MSW WEEK 4 CULTURE DISCUSSION

WEEK4READINGCHILDABUSENEGLECT.pdf

Home >Social Science homework help >MSW WEEK 4 CULTURE DISCUSSION

Child Abuse & Neglect 137 (2023) 106054

A developmental cascade model of adolescent peer relationships, substance use, and psychopathological symptoms from child maltreatment1

Dalhee Yoon a,*, Miyoung Yoon b, Xiafei Wang c, Ada A. Robinson-Perez d

a Department of Social Work, Binghamton University-State University of New York, USA b Department of Social Welfare, Pusan National University, South Korea c School of Social Work, Syracuse University, USA d Division of Diversity, Equity and Inclusion, Binghamton University-State University of New York, USA

A R T I C L E I N F O

Keywords: Child maltreatment Deviant peer affiliation Externalizing symptoms Internalizing symptoms Peer relationship Substance use

A B S T R A C T

Introduction: Although previous studies have demonstrated high intercorrelations among deviant peer affiliation, substance use, externalizing and internalizing symptoms in adolescence, these studies have been limited because they did not examine 1) the associations over time by assuming one particular sequence; and 2) child maltreatment effects. Methods: This study included 617 adolescents (54.3 % girls, 55.6 % Black) at-risk of maltreatment living in the U.S and primarily low-income. Deviant peer affiliation was assessed at ages 12, 14, and 16 using 13 items from the modified version of the Youth Risk Behavior and Monitoring the Future Survey. Externalizing and internalizing symptoms were measured at ages 12, 14, and 16 using the Child Behavior Checklist. The number of substances used (ages 12, 14, 16, and 18) were created by summing the self-reported alcohol, tobacco, and marijuana use. Each type of maltreatment (birth to age 12) was assessed using the self-report. Results: Autoregressive cross-lagged structural equation modeling explained the stability effects within each domain, as well as how different maltreatment types affect diverse developmental processes. Cross-lagged results showed the socialization effects of peers on substance use, whereas the peer selection effects on externalizing symptoms. Physical abuse was only associated with externalizing symptoms, while sexual abuse was associated with both externalizing and inter- nalizing symptoms. Additionally, emotional abuse was associated with deviant peer affiliation and substance use. Conclusions: Identifying the underlying reciprocal processes offers a deeper understanding of peer relationships in the substance use and externalizing symptoms among at-risk of maltreatment sample.

* Corresponding author at: Department of Social Work, Binghamton University-State University of New York, P.O. Box 6000, Binghamton, NY 13902-6000, USA.

E-mail address: [email protected] (D. Yoon). 1 This research was funded by the Institute for Justice and Well-Being with support from Binghamton University Division of Research. The current

study is based on data from the Consortium of Longitudinal Studies of Child Abuse and Neglect (LONGSCAN). The data were made available by the National Data Archive on Child Abuse and Neglect (NDACAN), Cornell University. All authors declare that they have no conflict of interest.

Contents lists available at ScienceDirect

Child Abuse & Neglect

journal homepage: www.elsevier.com/locate/chiabuneg

https://doi.org/10.1016/j.chiabu.2023.106054 Received 13 October 2022; Received in revised form 11 January 2023; Accepted 17 January 2023

Child Abuse & Neglect 137 (2023) 106054

1. Introduction

Adolescence is a developmental transition period marked by the heightened importance of peer relationships (Brown et al., 1997). Peer groups become the primary sources of social and emotional support for adolescents, as adolescents employ feedback and acceptance from peers to build a sense of self-concept (Brechwald & Prinstein, 2011). In line with the important role of peer re- lationships, a great body of research (Eiden et al., 2016; Gonzale et al., 2017) has explored the role of peers in adolescent behavioral problems, such as substance use and psychopathological symptoms (i.e., internalizing and externalizing symptoms). However, the direction of peer relationship—whether adolescents' behavioral problems are causes or consequences of peer relationships and vice versa—is unclear with two theoretical models (i.e., selection effects and socialization effects). Selection effects refer to adolescents' tendency to choose friends who are similar to themselves (Thompson et al., 2020); while the hypothesis of socialization effects posits that adolescents become more similar to their friends over time via attitude transference, as well as behaviors modeling and rein- forcement (Haynie, 2002). The unclear direction of peer relationship prohibits the development of effective preventive interventions, which brings massive social costs. Understanding the diverse factors of adolescent substance use with various developmental periods is salient to develop developmentally sensitive interventions.

Along with the vague directionality, relatively few studies have examined peer relationships within the context of child maltreatment, despite the fact adolescents who experienced child maltreatment are known to be at heightened risk for compromised physiological, behavioral, psychological, and interpersonal functioning across the life span (Cicchetti, 2016). The developmental psychopathology perspective explains that early adverse life experiences cause the emergence of a wide range of difficulties including impaired peer relationships, substance use, internalizing and externalizing symptoms. However, very limited research has investigated how peer relationships, substance use, and psychopathological symptoms develop and interrelate progressively over time within the context of child maltreatment (Duprey et al., 2017; Fava et al., 2019; Hudson et al., 2017; Jones et al., 2013). Drawing from the Longitudinal Studies of Child Abuse and Neglect (LONGSCAN), this study disentangles the directionality of the peer relationships (i.e., deviant peer affiliation), substance use, and psychopathological symptoms, as well as the effects of child maltreatment using the longitudinal designs with repeated measures.

1.1. Peer relationship, psychopathological symptoms, and adolescent substance use

Both selection and socialization effects have been supported by several empirical studies. Using a high-risk maltreatment sample (i. e., LONGSCAN), researchers (Yoon et al., 2019; Yoon, Snyder, and Yoon, 2020a) found that adolescents who were affiliated with deviant peers were more likely to be influenced by peer role models, suggesting socialization effects of deviant peers on substance use. However, it is also possible that adolescents using substances actively select friends who share their behaviors and attitudes (McDonough et al., 2016; Simons-Morton & Chen, 2006).

As for the link between pro-or deviant-peer affiliation and psychopathological symptoms, several studies have also elaborated the two theoretical frameworks. For example, Yoon, Snyder, Yoon, and Coxe (2020b) found that early adolescent externalizing symptoms were associated with increased deviant peer affiliation using at risk of maltreatment sample at ages 12 and 16. In addition, Allen et al. (2021) demonstrated that adolescent externalizing symptoms are more likely to be associated with decreased prosocial peer affiliation. On the contrary, Chen et al. (2015) articulated that deviant peer affiliation conferred risks for externalizing symptoms. Li et al. (2020) also supported the socialization effects for internalizing symptoms by identifying mediation of deviant peer affiliation in the associ- ation between emotional abuse and internalizing symptoms. Forming deviant peer affiliation are often interpreted as high conflicts and low-quality peer relationships, which may be associated with internalizing symptoms (Mrug, Hoza, & Bukowski, 2004).

Similar to the selection and/or socialization effects of peers, the direction of psychopathological symptoms and substance use is still under debate. Although the association between externalizing symptoms and substance use is robustly consistent in previous stud- ies—externalizing symptoms in early adolescence predicted substance use in late adolescence (Colder et al., 2018; Fava et al., 2019; Mezquita et al., 2014; Oshri et al., 2011; Traube et al., 2012; Yampolskaya et al., 2019), the opposite direction has also been reported (Rodriguez-Seijas et al., 2017). In addition, the ambiguous direction between internalizing symptoms and substance use has been reported throughout adolescence (Duprey et al., 2017; Hudson et al., 2017; Trim et al., 2007). The associations between substance use and psychopathological symptoms may be related to emotion dysregulation. On one hand, adolescents with internalizing and externalizing symptoms have difficulties in emotion regulation, which led them to rely on substance for coping (Siegel, 2015). On the other hand, using substances may compromise adolescents' executive functioning, which in turn exacerbate emotion dysregulation to cause internalizing and externalizing symptoms (Weiss et al., 2017).

Despite the mainstream trend (e.g., socialization effects of substance use in at-risk sample, and the prediction of adolescent sub- stance use from externalizing symptoms), it is premature to make a conclusion because previous studies did not examine the longi- tudinal changes of behavioral problems and peer relationships (Jones et al., 2013; Yoon, Snyder, and Yoon, 2020a). For example, only deviant peer affiliation was assessed longitudinally, whereas substance use was measured at one time (Yoon, Snyder, Yoon, and Coxe, 2020b). In addition, the prior studies used either the behavioral problems or peer relationships (Duprey et al., 2017; Fava et al., 2019), although the strong associations among the behavioral problems and peer relationships have been notably reported. Further, most studies have used general samples without consideration of traumatic experiences (McDonough et al., 2016; Mezquita et al., 2014).

1.2. Developmental cascade models

Developmental cascade refers to cumulative developmental outcomes shaped by transactions and interactions between various

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

domains of developing systems (Masten & Cicchetti, 2010). In order to understand the developmental progress of adolescent behavioral problems and peer relationships, it is crucial to examine cumulative effects. Although there are a few studies that inves- tigated the cumulative effects of deviant peer affiliation, psychopathological symptoms, and substance use using the developmental cascade model, they relied on a general sample without consideration of adverse childhood experiences, such as child maltreatment (Danzo et al., 2017; Eiden et al., 2016; Englund & Siebenbruner, 2012; Thompson et al., 2020).

To the best of our knowledge, only three studies have adapted a progressive developmental cascade model within the context of child maltreatment (Boyd et al., 2018; Negriff, 2018; Zhu et al., 2017). The three studies have generally reported within domain stability consistently (e.g., earlier deviant peer affiliation led to later deviant peer affiliation), but bidirectional relationships among domains were different. For example, Boyd et al. (2018), using a sample aged 13–18, and Zhu et al. (2017), using 7-9th grade students, documented a bidirectional relationship among externalizing symptoms and peer alcohol use across all study periods. However, Negriff (2018) indicated no cross-lagged effect among delinquency and peer delinquency across age 10–18 because sexual behavior plays an important role in the associations. Although three studies have guided the progressive development of substance use in high risk maltreated sample, these studies also have been limited by 1) excluding internalizing symptoms (Boyd et al., 2018; Negriff, 2018; Zhu et al., 2017); 2) using only one type of substance (Boyd et al., 2018); and 3) using a lumped maltreatment measure (Negriff, 2018). These limitations hinder understanding of the developmental progress of behavioral problems and peer relationships among high-risk of maltreatment sample and the effects of different types of child maltreatment.

1.3. The current study

Both extant empirical research and the theoretical frameworks—i.e., selection and socialization effects, as well as developmental psychopathology perspective—suggest the developmental cascade model of deviant peer affiliation, substance use, and psychopath- ological symptoms from child maltreatment in adolescence. To untangle the causal and directional relations using psychopathological symptoms, deviant peer affiliation, and substance use from child maltreatment, the current study's specific aims are: 1) to examine the stability effects within the domain (i.e., deviant peer affiliation, substance use, internalizing symptoms, and externalizing symptoms); 2) to scrutinize the stability and the role of different child maltreatment types; and 3) to investigate reciprocal relationships among deviant peer affiliation, substance use, and psychopathological symptoms from child maltreatment.

2. Method

2.1. Sample

The study data were derived from the Longitudinal Studies of Child Abuse and Neglect (LONGSCAN). The LONGSCAN (Larrabee & Lewis, 2014; Runyan et al., 1998) collected at-risk of child maltreatment sample from 1991 to 2012 from five study sites across the U.S. (N = 1354). Children, caregivers, and teachers reported comprehensive information through face-to-face interviews when the children were 4, 6, 8, 12, 14, 16, and 18 years old. Participants in this study included 617 adolescents, who completed at least two waves of age 12, 14, 16, and 18 data collection. Compared to the baseline sample (at age 4), participants in this study reported less deviant peer affiliation at age 16, t (787) = 2.58, p = .010, less externalizing symptoms at age 16, t (626.83) = 2.00, p = .046, and lower physical abuse experiences, z = − 2.96, p = .003. No other significant differences were reported.

2.2. Measures

2.2.1. Deviant peer affiliation Using the modified version of the Youth Risk Behavior and Monitoring the Future Survey (Knight et al., 2008; 13 items), ado-

lescents' self-reported peer groups' risk behaviors—including substance use, committing offenses—were used to assess deviant peer affiliation. The deviant peer affiliation was assessed at ages 12, 14, and 16, and each item was scaled from 0 (none of my friends) to 2 (most of my friends). To create a total deviant peer affiliation scale, all items were summed; higher scores represent greater levels of deviant peer affiliation. The Cronbach's alpha for the total deviant peer affiliation scale at ages 12, 14, and 16 in this sample were 0.79, 0.83, and 0.84, respectively.

2.2.2. Externalizing and internalizing symptoms Adolescents' externalizing and internalizing symptoms were measured at ages 12, 14, and 16 using the Child Behavior Checklist

(CBCL; Achenbach, 1991). The CBCL is a standardized measure for adolescent symptoms, which is reported by their primary caregivers on adolescents' psychopathological symptoms in the past 6 months. The response options were scored using the following categories: 0 = not true, 1 = somewhat or sometimes true, 2 = very true or often true. For this study, age- and gender-standardized T scores were used, with higher scores representing greater problems. Cronbach's alpha for externalizing symptoms were 0.91, 0.93, and 0.93 and internalizing symptoms were 0.87, 0.88, and 0.87 in this study sample. These scales demonstrated the content, construct, and criterion- related validity (Knight et al., 2008).

2.2.3. Child maltreatment Given the strength of a self-reported maltreatment measure that identifies underreporting maltreatment cases (Kobulsky et al.,

2018), LONGSCAN-developed self-report of child maltreatment measures (lifetime experiences up to age 12) were used to assess each

D. Yoon et al.

ChildAbuse& Neglect137(2023)106054

Table 1 Descriptive statistics and correlations among main study variables.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

1. PA 12 – 0.26*** 0.49*** 0.20*** 0.06 0.09* 0.16*** 0.11* 0.13** 0.13** 0.14** 0.15** 0.16*** 0.04 0.08 0.04 2. SA 12 – 0.31*** 0.13*** 0.06 0.01 0.20*** 0.10* 0.15** 0.16*** 0.17*** 0.16*** 0.10* 0.11* 0.07 0.09* 3. EA 12 – 0.25*** 0.18*** 0.13** 0.15*** 0.11* 0.17*** 0.12** 0.10* 0.11* 0.22*** 0.12** 0.13** 0.05 4. DP 12 – 0.27*** 0.27*** 0.07 0.04 0.06 0.05 0.06 0.02 0.32*** 0.12** 0.18*** 0.09* 5. DP 14 – 0.45*** 0.07 0.09* 0.10* 0.04 0.07 0.04 0.13** 0.38*** 0.36*** 0.21*** 6. DP 16 – 0.17*** 0.18*** 0.23*** 0.12** 0.15** 0.14** 0.14** 0.21*** 0.56*** 0.40*** 7. EXT 12 – 0.69*** 0.62*** 0.67*** 0.55*** 0.48*** 17*** 0.11* 0.21*** 0.16*** 8. EXT 14 – 0.70*** 0.47*** 0.69*** 0.53*** 0.13** 0.14** 0.23*** 0.18*** 9. EXT 16 – 0.44*** 0.52*** 0.71*** 0.14** 0.15*** 0.29*** 0.17*** 10. INT 12 – 0.67*** 0.57*** 0.13** 0.06 0.14** 0.15*** 11. INT 14 – 0.67*** 0.10* 0.09* 0.18*** 0.19*** 12. INT 16 – 0.14** 0.12* 0.15** 0.14** 13. SU 12 – 0.22*** 0.18*** 0.12** 14. SU 14 0.41*** 0.25*** 15. SU 16 – 0.53*** 16. SU 18 – M / n 0.35 0.36 1.32 1.06 2.55 4.20 54.32 54.07 52.63 50.63 49.72 48.33 0.11 0.27 0.79 1.14 SD / % 0.95 1.18 2.28 1.99 3.07 3.68 10.93 11.36 11.46 10.92 11.52 11.46 0.42 0.65 1.07 1.10

Note. PA physical abuse; SA sexual abuse; EA emotional abuse; DP deviant peer affiliation; EXT externalizing symptoms; INT internalizing symptoms; SU substance use. * p < .05. ** p < .01. *** p < .001.

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

type of maltreatment (15 dichotomous physical abuse items; α = 0.67, 11 dichotomous sexual abuse items; α = 0.83, and 18 dichotomous emotional abuse items; α = 0.82). To create a score for each type of maltreatment, responses within each type of maltreatment were summed; higher scores represent more abuse experiences.

2.2.4. Adolescent substance use Adolescent self-reported alcohol, tobacco, and marijuana use at ages 12, 14, 16, and 18 were assessed. For each substance, ado-

lescents reported whether they had been using alcohol, tobacco, and marijuana in the past year (1 = yes, 0 = no). The number of substances was created by summing the three questions (possible range: 0–3); higher scores indicate more substance use.

2.2.5. Covariates A high volume of research has documented that gender, race (Lansford et al., 2004; Yoon, Snyder, and Yoon, 2020a), household

income (Hinnant et al., 2019), and sexual activity (Negriff, 2018) were associated with adolescent deviant peer affiliation, psycho- pathological symptoms, and substance use. Thus, the present study used these variables as covariates. Adolescents self-reported their gender at age 12 and it was dummy coded in this study, 1 = “Girl” and 0 = “Boy.” Adolescent race was assessed at age 4 by their caregivers using seven categories (e.g., Hispanic, Native American, Asian, bi/multi-racial adolescents). In this study, it was dichoto- mized with 1 = “Black” and 0 = “Non-Black,” given the high proportion of Black adolescents (55.6 %). Household income was assessed at age 12 by their caregivers using 12 categories, from 1 = “Less than $5,000” to 12 = “Over $55,000 (with $5,000 increments).” Sexual activity was assessed at age 12, 14, and 16 using a single question, “Have you ever had sex?” with 1 = “yes” and 0 = “no.”

2.3. Statistical analyses

There are few models available to investigate stability and cross-lagged effects, such as cross-lagged panel model, random intercept cross lagged model, and autoregressive latent trajectory model. However, there is no common rule on which model should be chosen, but Orth et al. (2021) explained that the model should be determined by theoretical perspectives. This study is interested in within person effects (e.g., whether changes from an individual's expected substance use are predicted from preceding deviations on deviant peer affiliation), we opted to use cross-lagged panel model with MLR estimation was employed using Mplus 8.3. For the first aim (a stability model), the current study included each construct (i.e., deviant peer affiliation, substance use, internalizing and externalizing symptoms) over time (autoregressive paths) and synchronous associations of variables (within wave). For the second aim, this study added child maltreatment types (i.e., physical, sexual, and emotional abuse) as independent variables to examine how different maltreatment types set off diverse developmental processes and whether the rank-order developmental stability of each construct (i.e., deviant peer affiliation, substance use, internalizing and externalizing symptoms) is stable across four developmental periods. Building upon the second aim model, this study added the cross-lagged paths among constructs in the third aim model (deviant peer affiliation, externalizing and internalizing symptoms, and substance use) to investigate the reciprocal relationships. To examine the robustness of paths, all models were examined with covariates by regressing all constructs on covariates. In order to evaluate the model fit, multiple fit indices were used (West et al., 2012): Comparative Fit Index (CFI; >0.95), Root Mean Square Error of Approximation and 90 % Confidence Interval (RMSEA; <0.06 is close fitting model, <0.08 is an adequate fitting model), and Standardized Root Mean Square Residual (SRMR; <0.08). For parameter level (i.e., local fit), the statistical significances of individual parameters were checked and the standardized coefficients were calculated. Using the Little's Missing Completely At Random (MCAR) test, the missing data were examined (χ2 = 213.85, p = .490) and it indicates that data were missing completely at random. Missing data were handled using Full- information maximum likelihood (FIML) estimates because it produces less biased estimates (Enders & Bandalos, 2001).

Descriptive statistics and frequencies were conducted to check missing values and outliers. To determine the normality of variables, Curran et al. (1996)’s criteria of skewness (<2) and kurtosis (<7) were used. Zero-order correlations were also performed to check bivariate relationships and potential multicollinearity. In this study, all correlations among predictors were below 0.80, indicating that data has no multicollinearity issue (Allison, 1999). All preliminary analyses were conducted using SPSS v.25.

3. Results

3.1. Descriptive statistics

Table 1 presents means, standard deviations, and correlations among main study variables. Approximately half of the sample was female (54.3 %) and a majority of the adolescents were Black (55.6 %). The average household income for the sample ranged from $25,000 to $30,000. As adolescents age, the more deviant peer affiliation and substance use scores reported. Adolescents' externalizing and internalizing symptoms decreased as they aged.

3.2. Evaluation of cascade models

The first model's aim was to examine the stability of deviant peer affiliation, externalizing and internalizing symptoms, and substance use across four developmental periods using parallel autoregressive models. The model fit indicated relatively poor but acceptable: CFI = 0.95; RMSEA = 0.07 [90 % CI = 0.06–0.08]; SRMR = 0.08. Youth with higher levels of deviant peer affiliation, externalizing and internalizing symptoms, and substance use at age 12 were predicted to exhibit similarly high levels in these domains at subsequent time points.

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

The main aim of the second model was to evaluate how different types of maltreatment set forth various developmental processes of deviant peer affiliation, externalizing and internalizing symptoms, and substance use by adding child maltreatment types in the stability of constructs. The model yielded a good model fit to the data, CFI = 0.96; RMSEA = 0.05 [90 % CI = 0.04–0.06]; SRMR = 0.06. Results supported that the developmental stability of each domain is differently affected by the types of child maltreatment.

The final model of the cascade model (Aim 3) added cross-lagged relationships among domains and produced an excellent fit to the data, CFI = 0.98; RMSEA = 0.04 [90 % CI = 0.04–0.06]; SRMR = 0.04 (Table 2, Fig. 1). The stability paths from age 12 to 14, from age 14 to 16, and from 12 to 16 for deviant peer affiliation, externalizing, and internalizing symptoms were positive and significant. However, the recurrent paths of substance use from age 12 to 16, from age 14 to 18, and from age 12 to 18 were not significant. In terms of cross-lag reciprocal paths, externalizing symptoms at age 12 predicted internalizing symptoms at age 14 (β = 0.192, p < .001) after controlling for the prior levels of internalizing symptoms and within-time covariance. Externalizing symptoms at age 14 posi- tively predicted relations with substance use at age 16 (β = 0.156, p < .001) after controlling for the prior levels of substance use and within-time covariance. Deviant peer affiliation at age 14 positively predicted substance use at age 16 after controlling for the previous substance use and within-time covariance (β = 0.156, p < .001). Similarly, deviant peer affiliation at age 16 positively predicted substance use at age 18 controlling for the previous substance use (β = 0.157, p = .001) suggesting that adolescents who affiliated with deviant peers used more substances subsequently. As for the effects of child maltreatment, physical abuse experiences before age 12 were associated with deviant peer affiliation (β = 0.100, p = .031) and externalizing symptoms at age 12 (β = 0.110, p = .017). However, sexual abuse experiences before age 12 were associated with externalizing (β = 0.129, p = .003) and internalizing symptoms at age 12 (β = 0.114, p = .009). Youth who have been emotionally abused had higher levels of deviant peer affiliation (β = 0.191, p < .001) and substance use (β = 0.167, p < .001). There were several concurrent correlations among domains within wave. At age 12, only the associations between deviant peer affiliation and externalizing symptoms, as well as deviant peer affiliation and internalizing symptoms, were not statistically significant. At age 14, the association between deviant peer affiliation and substance use, as well as the association between externalizing symptoms and internalizing symptoms were significant. At age 16, deviant peer affiliation was not associated with internalizing symptoms. In addition, the correlation between internalizing symptoms and substance use was not

Table 2 Autoregressive Cross-Lagged Structural Equation Model for the Overall Study Sample (N = 617).

β p β p β p

Concurrent correlations 1. DP12 ↔ EXT12 0.027 0.501 7. DP14 ↔ EXT14 0.059 0.177 13. DP16 ↔ EXT16 0.137 0.002 2. DP12 ↔ INT12 0.016 0.695 8. DP14 ↔ INT14 0.048 0.265 14. DP16 ↔ INT16 0.080 0.082 3. DP12 ↔ SU12 0.285 <0.001 9. DP14 ↔ SU14 0.284 <0.001 15. DP16 ↔ SU16 0.413 <0.001 4. EXT12 ↔ INT12 0.659 <0.001 10. EXT14 ↔ INT14 0.578 <0.001 16. EXT16 ↔ INT16 0.605 <0.001 5. EXT12 ↔ SU12 0.125 0.002 11. EXT14 ↔ SU14 0.063 0.144 17. EXT16 ↔ SU16 0.165 <0.001 6. INT12 ↔ SU12 0.087 0.031 12. INT14 ↔ SU14 0.038 0.377 18. INT16 ↔ SU16 0.022 0.635

Recurrent paths 1. DP12 → DP14 0.224 <0.001 6. INT14 → INT16 0.454 <0.001 11. EXT12 → EXT16 0.230 <0.001 2. DP14 → DP16 0.333 <0.001 7. SU12 → SU14 0.136 0.002 12. INT12 → INT16 0.193 <0.001 3. EXT12 → EXT14 0.665 <0.001 8. SU14 → SU16 0.236 <0.001 13. SU12 → SU16 0.043 0.288 4. EXT14 → EXT16 0.507 <0.001 9. SU16 → SU18 0.404 <0.001 14. SU12 → SU18 − 0.006 0.872 5. INT12 → INT14 0.532 <0.001 10. DP12 → DP16 0.152 0.001 15. SU14 → SU18 0.022 0.610

Cross-lagged paths 1. DP12 → EXT14 0.000 0.989 15. INT12 → DP14 − 0.021 0.678 29. DP14 → INT16 − 0.006 0.865 2. DP12 → EXT16 − 0.007 0.847 16. INT12 → DP16 − 0.036 0.567 30. DP14 → SU16 0.156 <0.001 3. DP12 → INT14 0.038 0.228 17. INT12 → EXT14 0.031 0.452 31. DP14 → SU18 − 0.024 0.588 4. DP12 → INT16 − 0.050 0.167 18. INT12 → EXT16 0.012 0.801 32. EXT14 → DP16 0.103 0.140 5. DP12 → SU14 0.063 0.119 19. INT12 → SU14 − 0.069 0.205 33. EXT14 → INT16 0.098 0.076 6. DP12 → SU16 0.081 0.056 20. INT12 → SU16 − 0.039 0.519 34. EXT14 → SU16 0.134 0.043 7. DP12 → SU18 − 0.027 0.492 21. INT12 → SU18 − 0.030 0.600 35. EXT14 → SU18 − 0.031 0.659 8. EXT12 → DP14 0.009 0.856 22. SU12 → DP14 0.001 0.979 36. INT14 → DP16 0.018 0.783 9. EXT12 → DP16 0.024 0.718 23. SU12 → DP16 0.020 0.640 37. INT14 → EXT16 0.036 0.482 10. EXT12 → INT14 0.192 <0.001 24. SU12 → EXT14 0.014 0.734 38. INT14 → SU16 − 0.013 0.832 11. EXT12 → INT16 0.025 0.648 25. SU12 → EXT16 0.005 0.872 39. INT14 → SU18 0.071 0.285 12. EXT12 → SU14 0.069 0.201 26. SU12 → INT14 − 0.032 0.343 40. DP16 → SU18 0.157 0.001 13. EXT12 → SU16 0.033 0.607 27. SU12 → INT16 0.064 0.059 41. EXT16 → SU18 − 0.073 0.292 14. EXT12 → SU18 0.065 0.284 28. DP14 → EXT16 0.011 0.744 42. INT16 → SU18 0.052 0.427

Child maltreatment 1. PA → DP12 0.100 0.031 5. SA → DP12 0.036 0.416 9. EA → DP12 0.191 <0.001 2. PA → EXT12 0.110 0.017 6. SA → EXT12 0.129 0.003 10. EA → EXT12 0.025 0.601 3. PA → INT12 0.081 0.081 7. SA → INT12 0.114 0.009 11. EA → INT12 0.035 0.466 4. PA → SU12 0.077 0.101 8. SA → SU12 0.006 0.883 12. EA → SU12 0.167 <0.001

Note. DP deviant peer affiliation; EXT externalizing symptoms; INT internalizing symptoms; SU substance use; PA physical abuse; SA sexual abuse; EA emotional abuse.

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

statistically significant.

4. Discussion

The present study examined the developmental processes of deviant peer affiliation, substance use, externalizing symptoms, and internalizing symptoms from child maltreatment. These findings contribute to clarifying the temporal impact of deviant peer affili- ation, externalizing and internalizing symptoms, and adolescent substance use among at-risk of maltreatment population by con- trolling concurrent effects of the highly correlated domains, previous effects, gender, race, and household income. Additionally, the study's findings provide insights into how different maltreatment types are associated with behavior problems.

Consistent with previous literature (Boyd et al., 2018; Yoon et al., 2019; Yoon, Snyder, and Yoon, 2020a), these findings generally support the theoretical models of peer socialization effects on substance use (i.e., deviant peer affiliation initiated the use of sub- stances). Adolescents who are affiliated with deviant peers may have pressure to use substances because adolescence is the devel- opmentally sensitive period that escalates susceptibility to peer pressure (McCoy et al., 2019), which may be related to their own increased substance use. Another explanation is that adolescents may learn and perceive the substance use as “cool” in the peer group, because of the high rates of substance use by acquaintances. However, no socialization effects were found in early adolescence. It is possible that deviant peer affiliation and their problematic behaviors such as substance use in early adolescence may be carefully monitored by their parents. In other words, parents may foster communication with their child and facilitate early adolescents' disclosure of behavior issues, which may prohibit deviant peer affiliation and their substance use (Fosco et al., 2012).

As for the association between deviant peer affiliation and externalizing symptoms, this study demonstrated that no peer selection or socialization effects. These findings are consistent with Negriff (2018), but inconsistent with Yoon, Snyder, Yoon, and Coxe (2020b) study, which generally supported peer selection effects and Zhu et al. (2017) study that predominantly suggested socialization of aggression. This discrepancy may be explained by crucial role of sexual activity (Yoon and colleagues' study did not include sexual activity as a covariate) or different sampling: The present study used the at-risk of maltreatment sample in the U.S., whereas Zhu et al. (2017) used a general sample in China.

Regarding the role of child maltreatment types, physical abuse was associated with deviant peer affiliation and externalizing symptoms, while sexual abuse was associated with externalizing and internalizing symptoms. The connection between physical abuse and externalizing symptoms aligns with social learning theory (Bandura, 1978) and previous empirical studies (Barboza et al., 2017; Yoon et al., 2021). Physically abused youth may learn how to interact with others by observing perpetrators' behaviors and then eventually mimic the way they learned. In addition, physically abused youth may experience emotional numbing (i.e., youth normalize their peer groups' behavior issues) that lead to affiliation with deviant peers (Foster & Brooks-Gunn, 2009). Sexual abuse and its association with both externalizing and internalizing symptoms are also consistent with previous literature (Barboza et al., 2017; Lewis et al., 2016). Sexual abuse may adversely affect emotional regulation, which may further provoke internalizing and externalizing symptoms (Langevin et al., 2015). Moreover, this study supported the effects of emotional abuse on deviant peer affiliation and substance use, which is consistent with past research (Yoon, Shi, Yoon, Pei, Schoppe-Sullivan, and Snyder, 2020c; Yoon, Snyder, and Yoon, 2020a). Compared to other types of maltreatment, emotional abuse is more likely to jeopardize healthy re- lationships with peers, which may be associated with youth affiliating with deviant peers or substance use (Oshri et al., 2015).

Not surprisingly, the current study demonstrated the strong stability of each domain (i.e., deviant peer affiliation, externalizing and internalizing symptoms, and substance use) throughout the developmental periods. In other words, early onset of behavioral issues and deviant peer affiliation maintain during adolescence. These findings indicated the importance of early assessment of behavior issues

Fig. 1. Cross-Lagged Structural Equation Model from Child Maltreatment to Adolescent Substance Use for Overall Sample (N = 617). Note. In the interest of clarity, only significant cross-lagged paths and recurrent paths were depicted. In addition, correlations between deviant peer affiliation, externalizing symptoms, internalizing symptoms, and substance use at age 12, 14, and 16 are omitted from the figure. For all other parameter coefficients are in the Table 2. Model fit: CFI = 0.98; RMSEA = 0.04 [90 % CI = 0.04–0.06]; SRMR = 0.04.

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

that may prevent later behavior issues. However, we found no re-current paths from substance use at age 12 to 18, from age 12 to 16, as well as age 14 to 18. The reason for this insignificant association could be explained by other factors, such as race (In this study, a little over half of the sample was Black). Research has supported that Black youth are less likely to use substances (Johnston et al., 2020). In other words, they may have less interest in substance use and not use substances persistently even though they have been exposed. Future research is needed to examine how race is associated with the development of substance use.

4.1. Strengths and limitations

The current study had several distinct strengths. First of all, this study helps to understand the temporal ordering of deviant peer affiliation, substance use, and psychopathological symptoms from child maltreatment by using a large longitudinal at-risk of maltreatment sample, repeated measures, and a sophisticated model. In particular, as a theory-driven model, this study adapted both peer selection and peer socialization of behavior problems and substance use simultaneously, which helped to create a clear under- standing of the pathways among highly correlated domains. Thus, it provides preventive intervention points to eliminate adolescent substance use in at-risk of maltreatment population. Moreover, self-reported measures of child maltreatment allowed this study to detect and include child maltreatment cases that may not be identified in administrative records.

Although the present study had multiple strengths, there are some limitations to note. First, this study used parent-reported measures of adolescent externalizing and internalizing symptoms rather than adolescent self-reports, given the lack of self-reported measure at age 14 and 16 in the data. Since only behavioral problems that parents are aware of were included in this study, there may be the discrepancy between the assessed symptoms and the actual symptoms. Second, child maltreatment measures were assessed at age 12, although they are lifetime measures. The same assessment period with deviant peer affiliation, substance use, and behavior problems may undermine the longitudinal effects of child maltreatment on subsequent behavior problems. Additionally, this study used only the type of maltreatment without neglect (the dataset does not have lifetime self-reported neglect at age 12). Future studies may consider diverse child maltreatment effects such as timing onset or chronicity, as well as neglect. Third, this study used observed variables, instead of a latent variable. Future studies may consider latent variables using multiple observed indicators because it helps to control measurement error. Fourth, the study sample is at-risk of child maltreatment with relatively low income. Thus, the results of the present study may not be generalizable to adolescents in the general population. Fifth, adolescent self-reports of deviant peer affiliation may have been influenced by social desirability bias.

4.2. Implications and conclusions

These findings indicate that age and types of child maltreatment play a critical role in specific symptoms including substance use and deviant peer affiliation. In particular, findings on different paths set off by adolescents' maltreatment experiences highlight the importance of practitioners to expediently refer maltreated adolescent for systematic screening and linking them to appropriate services in the youth's community. For example, Risk Reduction through Family Therapy (RRFT)—a multicomponent family-focused treatment that is based on the ecological model—may provide a comprehensive approach individualized to the needs, strengths, developmental factors, and cultural background of each adolescent and family (Danielson et al., 2010). The current study's findings also suggest that externalizing symptoms in mid-adolescence (at age 14) are important, as they are associated with several risk be- haviors. To reduce externalizing symptoms, use of the Trauma Focused Cognitive Behavioral Therapy (TF-CBT) could be another effective intervention. TF-CBT encourages adolescents to recognize thoughts that impact their feelings in an effort to disrupt or deter a pattern of risky or disruptive behaviors (Gauthier-Duchesne et al., 2017). By helping maltreated youth understand that they are not responsible for the abuse incurred, their awareness is heightened, and emotional regulation and behavioral management is shown to improve. Lastly, in response to the socialization effects of deviant peer affiliation on substance use, bolstering positive pro-social activities to model healthy peer engagement for youth should be considered in conjunction with these trauma-informed treatments.

Data availability

The authors do not have permission to share data.

References

Achenbach, T. M. (1991). Manual for the child behavior checklist/4-18 and 1991 profile. Department of Psychiatry, University of Vermont Burlington, VT. Allen, E. K., Desir, M. P., & Shenk, C. E. (2021). Child maltreatment and adolescent externalizing behavior: Examining the indirect and cross-lagged pathways of

prosocial peer activities. Child Abuse & Neglect, 111, Article 104796. Allison, P. (1999). Multiple regression: A primer. Thousand Oaks, CA: Sage. Bandura, A. (1978). Social learning theory of aggression. Journal of Communication, 28(3), 12–29. Barboza, G. E., Dominguez, S., & Pinder, J. (2017). Trajectories of post-traumatic stress and externalizing psychopathology among maltreated foster care youth: A

parallel process latent growth curve model. Child Abuse & Neglect, 72, 370–382. Boyd, S. J., Sceeles, E. M., Tapert, S. F., Brown, S. A., & Nagel, B. J. (2018). Reciprocal relations between positive alcohol expectancies and peer use on adolescent

drinking: An accelerated autoregressive cross-lagged model using the NCANDA sample. Psychology of Addictive Behaviors, 32(5), 517–527. Brechwald, W. A., & Prinstein, M. J. (2011). Beyond homophily: A decade of advances in understanding peer influence processes. Journal of Research on Adolescence,

21(1), 166–179. Brown, B. B., Dolcini, M. M., & Leventhal, A. (1997). Transformations in peer relationships at adolescence: Implications for health-related behavior. In J. Schulenberg,

J. L. Maggs, & K. Hurrelmann (Eds.), Health risks and developmental transitions during adolescence (pp. 161–189). Cambridge University Press.

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

Chen, D., Drabick, D. A., & Burgers, D. E. (2015). A developmental perspective on peer rejection, deviant peer affiliation, and conduct problems among youth. Child Psychiatry & Human Development, 46(6), 823–838.

Cicchetti, D. (2016). Socioemotional, personality, and biological development: Illustrations from a multilevel developmental psychopathology perspective on child maltreatment. Annual Review of Psychology, 67, 187–211.

Colder, C. R., Frndak, S., Lengua, L. J., Read, J. P., Hawk, L. W., & Wieczorek, W. F. (2018). Internalizing and externalizing problem behavior: A test of a latent variable interaction predicting a two-part growth model of adolescent substance use. Journal of Abnormal Child Psychology, 46(2), 319–330.

Curran, P. J., West, S. G., & Finch, J. F. (1996). The robustness of test statistics to nonnormality and specification error in confirmatory factor analysis. Psychological Methods, 1, 16–29.

Danielson, C. K., McCart, M. R., de Arellano, M. A., Macdonald, A., Doherty, L. S., & Resnick, H. S. (2010). Risk reduction for substance use and trauma-related psychopathology in adolescent sexual assault victims: Findings from an open trial. Child Maltreatment, 15(30), 261–268.

Danzo, S., Connell, A. M., & Stormshak, E. A. (2017). Associations between alcohol-use and depression symptoms in adolescence: Examining gender differences and pathways over time. Journal of Adolescence, 56, 64–74.

Duprey, E. B., Oshri, A., & Caughy, M. O. (2017). Childhood neglect, internalizing symptoms and adolescent substance use: Does the neighborhood context matter? Journal of Youth and Adolescence, 46(7), 1582–1597.

Eiden, R. D., Lessard, J., Colder, C. R., Livingston, J., Casey, M., & Leonard, K. E. (2016). Developmental cascade model for adolescent substance use from infancy to late adolescence. Developmental Psychology, 52(10), 1619–1633.

Enders, C., & Bandalos, D. (2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling, 8(3), 430–457.

Englund, M. M., & Siebenbruner, J. (2012). Developmental pathways linking externalizing symptoms, internalizing symptoms, and academic competence to adolescent substance use. Journal of Adolescence, 35(5), 1123–1140.

Fava, N. M., Trucco, E. M., Martz, M. E., Cope, L. M., Jester, J. M., Zucker, R. A., & Heitzeg, M. M. (2019). Childhood adversity, externalizing behavior, and substance use in adolescence: Mediating effects of anterior cingulate cortex activation during inhibitory errors. Development and Psychopathology, 31(4), 1439–1450.

Fosco, G. M., Stormshak, E. A., Dishion, T. J., & Winter, C. E. (2012). Family relationships and parental monitoring during middle school as predictors of early adolescent problem behavior. Journal of Clinical Child & Adolescent Psychology, 41(2), 202–213.

Foster, H., & Brooks-Gunn, J. (2009). Toward a stress process model of children’s exposure to physical family and community violence. Clinical Child and Family Psychology Review, 12(2), 71–94.

Gauthier-Duchesne, A., Hébert, M., & Daspe, M.È. (2017). Gender as a predictor of posttraumatic stress symptoms and externalizing behavior problems in sexually abused children. Child Abuse & Neglect, 64, 79–88.

Gonzale, N. A., Liu, Y., Jensen, M., Tein, J. Y., White, R. M. B., & Deardorff, J. (2017). Externalizing and internalizing pathways to Mexican American adolescents’ risk taking. Development and Psychopathology, 29, 1371–1390.

Haynie, D. L. (2002). Friendship networks and delinquency: The relative nature of peer delinquency. Journal of Quantitative Criminology, 18(2), 99–134. Hinnant, J. B., Erath, S. A., Shimizu, M., & El- Sheikh, M. (2019). Parenting, deviant peer affiliation, and externalizing behavior during adolescence: Processes

conditional on sympathetic nervous system reactivity. Journal of Child Psychology and Psychiatry, 60(7), 793–802. Hudson, A., Wekerle, C., Goldstein, A. L., Ellenbogen, S., Waechter, R., Thompson, K., & Stewart, S. H. (2017). Gender differences in emotion-mediated pathways from

childhood sexual abuse to problem drinking in adolescents in the child welfare system. Journal of Child & Adolescent Trauma, 10, 19–28. Johnston, L. D., Miech, R. A., O’Malley, P. M., Bachman, J. G., Schulenberg, J. E., & Patrick, M. E. (2020). Monitoring the Future national survey results on drug use. 1975-

2019: Overview, key findings on adolescent drug use. Institute for Social Research, The University of Michigan. Jones, D. J., Lewis, T., Litrownik, A., Thompson, R., Proctor, L. J., Isbell, P., & Runyan, D. K. (2013). Linking childhood sexual abuse and early adolescent risk

behavior: The intervening role of internalizing and externalizing problems. Journal of Abnormal Child Psychology, 41, 139–150. Knight, E. D., Smith, J. S., Martin, L. M., Lewis, T., & the LONGSCAN Investigators. (2008). Measures for assessment of functioning and outcomes in longitudinal research

on child abuse-volume 3: Early adolescence (ages 12-14). Kobulsky, J. M., Kepple, N. J., & Jedwab, M. (2018). Abuse characteristics and the concordance of child protective service determinations and adolescent self-reports

of abuse. Child Maltreatment, 23(3), 269–280. Langevin, R., Hébert, M., & Cossette, L. (2015). Emotion regulation as a mediator of the relation between sexual abuse and behavior problems in preschoolers. Child

Abuse & Neglect, 46, 16–26. Lansford, J. E., Deater- Deckard, K., Dodge, K. A., Bates, J. E., & Pettit, G. S. (2004). Ethnic differences in the link between physical discipline and later adolescent

externalizing behaviors. Journal of Child Psychology and Psychiatry, 45(4), 801–812. Larrabee, H. M., & Lewis, T. (2014). Longitudinal studies of child abuse and neglect (LONGSCAN) assessments 0–18: User's guide. Ithaca, NY: National Data Archive on

Child Abuse and Neglect, Cornell University. Lewis, T., McElroy, E., Harlaar, N., & Runyan, D. (2016). Does the impact of child sexual abuse differ from maltreated but non-sexually abused children? A prospective

examination of the impact of child sexual abuse on internalizing and externalizing behavior problems. Child abuse & neglect, 51, 31–40. Li, S., Zhao, F., & Yu, G. (2020). Childhood emotional abuse and depression among adolescents: Roles of deviant peer affiliation and gender. Journal of Interpersonal

Violence, 1–21. Masten, A. S., & Cicchetti, D. (2010). Developmental cascades. Development and Psychopathology, 22(3), 491–495. McCoy, S. S., Dimler, L. M., Samuels, D. V., & Natsuaki, M. N. (2019). Adolescent susceptibility to deviant peer pressure: Does gender matter? Adolescent Research

Review, 4(1), 59–71. McDonough, M. H., Jose, P. E., & Stuart, J. (2016). Bidirectional effects of peer relationships and adolescent substance use: A longitudinal study. Journal of Youth and

Adolescence, 45(8), 1652–1663. Mezquita, L., Ibánez, M. I., Moya, J., Villa, H., & Ortet, G. (2014). A longitudinal examination of different etiological pathways to alcohol use and misuse. Alcoholism:

Clinical and Experimental Research, 38(6), 1770–1779. Mrug, S., Hoza, B., & Bukowski, W. M. (2004). Choosing or being chosen by aggressive-disruptive peers: Do they contribute to children’s externalizing and

internalizing problems? Journal of Abnormal Child Psychology, 32(1), 53–65. Negriff, S. (2018). Developmental pathways from maltreatment to risk behavior: Sexual behavior as a catalyst. Development and Psychopathology, 30(2), 683–693. Orth, U., Clark, D. A., Donnellan, M. B., & Robins, R. W. (2021). Testing prospective effects in longitudinal research: Comparing seven competing cross-lagged models.

Journal of Personality and Social Psychology, 120(4), 1013–1034. https://doi.org/10.1037/pspp0000358. Oshri, A., Rogosch, F. A., Burnette, M. L., & Cicchetti, D. (2011). Developmental pathways to adolescent cannabis abuse and dependence: Child maltreatment,

emerging personality, and internalizing versus externalizing psychopathology. Psychology of Addictive Behaviors, 25(4), 634–644. Oshri, A., Sutton, T. E., Clay-Warner, J., & Miller, J. D. (2015). Child maltreatment types and risk behaviors: Associations with attachment style and emotion

regulation dimensions. Personality and Individual Differences, 73, 127–133. Rodriguez-Seijas, C., Arfer, K. B., Thompson, R. G., Jr., Hasin, D. S., & Eaton, N. R. (2017). Sex-related substance use and the externalizing spectrum. Drug & Alcohol

Dependence, 174, 39–46. Runyan, D. K., Curtis, P. A., Hunter, W. M., Black, M. M., Kotch, J. B., Bangdiwala, S.Landsverk, J., … (1998). LONGSCAN: A consortium for longitudinal studies of

maltreatment and the life course of children. Aggression and Violent Behavior, 3(3), 275–285. Siegel, J. P. (2015). Emotional regulation in adolescent substance use disorders: Rethinking risk. Journal of Child & Adolescent Substance Abuse, 24(2), 67–79. Simons-Morton, B., & Chen, R. S. (2006). Over time relationships between early adolescent and peer substance use. Addictive Behaviors, 31(7), 1211–1223. Thompson, E. L., Mehari, K. R., & Farrell, A. D. (2020). Deviant peer factors during early adolescence: Cause or consequence of physical aggression? Child Development,

91(2), e415–e431. Traube, D. E., James, S., Zhang, J., & Landsverk, J. (2012). A national study of risk and protective factors for substance use among youth in the child welfare system.

Addictive Behaviors, 37(5), 641–650.

D. Yoon et al.

Child Abuse & Neglect 137 (2023) 106054

Trim, R. S., Meehan, B. T., King, K. M., & Chassin, L. (2007). The relation between adolescent substance use and young adult internalizing symptoms: Findings from a high-risk longitudinal sample. Psychology of Addictive Behaviors, 21(1), 97–107.

Weiss, N. H., Bold, K. W., Sullivan, T. P., Armeli, S., & Tennen, H. (2017). Testing bidirectional associations among emotion regulation strategies and substance use: A daily diary study. Addiction, 112(4), 695–704.

West, S., Taylor, A., & Wu, W. (2012). Model fit and model selection in structural equation modeling. In R. Hoyle (Ed.), Handbook of structural equation modeling (pp. 209–231). New York, NY: Guilford Press.

Yampolskaya, S., Chuang, E., & Walker, C. (2019). Trajectories of substance use among child welfare-involved youth: Longitudinal associations with child maltreatment history and emotional/behavior problems. Substance Use & Misuse, 54(3), 437–448.

Yoon, D., Yoon, S., Yoon, M., & Snyder, S. M. (2019). Developmental trajectories of deviant peer affiliation in adolescence: Associations with types of child maltreatment and substance use. Children and Youth Services Review, 105, 1–8.

Yoon, D., Snyder, S. M., & Yoon, S. (2020). Child maltreatment types and adolescent substance use: The role of deviant peer affiliation. Child & Family Social Work, 25 (2), 355–363.

Yoon, D., Snyder, S. M., Yoon, S., & Coxe, K. A. (2020). Longitudinal association between deviant peer affiliation and externalizing behavior problems by types of child maltreatment. Child Abuse & Neglect, 109, 1–11.

Yoon, S., Shi, Y., Yoon, D., Pei, F., Schoppe-Sullivan, S., & Snyder, S. M. (2020). Child maltreatment, fathers, and adolescent alcohol and marijuana use trajectories. Substance Use & Misuse, 55(5), 721–733.

Yoon, D., Yoon, S., Pei, F., & Ploss, A. (2021). The roles of child maltreatment types and peer relationships on behavior problems in early adolescence. Child Abuse & Neglect, 112, 1–11.

Zhu, J., Yu, C., Bao, Z., Jiang, Y., Zhang, W., Chen, Y.Zhang, J., … (2017). Deviant peer affiliation as an explanatory mechanism in the association between corporal punishment and physical aggression: A longitudinal study among Chinese adolescents. Journal of Abnormal Child Psychology, 45(8), 1537–1551.

D. Yoon et al.