria resulted in a sample of 733 youth who were 10, 11, or 12 years old when they entered the sample at T1 (2002 or 2004).

Measures Dependent variables. Aggressive and non-aggressive forms of adverse peer rela- tionships were measured at T2 (2004 or 2006) using two items from the Antiso- cial Behavior Subscale of the Behavior Problem Index (BPI; Center for Human Resource Research, 2004): “child bullies or is cruel/mean to others” and “child has trouble getting along with other children.” Mothers rated these items as 3 = “often true,” 2 = “sometimes true,” or 1 = “not true” in the previous 3 months.

The two items were correlated at r = .70, and were summed and collapsed into a three-category variable to create an overall measure that captured both aggressive and non-aggressive adverse peer relationships. The new variable, however, failed to meet the assumption of proportional odds assumption for conducting ordinal regression. Therefore, the dependent variable was dichot- omized into 0 (mothers’ responded “not true” to both items) and 1 (mothers’ responded “somewhat true” or “often true” to at least one of the two items). Multivariate logistic regression was used to estimate the models.

Independent variables. Socio-demographic characteristics of youth and mother participants were controlled for in each logistic regression model.

Hong et al. 1007

Microsystem (parenting, peer, and school) and exosystem (neighborhood) variables were also entered in each model to assess the influence of ecologi- cal factors on adverse peer relationships. These variables were all measured at T1. As the ecological framework suggests, socio-demographic characteris- tics of the youth and mother can affect microsystem factors in the home and school and exosystem factors in the neighborhood, increasing or decreasing the odds of adverse peer relationships.

The socio-demographic characteristics of youth were (a) age in months (120-144); (b) race/ethnicity, based on the mothers’ racial/ethnic identifier (Black, Hispanic, White [reference category]), gender (female used as refer- ence category); and (c) presence of a learning disorder, based on mothers’ responses to whether the child has a learning problem/disability, dyslexia, reading, or speech problem (1 = yes and 0 = no).

Socio-demographic characteristics of the mother and household included (a) mothers’ marital status (never married; other [divorced, separated]; mar- ried, spouse present [reference variable]), (b) mothers’ educational status (high school, more than high school, less than high school [reference vari- able]), and (c) poverty status (last year before the interview). Poverty status was defined by a NLSY-constructed variable, using the Federal definition of poverty (1 = “in poverty”; 0 = “not in poverty”). The final control variable was adverse peer relationships measured at T1 (2 years prior).

The microsystem level refers to immediate environments (e.g., home and school) in which youth interact with family, peers, and teachers. Parenting was measured using items from the Home Observation for Measurement of the Environment–Short Form (HOME-SF) Inventory (Caldwell & Bradley, 1984). Based on maternal report and interviewer observations, the HOME-SF Inventory measured the nature and quality of children’s home environment from birth to adolescence, including age-appropriate cognitive stimulation and emotional support subscales for children, ages 0 to 15 years. Cognitive stimulation scale items include outings, reading, playing, and other parent– youth interactions. The emotional support subscale items include family rela- tionships and disciplining (e.g., spanking). The raw scores for both scales were normalized so that a one-unit change in the variable represents a one standard deviation change in the outcome variables (Zimmerman, Glew, Christakis, & Katon, 2005).

Negative peer influence was assessed with five items asking youth whether they felt pressured from friends to “try cigarettes,” “try alcohol,” “try marijuana/other drugs,” “skip school,” and “commit crime/violence” (1 = yes, 0 = no). A count variable was created across the five items and three categories were developed: no peer influence (0), moderate (1-2), and high (3-5). Perceptions of school climate were assessed with two

1008 Youth & Society 49(8)

items asking the youth about teacher involvement (“most of the teachers are willing to help with personal problems”) and ease of making friends (“it’s easy to make friends at this school”). Response options were on a 4-point Likert-type scale (1 = not at all true, 2 = not too true, 3 = some- what true, 4 = very true). However, due to low frequencies, two of the categories (i.e., “not at all true,” “not too true”) were collapsed. A Principal Components Analysis using PRINQUAL and PRINCOMP pro- cedures was conducted on these two items. However, due to low internal consistency (α = .51), the school climate items were entered separately into the statistical models.

Exosystem variables included are youth’s responses to neighborhood safety (“how safe do you feel walking and playing in your neighborhood”; 1 = not at all safe, 2 = somewhat safe, 3 = reasonably safe, 4 = very safe) and area of residence (e.g., urban, non-urban). Neighborhood safety was collapsed into a three-level variable due to low frequency responses for “not at all safe” and “somewhat safe.” Youth’s area of residence was assessed as one of three types: “not in a standard metropolitan statistical area (SMSA)”; “in a SMSA, not in a central city”; and “in a SMSA, in a central city” (reference variable). These variables were defined by NLSY, according to data from the Census Bureau (U.S. Census Bureau, 2011). SMSA includes a core urbanized area of at least 50,000 residents and adjacent communities that have a high degree of economic and social integration with that core area ( National Longitudinal Surveys n.d.).

Missing data. Approximately half of the respondents (51.84%) had no miss- ing data on any of the variables. However, 196 cases (26.74%) had data miss- ing on at least one variable, and one case (0.14%) had data missing on eight variables. Regarding specific variables, negative peer influence had the high- est number of cases missing (n = 117, 15.96%), while presence of a learning disorder and mothers’ educational status both had the lowest number of cases missing (n = 1, 0.14%). Because missing data or non-response can produce a threat to the validity of inference (Shadish, Cook, & Campbell, 2002), miss- ing data were addressed using the imputation methods available in SAS 9.1. PROC MI and PROC MIANALYZE were used, incorporating procedures suggested by other researchers, for imputing data for categorical variables (R. A. Rose & Fraser, 2008). The MI procedure replaces missing values with values repeatedly drawn from conditional probability distributions by using the Markov Chain Monte Carlo simulation method. The five implicates that were created using the PROC MI procedure were combined, using the MIANALYZE procedure to generate valid statistical inferences (R. A. Rose & Fraser, 2008).

Hong et al. 1009

Data Analysis SAS 9.1 was used to conduct the data analyses. Weighted descriptive statis- tics (means and standard deviations, percentages) for all of the variables were calculated. For a multivariate logistic model, each odds ratio (OR) can be interpreted as the effect of each variable on the odds of having adverse peer relationships, adjusted for the effects of the other independent variables (Allison, 2001). Multivariate models were not weighted (Center for Human Resource Research, 2004). Controlling for poverty status and race/ethnicity accounted for the oversampling of participants.

Any relationship found between the independent variables and adverse peer relationships T2 may possibly be the result of selection bias. That is, the associations are caused by some unmeasured characteristics not controlled for in the analysis and can result in biased estimates. To adjust for this, in addition to placing the socio-demographic variables into the models that might be related both to the systems level factors and to the adverse peer relationships, residualized change models (lagged dependent variable or regressor variable methods) were estimated (Berger, Bruch, Johnson, James, & Rubin, 2009). In the residualized change model, adverse peer relationships T1 was entered into the multivariate logistic regression models. This method adjusted for persistent characteristics of youth (e.g., genetic factors) that are assumed to have consistent effects on adverse peer relationships at both T1 and T2. The estimates should then be less subject to bias than those estimated with traditional multivariate logistic models.

Consistent with the ecological model, the effects of four groups of vari- ables on adverse peer relationships were investigated by estimating three hierarchical logistic models. The first model included variables measuring the socio-demographic characteristics, followed by adding the variables mea- suring the microsystem and exosystem. To assess model improvement when variables were entered, whether adding each group of variables and interac- tions among variables, we subtracted the −2 × log likelihood (−2LL) value for the model including the additional ecological systems variables from the −2LL value for the previous model. Whether there was a significant differ- ence in the −2LL between the two models was determined with the differ- ences in the degrees of freedom between the two models utilizing a chi-square table. A significant decrease in −2LL indicated a better fitting model.

Results The average age of the youth was 132.20 months (11.02 years). The majority of youth were White (54.30%), followed by Black (26.33%) and Hispanic

1010 Youth & Society 49(8)

(19.37%). Slightly over half of the sample was male (51.02%), and approxi- mately 5% were identified as having a learning disorder. Among the mothers, 68.22% were married (spouse present), 22.78% were divorced or separated, and 9.0% were never married. In terms of educational status, few mothers had less than a high school education (8.43%), compared with having a high school education (27.72%) and more than a high school education (63.85%). Approximately 10% of the families experienced poverty the year prior to the interview. Approximately 19% of youth had adverse peer relationships at T1.

At the microsystem level, the average for the cognitive stimulation sub- scale was 101.88 (range = 37.40-120.80) and the emotional support subscale was 100.97 (range = 41.30-123.30). Although the majority of youth reported no negative peer influences (90.47%), such as pressure from friends to engage in illegal behavior and to skip school, 4.25% experienced one to two types, and 5.28% experienced three to five types. For teacher involvement, the responses ranged from a low of approximately 14% for “not at all true/not too true” to a high of approximately 54% for “very true,” suggesting that the majority of these youth perceived that teachers helped them with personal problems. Similarly, a low percentage (9.77%) reported “not too true or not at all true” that it was easy to make friends at their schools, while the remain- ing youth reported “somewhat or very true.”

At the exosystem level, the majority of the youth (54.75%) (Table 1) per- ceived their neighborhood as “very safe.” This was followed by 28.50% who felt their neighborhood was “reasonably safe,” and 16.74% who felt their neighborhood was “somewhat or not at all safe.” In regards to area of resi- dence, 65.28% of the youth resided in a SMSA, not central city; 21.66% in a SMSA, in central city; and 13.06% did not reside in a SMSA.

Model 1 indicated that the presence of a learning disorder and adverse peer relationships at T1 were associated with an increased adjusted odds of having adverse peer relationships at T2. Youth with a learning disorder had more than 3 times the risk of having adverse peer relationships 2 years later than did those with no learning disorder (OR = 3.22, p < .01). Youth with adverse peer relationships at T1 were 7.92 times more likely to have adverse peer relationships 2 years later (OR = 7.92, p < .001) than were those without adverse peer relationships at T1.

The microsystem variables included in Model 2 resulted in a significant improvement of fit over Model 1 (change in −2LL = 12.43, df = 5, p < .05). Both learning disorder and adverse peer relationships at T1 remained signifi- cant. Only one microsystem variable was associated with adverse peer rela- tionships at T2. Youth who perceived it was easy to make friends at their school were less likely to experience adverse peer relationships at T2 (OR = −.45, p < .01).

Hong et al. 1011

Table 1. Weighted Means (SD) or Percentages of the Sample (N = 733).

Variable % M SD

Dependent variable Adverse peer relations at T2 Not at all true 85.18 Somewhat true/often true 14.82 Independent variables Socio-demographics Age in months (range = 120-144) 132.20 6.81 Race/ethnicity Hispanic 19.37 Black 26.33 White 54.30 Gender Male 51.02 Female 48.98 Learning disorder No 94.91 Yes 5.09 Mothers’ marital status Never married 9.00 Married, spouse present 68.22 Other (divorced, separated) 22.78 Mothers’ educational status Less than high school 8.43 High school 27.72 More than high school 63.85 Poverty status No 90.36 Yes 9.64 Adverse peer relations at T1 Not at all true 81.26 Somewhat true/often true 18.74 Microsystem Parenting (HOME scale) Cognitive stimulation (range = 37.40-120.80) 101.88 15.95 Emotional support (range = 41.30-123.30) 100.97 15.61 Negative peer influence No peer influence 90.47 Moderate (1-2) 4.25 High (3-5) 5.28

(continued)

1012 Youth & Society 49(8)

When the exosystem variables were added to Model 3 (change in −2LL = 11.17, df = 3, p < .01), learning disorder and adverse peer relationships at T1 remained significant. Youth’s neighborhood safety (OR = .73, p < .01), resi- dence not in a SMSA (OR = 2.64, p < .01), and in a SMSA, not central city (OR = 1.95, p < .01), compared with residence in a SMSA, central city, were associated with adverse peer relationships T2. Youth who did not reside in a SMSA and youth who resided in a SMSA, not central city were both approxi- mately 2 times more likely to experience adverse peer relationships at T2, in comparison with youth who resided in a SMSA, central city (Table 2).

Discussion Consistent with our hypothesis, we found that the presence of a learning dis- order and adverse peer relationships at T1 were significantly related to

Variable % M SD

Perceptions of school climate Teacher involvement Not too true, not at all true 14.32 Somewhat true 31.96 Very true 53.72 Ease of making friends Not too true, not at all true 9.77 Somewhat true 33.37 Very true 56.86 Exosystem Neighborhood Neighborhood safety Somewhat safe, not at all safe 16.74 Reasonably safe 28.50 Very safe 54.75 SMSA residence Not in SMSA 13.06 SMSA, not central city 65.28 SMSA, in central city 21.66

Note. Percentages for some variables do not add up to 100% due to rounding error. HOME = Home Observation for Measurement of the Environment; SMSA = standard metropolitan statistical area.

Table 1. (continued)

Hong et al. 1013

Table 2. Hierarchical Logistic Regression of Adverse Peer Relations (N = 733).

Model 1 Model 2 Model 3

Variable B (SE) Exp(B)

OR B (SE) Exp(B)

OR B (SE) Exp(B)

OR

Socio-demographics Age 0.00 (.02) 1.00 0.00 (.02) 1.00 0.00 (.02) 1.00 Race/ethnicity (White) Hispanic −0.43 (.35) 0.65 −0.48 (.37) 0.62 −0.33 (.38) 0.72 Black 0.25 (.28) 1.28 0.20 (.30) 1.22 0.39 (.32) 1.48 Gender (female) Male 0.05 (.24) 1.05 0.11 (.24) 1.12 0.15 (.25) 1.16 Learning disorder 1.17** (.39) 3.22 0.97** (.41) 2.64 1.01** (.25) 2.75 Mothers’ marital status Never married 0.19 (.44) 1.21 −0.03 (.46) 0.97 −0.15 (.48) 0.86 Other 0.01 (.31) 1.01 −0.14 (.34) 0.87 −0.12 (.35) 0.89

Mothers’ educational status High school −0.37 (.38) 0.69 −0.24 (.39) 0.79 −0.37 (.40) 0.69 More than high school −0.60 (.38) 0.55 −0.41 (.40) 0.66 −0.44 (.41) 0.64 Poverty status 0.07 (.39) 1.07 −0.05 (.39) 0.95 0.05 (.40) 1.05 Adverse peer relations at T1 2.07*** (.23) 7.92 1.96*** (.24) 7.10 2.05*** (.25) 7.77 Microsystem Parenting (HOME scale) Cognitive stimulation −0.00 (.01) 1.00 −0.00 (.01) 1.00 Emotional support −0.01 (.01) 0.99 −0.01 (.01) 0.99 Negative peer influence 0.12 (.20) 1.13 0.03 (.21) 1.03 Perceptions of school climate Teacher involvement −0.01 (.16) 1.00 0.01 (.17) 1.01 Ease of making friends −0.45**(.18) 0.64 −0.35† (.19) 0.70 Exosystem Neighborhood Neighborhood safety −0.31** (.14) 0.73 Not in SMSA 0.97** (.42) 2.64 SMSA, not central city 0.67** (.30) 1.95 −2LL 511.55 499.12 487.95 df 11 16 19

Note. Reference categories are denoted in parentheses. For Model 2, change in −2LL = 12.43, df = 5, p < .03; and for Model 3, change in −2 LL = 11.17, df = 3, p < .01. −2LL was averaged for the five implicates for each model. OR = odds ratio; HOME = Home Observation for Measurement of the Environment; SMSA = standard metropolitan statistical area; LL = log likelihood. †p < .10. *p < .05. **p < .01. ***p < .001.

adverse peer relationships at T2. This finding is consistent with past research, which indicates that early adolescents with a learning disorder are more likely to display bullying and experience adverse peer relationships (Carter & Spencer, 2006; Kaukiainen et al., 2002; Mishna, 2003; Nabuzoka, 2003; C.

1014 Youth & Society 49(8)

A. Rose & Espelage, 2012). Carter and Spencer (2006), for example, pro- vided a review of research literature on the link between bullying and stu- dents with learning disorders. Their findings indicated that students with a learning disorder experienced bullying and victimization more frequently than their peers without a learning disorder. Children with a learning disorder are more likely to display poor social and communication skills and impul- sive behavioral tendencies resulting from social isolation due to rejection by their peers, which heightens their risk of involvement in adverse peer rela- tionships (Kavale & Forness, 1995; C. A. Rose et al., 2011; Whitney et al., 1994).

In addition, early adolescents who previously had adverse peer relation- ships were more than 7 times more likely to experience the same problem 2 years later. Research has reported that children with adverse peer relation- ships at an early age are at greater risk of these behaviors later in life (Campbell, Shaw, & Gilliom, 2000). Children who frequently experienced adverse peer relationships early in life were, if untreated, likely to continue having these problems at school (Campbell et al., 2000). Contrary to our hypothesis, poverty status was not associated with adverse peer relationships at T2, which is also inconsistent with previous research (Civita et al., 2007). One possible explanation is that these studies only examined poverty status in the last year, which might not adequately detect the full range of economic difficulties.

Inconsistent with our hypotheses and previous findings (Silver et al., 2005), we found no evidence that parenting practices, negative peer influ- ence, or teacher involvement influence early adolescents’ adverse peer rela- tionships. One possible explanation is that adverse peer relationships were measured using mothers’ reports on only two questions, rather than direct observation, normed scales, or reports from peers and teachers, which may have resulted in unmeasured biases. Moreover, early adolescence is a period in which youth rely less on their caregivers for emotional support (Ayyash- Abdo, 2002), which might account for the lack of association between mater- nal cognitive stimulation and emotional support on adverse peer relationships. Interestingly, in the second model, we found that early adolescents who per- ceived their school climate as one in which students could make friends eas- ily were less likely to experience adverse peer relationships at T2. Although this variable was no longer significant when the exosystem variables were entered into the third model, the statistical significance of ease of making friends was consistent with previous study findings. For instance, Hartup (1992) reported that feeling that it was easy to make friends could result in better social adjustment and a lower likelihood of adverse peer relationships.

Hong et al. 1015

All of the neighborhood variables at the exosystem level were associated with adverse peer relationships at T2. Consistent with our hypothesis, youth who feel safer in their neighborhoods were at lower risk of adverse peer rela- tionships. Although research on the relation between neighborhood factors and adverse peer relationships is scant, our finding was consistent with accu- mulating evidence that exposure to neighborhood violence and unsafety escalates into maladaptive and antisocial behaviors (e.g., Bacchini, Miranda, & Affuso, 2011; Copeland-Linder, Lambert, Chen, & Ialongo, 2011; Low & Espelage, 2014) and adverse peer relationships, such as bullying (Khoury- Kassabri, Benbenishty, Astor, & Zeira, 2004; Schwartz & Gorman, 2003; Schwartz & Proctor, 2000). Our finding is significant, given that few studies have examined how youth’s adverse peer relationships, such as bullying, are influenced by their experiences and exposures outside of school (Low & Espelage, 2014). Exposure to neighborhood violence can compromise emo- tional regulation and impulse control (Schwartz & Proctor, 2000), which may preclude or diminish opportunities for developing and maintaining positive peer relations (Low & Espelage, 2014). Contrary to our hypotheses, we found that early adolescents living in areas other than a central city are more likely than those living in a central city to have adverse peer relations 2 years later. This might be the result of mothers living in central cities reporting fewer children’s behavior problems because they were unaware of these behaviors, which may have led to different standards for judging behaviors and peer relations.

This study has limitations, many related to the available data. The first is the measures of the dependent variable, which came from only two BPI items and relied on mothers’ assessments, rather than on multiple informants (i.e., self, peer, and teacher reports). Using multiple items and items from other validated scales could have yielded greater accuracy. Another limitation is also related to the dependent variable. A youth could have difficulty getting along with his or her peers due to chronic victimization, lack of social or communication skills, access to other students (e.g., self-contained settings), and so on. The item “child bullies or is cruel/mean to others” addresses a learned behavior, whereas the item “child has trouble getting along with other children” could represent either a direct or learned behavior or a situation that is a manifestation of the youth’s cognitive or functional abilities or the youth’s environment. Furthermore, because the proportional odds assumption was not met for many of the models when ordinal regression models were esti- mated, the dependent variable was dichotomized and logistics regression models were estimated. These processes do not allow for examining the degree of adverse peer relationships exhibited by the early adolescents. In addition, the presence of a learning disorder is a broad variable that is

1016 Youth & Society 49(8)

difficult to define by the selected variables. To illustrate, a student with one particular learning disorder (e.g., speech problems) may have average or above average functioning in other domains (e.g., cognitive). The absence of mesosystem and macrosystem variables is also a serious limitation. Furthermore, this study did not control for other parent-related factors (e.g., mothers’ psychological health), which might influence adverse peer relation- ships (Goodman et al., 2011). Finally, the results can only be generalized to youth living with their mothers.

These limitations aside, the current findings suggest implications for future research. Because much of scholarship on adverse peer relationships among adolescents has focused on bullying, this study contributes by concep- tualizing and examining a broader range of aggressive and non-aggressive forms of adverse peer relationships. Furthermore, little is known about the mesosystem factors such as parents’ school involvement. Home and school represent two primary systems in youth’s lives (Sheridan, Warnes, & Dowd, 2004). Our findings also suggest the importance of considering neighbor- hood influences. Future studies might collect data on or utilize a data set with additional measures of neighborhood characteristics and environments to determine how they might contribute.

In sum, adverse peer relationships among early adolescents are a mul- tifaceted problem. This study demonstrates the utility of the ecological framework in this research area and points us toward the importance of (a) addressing early youth exposure to adverse peer relationships, (b) dis- entangling the specific aspects of learning disorders (e.g., social skills, language abilities, emotional dysregulation) that increase the likelihood of this problem, and (c) impacting perceptions of school climate and neighborhood safety. Bronfenbrenner’s (1977) ecological theory can serve as a viable tool for enhancing our understanding of adolescent peer relationships.

Authors’ Note This article is based on the first author’s dissertation, written under the guidance of Drs. Mary Keegan Eamon (chair), Dorothy L. Espelage, Wynne Sandra Korr, and Joseph P. Ryan. Both Dorothy L. Espelage and Paul R. Sterzing contributed equally to the article.

Acknowledgments The first author wishes to express his deepest gratitude to Mary Keegan Eamon, Associate Professor Emerita, for her guidance and support, which contributed tremen- dously to this article. The first author also wishes to thank Mr. Al Acker for proofread- ing this article.

Hong et al. 1017

Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The author(s) received no financial support for the research, authorship, and/or publi- cation of this article.

References Alexander, J., & McConnell, S. C. (1993). Children as peacemakers: Promoting

the development of cooperation and conflict resolution. In V. K. Kool (Ed.), Nonviolence: Social and psychological issues (pp. 107-128). Lanham, MD: University Press of America.

Allison, P. D. (2001). Logistic regression using the SAS Institute, Inc.(SAS) system: Theory and application. Cary, NC: SAS Institute.

Ayyash-Abdo, H. (2002). Adolescent suicide: An ecological approach. Psychology in the Schools, 39, 459-475.

Bacchini, D., Miranda, M. C., & Affuso, G. (2011). Effects of parental monitoring and community violence on antisocial behavior and anxiety/depression among adolescents. Journal of Interpersonal Violence, 26, 269-292.

Baldry, A. C., & Farrington, D. P. (1998). Parenting influences on bullying and vic- timization. Legal and Criminological Psychology, 3, 237-254.

Berger, L. M., Bruch, S. K., Johnson, E. I., James, S., & Rubin, D. (2009). Estimating the “impact” of out-of-home placement on child well-being: Approaching the problem of selection bias. Child Development, 80, 1856-1876.

Black, K. A. (2000). Gender differences in adolescents’ behavior during conflict reso- lution tasks with best friends. Adolescence, 35, 499-512.

Bollmer, J. M., Milich, R., Harris, M. J., & Maras, M. A. (2005). A friend in need: The role of friendship quality as a protective factor in peer victimization and bul- lying. Journal of Interpersonal Violence, 20, 701-712.

Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, 513-531.

Bronfenbrenner, U. (1994). Ecological models of human development. In T. Husen & T. N. Postlethwaite (Eds.), The international encyclopedia of education (2nd ed., pp. 1643-1647). New York, NY: Elsevier Sciences.

Bukowski, W. M., Hoza, B., & Boivin, M. (1994). Measuring friendship quality dur- ing pre- and early adolescence: The development and psychometric properties of the Friendship Qualities Scale. Journal of Social and Personal Relationships, 11, 471-484.

Caldwell, B., & Bradley, R. (1984). Home observation for measurement of the envi- ronment. Little Rock: University of Arkansas.

Campbell, S. B., Shaw, D. S., & Gilliom, M. (2000). Early externalizing behavior prob- lems: Toddlers and preschoolers at risk for later maladjustment. Development and Psychopathology, 12, 467-488.

1018 Youth & Society 49(8)

Carter, B. B., & Spencer, V. G. (2006). The fear factor: Bullying and students with disabilities. International Journal of Special Education, 21, 11-23.

Center for Human Resource Research. (2004). NLSY 79 child & young adult data users guide. Washington, DC: Author.

Civita, M. D., Pagani, L. S., Vitaro, F., & Tremblay, R. E. (2007). Does maternal supervision mediate the impact of income source on behavioral adjustment in children from persistently poor families? The Journal of Early Adolescence, 27, 40-66.

Copeland-Linder, N., Lambert, S. F., Chen, Y., & Ialongo, N. S. (2011). Contextual stress and health risk behaviors among African American adolescents. Journal of Youth and Adolescence, 40, 158-173.

Curtner-Smith, M. E., Culp, A. M., Culp, R., Scheib, C., Owen, K., Tilley, A., . . .Coleman, P. W. (2006). Mothers’ parenting and young economically disad- vantaged children’s relational and overt bullying. Journal of Child and Family Studies, 15, 181-193.

Dishion, T. J., Patterson, G. R., & Griesler, P. C. (1994). Peer adaptations in the devel- opment of antisocial behavior: A confluence model. In L. R. Huesmann (Ed.), Plenum Series in Social/Clinical Psychology. Aggressive behavior: Current per- spectives (pp. 61-95). New York, NY: Plenum Press.

Duncan, G. J., & Brooks-Gunn, J. (1997). Consequences of growing up poor. New York, NY: Russell Sage.

Duncan, R. D. (2004). The impact of family relationships on school bullies and vic- tim. In D. L. Espelage & S. M. Swearer (Eds.), Bullying in American schools: A social-ecological perspective on prevention and intervention (pp. 227-244). Mahwah, NJ: Lawrence Erlbaum.

Eamon, M. K. (2001a). The effects of poverty on children’s socioemotional develop- ment: An ecological systems analysis. Social Work, 46, 256-266.

Eamon, M. K. (2001b). Poverty, parenting, peer, and neighborhood influences on young adolescent antisocial behavior. Journal of Social Service Research, 28, 1-23.

Eamon, M. K., & Zuehl, R. M. (2001). Maternal depression and physical punishment as mediators of the effect of poverty on socioemotional problems of children in single-mother families. American Journal of Orthopsychiatry, 71, 218-226.

Espelage, D. L., Bosworth, K., & Simon, T. R. (2000). Examining the social context of bullying behaviors in early adolescence. Journal of Counseling & Development, 78, 326-333.

Espelage, D. L., Holt, M. K., & Henkel, R. R. (2003). Examination of peer group contextual effects on aggressive behavior during early adolescence. Child Development, 74, 205-220.

Fekkes, M., Pijpers, F. I. M., & Verloove-Vanhorick, S. P. (2005). Bullying: Who does what, when and where? Involvement of children, teachers and parents bul- lying behavior. Health Education Research, 20, 81-91.

Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, M. R., Hall, C. M., & Heyward, D. (2011). Maternal depression and child psychopathology: A meta-analytic review. Clinical Child and Family Psychology Review, 14, 1-27.

Hong et al. 1019

Harris, J. R. (2000). Socialization, personality development, and the child’s environ- ments. Developmental Psychology, 36, 699-710.

Hartup, W. W. (1992). Having friends, making friends, and keeping friends: Relationships as educational contexts (ERIC Digest). Retrieved from http:// ericeece.org/pubs/digests/1992/hartup92.html

Holt, M. K., & Espelage, D. L. (2007). Perceived social support among bullies, vic- tims, and bully-victims. Journal of Youth and Adolescence, 36, 984-994.

Hong, J. S., & Espelage, D. L. (2012). A review of research on bullying and peer victimization in school: An ecological system analysis. Aggression and Violent Behavior, 17, 311-322.

Juvonen, J., Graham, S., & Shuster, M. A. (2003). Bullying among young adoles- cents: The strong, the weak, and the troubled. Pediatrics, 112, 1231-1237.

Kaukiainen, A., Salmivalli, C., Lagerspetz, K., Tamminen, M., Vauras, M., Maki, H., & Poskiparta, E. (2002). Learning difficulties, social intelligence, and self-concept: Connections to bully-victim problems. Scandinavian Journal of Psychology, 43, 269-278.

Kavale, K. A., & Forness, S. R. (1995). Social skill deficits and training: A meta-anal- ysis of the research in learning disabilities. In T. E. Scruggs & M. A. Mastropieri (Eds.), Advances in learning and behavioral disabilities (Vol. 9, pp. 119-160). Greenwich, CT: JAI Press.

Kawachi, I., & Kennedy, B. P. (2002). The health of nations: Why inequality is harm- ful to your health. New York, NY: New Press.

Khoury-Kassabri, M., Benbenishty, R., Astor, R. A., & Zeira, A. (2004). The con- tributions of community, family, and school variables to student victimization. American Journal of Community Psychology, 34, 187-204.

Laursen, B. (1993). The perceived impact of conflict on adolescent relationships. Merrill-Palmer Quarterly, 39, 535-550.

Laursen, B., & Collins, W. A. (1994). Interpersonal conflict during adolescence. Psychological Bulletin, 115, 197-209.

Laursen, B., Hartup, W. W., & Koplas, A. L. (1996). Towards understanding peer conflict. Merrill-Palmer Quarterly, 42, 76-102.

Lee, C. H. (2010). Personal and interpersonal correlates of bullying behaviors among Korean middle school students. Journal of Interpersonal Violence, 25, 152-176.

Low, S., & Espelage, D. L. (2014). Conduits from community violence exposure to peer aggression and victimization: Contributions of parental monitoring, impul- sivity and deviancy. Journal of Counseling Psychology, 61, 221-231.

Malloy, H. L., & McMurray, P. (1996). Conflict strategies and resolutions: Peer conflict in an integrated early childhood classroom. Early Childhood Research Quarterly, 11, 185-206.

Mikami, A. Y., Szwedo, D.E., Allen, J.P., Evans, M.A., & Hare, A.L. (2010). Adolescent peer relationships and behavior problems predict young adults’ com- munication on social networking websites. Developmental Psychology, 46(1), 46-56.

1020 Youth & Society 49(8)

Miller, P., Danaher, D., & Forbes, D. (1986). Sex-related strategies for coping with interpersonal conflicts in children aged five and seven. Developmental Psychology, 22, 543-548.

Mishna, F. (2003). Learning disabilities and bullying: Double jeopardy. Journal of Learning Disabilities, 36, 336-347.

Murphy, B. C. (2002). An integrative examination of peer conflict: Children’s reported goals, emotions, and behaviors. Social Development, 11, 534-557.

Nabuzoka, D. (2003). Teacher ratings and peer nominations of bullying and other behavior of children with and without learning difficulties. Educational Psychology, 23, 307-321.

Nansel, T. R., Overpeck, M., Pilla, R. S., Ruan, W. J., Simons-Morten, B., & Scheidt, P. (2001). Bullying behaviors among U.S. youth: Prevalence and association with psychosocial adjustment. Journal of the American Medical Association, 285, 2094-2100.

National Longitudinal Survey of Youth Codebook Supplement. (n.d.). NLSY79 Appendix 6:1979-2004 urban-rural variables. Retrieved from https://www. nlsinfo.org/content/cohorts/nlsy79/other-documentation/codebook-supplement/ nlsy79-appendix-6-urban-rural-and

Noakes, M. A., & Rinaldi, C. M. (2006). Age and gender differences in peer conflict. Journal of Youth and Adolescence, 35, 881-891.

Pellegrini, A. D., & Bartini, M. (2000). A longitudinal study of bullying, victimiza- tion, and peer affiliation during the transition from primary school to middle school. American Educational Research Journal, 37, 699-725.

Ray, G. E., & Cohen, R. (2000). Children’s evaluations of peer group entry and lim- ited resource situations. Merrill-Palmer Quarterly, 46, 71-89.

Rodkin, P. C., & Hodges, E. V. E. (2003). Bullies and victims in the peer ecology: Four questions for psychologists and school professionals. School Psychology Review, 32, 384-400.

Rose, C. A., & Espelage, D. L. (2012). Risk and protective factors associated with the bullying involvement of students with disabilities. Behavioral Disorders, 37, 133-148.

Rose, C. A., Monda-Amaya, L. E., & Espelage, D. L. (2011). Bullying perpetration and victimization in special education: A review of the literature. Remedial and Special Education, 32, 114-130.

Rose, R. A., & Fraser, M. W. (2008). A simplified framework for using multiple imputation in social work research. Social Work Research, 32, 171-178.

Salzinger, S., Feldman, R. S., Stockhammer, T., & Hood, J. (2002). An ecological framework for understanding risk for exposure to community violence and the effects of exposure on children and adolescents. Aggression and Violent Behavior, 7, 423-451.

Sampson, R. J. (2012). Great American city: Chicago and the enduring neighborhood effect. Chicago, IL: University of Chicago Press.

Schwartz, D., & Gorman, A. H. (2003). Community violence exposure and children’s academic functioning. Journal of Educational Psychology, 95, 163-173.

Hong et al. 1021

Schwartz, D., & Proctor, L. (2000). Community violence exposure and children’s social adjustment in the school peer group: The mediating roles of emotion regu- lation and social cognition. Journal of Consulting and Clinical Psychology, 68, 670-683.

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi- experimental designs for generalized causal inference. Boston, MA: Houghton- Mifflin.

Sheridan, S. M., Warnes, E., & Dowd, S. (2004). Home-school collaboration and bullying: An ecological approach to increase social competence in children and youth. In D. L. Espelage & S. M. Swearer (Eds.), Bullying in American schools: A social-ecological perspective on prevention and intervention (pp. 245-267). Mahwah, NJ: Lawrence Erlbaum.

Sidorowicz, K., & Hair, E. C. (2009). Assessing peer conflict and aggressive behav- iors: A guide for out-of-school time program practitioners. Child Trends. Retrieved from ERIC database. (ED510684)

Silver, R. B., Measelle, J. R., Armstrong, J. M., & Essex, M. J. (2005). Trajectories of classroom externalizing behavior: Contributions of child characteristics, family characteristics, and the teacher-child relationship during the school transition. Journal of School Psychology, 43, 39-60.

Smith, P. K., Madsen, K. C., & Moody, J. C. (1999). What causes the age decline in reports of being bullied at school? Toward a developmental analysis of risks of being bullied. Educational Research, 41, 267-285.

Spriggs, A. L., Iannotti, R. J., Nansel, T. R., & Haynie, D. L. (2007). Adolescent bully- ing involvement and perceived family, peer and school relations: Commonalities and differences across race/ethnicity. Journal of Adolescent Health, 41, 283-293.

Stocker, C. M., & Richmond, M. K. (2007). Longitudinal associations between hostil- ity in adolescents’ family relationships and friendships and hostility in romantic relationships. Journal of Family Psychology, 21, 490-497.

Storch, E. A., Brassard, M. R., & Masia-Warner, C. L. (2003). The relationship of peer victimization to social anxiety and loneliness in adolescence. Child Study Journal, 33, 1-18.

Tobin, T., & Sprague, J. (2000). Alternate education strategies: Reducing violence in school and the community. Journal of Emotional and Behavioral Disorders, 8, 177-186.

Tolan, P. H., Gorman-Smith, D., & Henry, D. B. (2003). The developmental ecology of urban males’ youth violence. Developmental Psychology, 39, 274-291.

U.S. Census Bureau. (2011). Poverty: Highlights. U.S. Department of Commerce. Retrieved from http://www.census.gov/hhes/www/poverty/about/overview/

Veronneau, M. H., & Dishion, T. J. (2010). Predicting change in early adolescent problem behavior in the middle school years: A mesosystemic perspective on parenting and peer experiences. Journal of Abnormal Child Psychology, 38, 1125-1137.

Welsh, W., Stokes, R., & Greene, J. (2000). A macro-level model of school disorder. Journal of Research in Crime & Delinquency, 37, 243-283.

1022 Youth & Society 49(8)

Werner, N. E., & Crick, N. R. (2004). Maladaptive peer relationships and the devel- opment of relational and physical aggression during middle childhood. Social Development, 13, 495-514.

Whitney, I., Smith, P. K., & Thompson, D. (1994). Bullying and children with special educational needs. In P. K. Smith & S. Sharp (Eds.), School bullying: Insights and perspectives (pp. 213-240). London, England: Routledge.

Zimmerman, F. J., Glew, G. M., Christakis, D. A., & Katon, W. (2005). Early cog- nitive stimulation, emotional support, and television watching as predictors of subsequent bullying among grade-school children. Archives of Pediatrics & Adolescent Medicine, 159, 384-388.

Author Biographies Jun Sung Hong, PhD, is an assistant professor in the School of Social Work at Wayne State University. His research interests include school violence, school-based inter- vention, juvenile delinquency, child welfare, and cultural competency in social work practice.

Dorothy L. Espelage, PhD, is an Edward William Gutgsell & Jane Marr Gutgsell Endowed professor and Hardie Scholar of Education, in the Department of Educational Psychology at the University of Illinois, Urbana-Champaign. She has conducted research on bullying, homophobic teasing, sexual harassment, and dating violence for the last 20 years. She has authored on over 130 peer-reviewed journal articles and 25 chapters.

Paul R. Sterzing, PhD, is the co-director of the Center for Prevention Research in Social Welfare (University of California, Berkeley) and the Co-PI on a three-year study funded by the National Institute of Justice to examine polyvictimization among gender and sexual minority youth. This study will be the first to provide a comprehen- sive examination of their victimization experiences across different contexts and perpetrators