Parental aggression as a predictor of boys’ hostile attribution across the transition to middle school

Abstract

Aggression among youth is a public health problem that is often studied in the context of how youth interpret social information. Social cognitive factors, especially hostile attribution biases, have been identified as risk factors for the development of youth aggression, particularly across the transition to middle school. Parental behaviors, including parental aggression to children in the form of corporal punishment and other aggressive behavior, have also been linked to aggressive behavior in children at these ages. Despite the important role played by these two risk factors, the connection between the two has not been fully studied in the literature. This study examined the link between parental aggression and children’s hostile attributions longitudinally among a diverse sample of 123 boys as they entered middle school. Results support acceptance of a model in which parental aggression to children prior to entering middle school predicted children’s hostile attributions after the transition to middle school above and beyond that which was predicted by previous levels of hostile attributions. As expected, hostile attributions also predicted change in parent- and teacher-rated child aggression. These findings provide important evidence of the role that parental behavior plays in youth social cognition at this critical age, which has implications for understanding the development of aggressive behavior.

Keywords

Hostile attribution social cognition corporal punishment parenting behaviors

Introduction

In efforts to prevent acts of youth violence, researchers have suggested that a mix of internal social-cognitive processes, including social information processing, and external environmental contexts contribute to aggressive behavior in children (e.g. Crick & Dodge, 1994; Lochman, 2006). Studies of one social cognitive process, hostile attribution bias, indicate that children who interpret social cues as threating and hostile react with more aggressive behavior (see Orobio de Castro, Veerman, Koops, Bosch, & Monshouwer, 2002 for a meta-analysis). One study found that this relation is robust with hostile attributions and other social information-processing problems predicting changes in aggression from eighth to eleventh grade (Lansford et al., 2006). In this same study, elementary school social information-processing problems did not predict aggression in the eleventh grade, suggesting that the relation between these variables may change during that period (Lansford et al., 2006). Evidence indicates that hostile attribution style may develop from hostile schemas in children owing to a variety of factors including physical abuse, adult modeling of hostile attribution bias, failure in life tasks, and rearing in a culture that values self-defense and personal honor (Dodge, 2006). Despite these theoretical underpinnings of hostile attribution bias, the role of parenting in the development of hostile attributions has rarely been directly tested, especially using longitudinal data across middle childhood.

The transition to middle school appears to be a particularly important developmental period, including in the development of hostile attributions. A large meta-analysis found that studies of children ages 8 to 12 years old had larger effect sizes in the relation between hostile attributions and aggressive behavior than younger or older children (Orobio de Castro et al., 2002). Some additional evidence suggests that preadolescence has been linked to a number of variables including decreases in parental monitoring and increases in the hostility of the school environment, child autonomy within the school environment, and peer influence on children (Blyth, Simmons, & Bush, 1978; Eccles et al., 1993; Fite, Colder, Lochman, & Wells, 2006). When taken together, the unique changes of this developmental age suggest a need for additional research on the way that hostile attribution develops in this preadolescent age group.

Another body of literature focuses on the influence of parenting behaviors on child aggression. Parenting behaviors such as poor parental monitoring (e.g. Stattin & Kerr, 2000), inconsistent discipline (e.g. Stormshak, Bierman, McMahon, & Lengua, 2000), and corporal punishment (see Gershoff, 2002) are associated with increased risk for aggressive behaviors. Parent-to-child aggression, especially in the form of physical or corporal punishment, has been repeatedly linked to increased child aggression (Conger, Neppl, Kim, & Scaramella, 2003; Dodge, Pettit, Bates, & Valente, 1995; Gershoff, 2002), but its role in the development of hostile attributions during middle childhood has not been adequately tested.

As two primary risk factors for aggression, parenting factors and social cognitive styles are important in their own right, but most theoretical models suggest that they likely impact one another (Crick & Dodge, 1994; Dodge, 2006; Lochman, 2006). A recent cross-sectional study of parent behavior and child hostile attributions included relational provocation from parents such as psychological control (Nelson & Coyne, 2009). This study found that paternal psychological control related to boys’ instrumental and relational hostile attributions and paternal corporal punishment related to girls’ instrumental hostile attribution (Nelson & Coyne, 2009). A negative relation between paternal corporal punishment and boys’ instrumental hostile attribution was found to have a suppressor effect (Nelson & Coyne, 2009).

To clarify the direction of effects, some studies have used longitudinal methods and tests of indirect effects, or mediation, to show that social information processing in children and later child aggressive behavior is affected by earlier physical abuse, characterized by an adult physically harming a child or the child needing medical attention (Dodge, Bates, & Pettit, 1990; Dodge, Bates, Pettit, & Valente, 1995). For the current study, we are interested in testing the role of ‘parental aggression’ which includes serious physical abuse such as that tested in these studies, but also includes milder, less severe forms of aggression toward children (e.g., corporal punishment; pushing, grabbing, or shoving a child during a parent–child conflict). There are studies that have investigated this (e.g. Weiss, Dodge, Bates, & Pettit, 1992), labeling it as ‘harsh discipline’ or ‘negative behavior’. Two longitudinal studies tested the mediational role of children’s hostile attributions about their mothers’ behavior (MacKinnon-Lewis, Lamb, Hattie, & Baradaran, 2001; MacKinnon-Lewis, Lindsey, Frabutt, & Chambers, 2014). One study found mothers’ negative behaviors (i.e., verbalizations, actions, or facial expressions) in a parent-child interaction task predicted later negative attributions by boys about their mothers, which subsequently predicted aggression by the boys in the interaction task (MacKinnon-Lewis et al., 2001). More recent work by MacKinnon-Lewis and colleagues found that the indirect effect of maternal corporal punishment on maternal–child conflict was mediated by hostile attributions about mothers (MacKinnon-Lewis et al., 2014). Both of these studies provide important information about how corporal punishment and maternal negative behaviors contribute to the development of hostile attributions, but the measures of hostile attributions and child behavior in these studies are specific to scenarios with the child’s mother. The current study endeavors to determine if parents’ aggressive behavior can affect both children’s hostile attributions and their behavior in parent–child context as well as in peer contexts.

Several studies have looked at hostile attributions and aggressive behavior in parent–child and peer contexts. One study tested a similar dual-mediational model using a hostile attribution measure that included peer vignettes (Heidgerken, Hughes, Cavell, & Willson, 2004). They found that the relation between harsh parenting and child aggression was mediated through hostile social goals, hostile attributions, and other social information processing steps. Their model, using second and third-grade participants, was cross-sectional and was not able to test change in variables over time. The role of harsh discipline in the change in aggression over time was examined in another study which identified a mediating role of social information-processing, including hostile attribution and three additional social information-processing steps (Weiss et al., 1992). The current study seeks to replicate these findings while isolating hostile attributions from other types of social information-processes and testing hostile attribution change over time. Whereas Weiss and colleagues (1992) focused on children ages 5 and 6, the current study seeks to explore these variables in the transition to middle school when aggression and social cognitions are becoming more sophisticated and the role of peers and parents are changing (Steinberg & Morris, 2001).

In summary, the current study builds upon previous literature by investigating parental aggression, broadly defined to include physical aggression and corporal punishment, and its effect on hostile attribution bias over time, and child aggression over time. Both hostile attribution and child aggression will be examined with respect to parent–child interactions and peer interactions. Moreover, hostile attribution and child aggression will be examined while controlling for previous levels of the same variables, allowing us to interpret the role of parental aggression on the change in these constructs across a distinct developmental period—the transition to middle school. The indirect effect of parental aggression on later child aggression with child hostile attribution as a mediator will also be tested.

Hypotheses

Given the theoretical and empirical literature supporting the relations between parental behavior and children’s hostile attributions, it is anticipated that parental aggression will predict children’s hostile attributions over time. Similarly, parental aggression is expected to predict increases in children’s aggression across the transition to middle school. The theorized role of parental aggression to predict both hostile attribution and child aggression across time might suggest that hostile attribution could serve as a mediator of the relation between parental and child aggression. The current study which consists of only two timepoints, allows for a half-longitudinal design model of mediation to be tested (Fritz & MacKinnon, 2012).

Method

This study used data from a previous longitudinal evaluation of the Coping Power program collected annually during the summers of six consecutive years (Lochman & Wells, 2002b). The study was conducted in an urban school district of a mid-sized city in the Southeastern United States.

Sample

Participants in the original study were enrolled after being identified using a multiple-gating screening procedure (detailed in Lochman & Wells, 2002b). The screening process asked teachers of fourth- and fifth-grade boys to rate their verbal aggression, physical aggression, and disruptive behavior. The highest rated 22% of the initial screening sample was included in the second screen wherein teachers and parents assessed aggression using the Aggression subscale of the Teacher Report Form (TRF; Achenbach, 1991b) and the Child Behavior Checklist (CBCL; Achenbach, 1991a), respectively. Boys with high scores on both forms were enrolled in the study and randomized to one of three treatment groups: Child Intervention (n = 60), Child and Parent Intervention (n = 60), and Risk Comparison (n = 62). In addition, boys who did not meet the initial teacher-rating cutoff were included as a control in the Non-Risk Comparison group (n = 61). Because the current study did not evaluate effects of the preventive intervention, only participants who did not receive the intervention were included in this project. The Risk Comparison and Non-Risk Comparison groups combined (n = 123) made up the sample for the current project. The final sample included 123 boys with almost exactly half identified as at-risk for aggression (n = 61) and half as non-risk (n = 62). The final sample included 45.5% White, 53.7% African American and 0.8% Other race boys. Caregivers completing the survey consisted mostly of mothers (63.4%) and grandmothers (22.8%). Other caregivers included stepmothers (3.3%), fathers (2.4%), grandfathers (1.6%), uncles (1.6%), and one stepfather (0.8%). Within the sample, the mean income level was between $30,000 and $34,999. Assessments were first conducted at baseline with each child and his parent separately when the child was in the fourth or fifth grade. The ‘Time 2’ data referred to in the current study was collected two years after baseline, when children in the sample had advanced to the sixth or seventh grade.

Attrition was moderately high across the sample, with 22% of participants leaving the study by Time 2 (n = 27). No significant differences existed between attritted and nonattritted participants in levels of child aggression, parental aggression, or child hostile attribution, or in demographic variables at Time 1.

Measures

Child aggression

Child aggression was measured by creating a latent variable using both parent and teacher report of aggressive behavior. For parent-report, the Child Behavior Checklist- Aggressive behavior subscale (CBCL; Achenbach, 1991a) is used. This 20-item scale asked parents to rate their children’s aggressive behaviors as 0 = not true, 1 = somewhat/sometimes true or 2 = very/often true. Examples of aggressive behavior measured on the CBCL included screams a lot and physically attacks people. Higher scores on the scale indicated more parent-rated aggressive and defiant behavior. Internal consistencies of the Aggression subscale of the CBCL were high in each time point in the sample (alpha ranged from 0.86 to 0.91).

For teacher report of child aggression, teachers completed the Teacher Report Form—Aggressive Behavior subscale (TRF; Achenbach, 1991b), which is a 25-item teacher report measure of child aggression as observed at school in the classroom. The 3-point response scale was as follows: 0 = not true (as far as you know), 1 = somewhat or sometimes true, 2 = very true or often true. Items on the aggressive behavior subscale include proactively aggressive behaviors such as teases a lot and more disruptive behaviors such as fidgets. Overall, higher scores on this scale indicated more aggressive and defiant behaviors. The internal consistency of the Teacher Report Form for this study was strong across timepoints, α = 0.95 to 0.96. Research has indicated that the TRF Aggressive behavior subscale is significantly correlated with the CBCL in a national sample (Achenbach & Rescorla, 2001).

Child hostile attribution

The Child Attribution Measure (CAM) used four vignettes of parent–child and child–peer interactions (Lochman & Dodge, 1994; Lochman & Wells, 2002b). The first two vignettes presented conflicts between children and their peers while the last two illustrated child–parent conflicts. For the purposes of this study, only the Attribution subscale was used, as it was the most relevant assessment tool of the hostile attribution bias construct. The Attribution questions asked the child why the peer or mother acted the way that they did. Answers were scored 0 to 2 (0 = accident, 1 = don’t know, 2 = protagonist was angry) with a higher score indicating more hostile attribution bias.

Given that each vignette is thematically different from the others, Chronbach’s alpha is not expected to be an appropriate measure of internal consistency. Instead of using the vignettes as a cohesive scale, the items are used as observed variables to create a latent variable measuring children’s hostile attribution at Time 1 and at Time 2 within the larger structural equation model testing the hypotheses.

Parental aggression

Parental aggression was measured by standardizing and adding scores on two similar measures—the Physical Aggression subscale of the Conflict Tactics Scale–Child (CTS-C; Straus, 1979) and the Corporal Punishment subscale of the Alabama Parenting Questionnaire (APQ; Shelton, Frick & Wootton, 1996). The Physical Aggression subscale of the Conflict Tactics Scale–Child (CTS-C; Straus, 1979) asked responding parents how often they engaged in physically aggressive behaviors in the last year while experiencing a conflict with their child. This scale includes 10 items with response choices on a 7-point scale (never, once, twice, 3–5 times, 6–10 times, 11–20 times, more than 20 times). The Corporal Punishment subscale of the Alabama Parenting Questionnaire (APQ; Shelton et al., 1996)–Parent Version consists of three items with higher scores indicating more corporal punishment. Parents responded on a 5-point scale (Never, Almost Never, Sometimes, Often, and Always). The two standardized scores at Time 1 were then averaged to create a unitary measure of parent-to-child aggression for Time 1. This procedure was repeated at Times 2. The average parent-to-child aggression score focused on the physical aspect of aggression and had moderate internal consistency, α = 0.65 to α = 0.67.

Data Analyses

In order to understand levels of parental aggression, child aggression, and child hostile attributions, descriptive statistics were completed treating each variable as continuous. Correlations between each variable were also completed to explore bivariate relations. In order to test the hypotheses linking variables over time, we conducted structural equation modeling with the observed variable of parental aggression at Time 1 and latent variables measuring child hostile attribution and child aggression at Times 1 and 2 using MPlus 6 software (Muthén & Muthén, 1998–2011). To measure child hostile attribution at Time 1 and Time 2, latent variables were created using four vignettes at each time point as observed variables. To measure child aggression, the observed variables of the CBCL Aggressive behavior subscale raw score and the TRF Aggressive behavior subscale raw score were used in a measurement model creating the latent variable at Times 1 and 2. Each of the Time 1 variables—child aggression, child hostile attribution, and parental aggression—were allowed to correlate in the model, as were CBCL scores at Time 1 and Time 2. The overall fit of the structural equation model was tested using the chi-square test of model fit, the Root Mean Square Error of Approximation (RMSEA), the comparative fit index (CFI), the Tucker Lewis Index (TLI) and the standardized root mean squared residual (SRMR).

Results

The means and standard deviations for the sample (n = 123) are presented in Table 1 and correlations showing cross-sectional and longitudinal relations among the variables are in Table 2. For combining informants of child aggression to create a latent variable, the correlations between the parent-rated CBCL and teacher-rated TRF were examined. They ranged from r = 0.486 to r = 0.504 across time and were significant (p < 0.01).

Table 1.

Means and Standard Deviations of Variables at Times 1 and 2.

				Time 1			Time 2
	Timepoint	Range	n	M	SD	n	M	SD
Child Hostile Attribution Bias	Peer Vignette 1	0–2	122	0.61	0.569	85	0.48	0.548
	Peer Vignette 2	0–2	122	0.69	0.708	86	0.74	0.657
	Parent Vignette 1	0–2	122	0.21	0.548	86	0.12	0.448
	Parent Vignette 2	0–2	122	0.36	0.562	86	0.27	0.471
Child Aggression	Child Behavior Checklist Aggression	0–37	121	11.55	7.619	88	9.57	7.029
Child Aggression	Teacher Report Form Aggression	0–50	120	16.93	12.360	96	12.85	13.786
Parental Aggression^a		−1.43–2.51	121	0	1	n/a	n/a	n/a

Note: n/a notes variables not included in the model. Standardized score averaging Conflict Tactics Scale Respondent to Child Physical Aggression subscale and Alabama Parenting Questionnaire Corporal Punishment subscale.

Table 2.

Correlations Between Variables.

			Child Hostile Attribution Latent Variable		Child Aggression (Teacher Report Form)		Child Aggression (Child Behavior Checklist)		Parental Aggression
			Time 1	Time 2	Time 1	Time 2	Time 1	Time 2	Time 1
Child Hostile Attribution Latent Variable	Time 1	r	–	0.461**	0.136	0.363**	0.080	0.331**	−0.043
		95% CI		(0.309, 0.590)	(−0.044, 0.308)	(0.175, 0.525)	(−0.100, 0.255)	(0.131, 0.505)	(−0.220, 0.137)
		n		123	120	96	121	88	121
	Time 2	r		−	0.116	.000	0.141	0.032	0.646**
		95% CI			(−0.065, 0.289)	(−0.200, 0.200)	(−0.038, 0.312)	(−0.179, 0.240)	(0.528, 0.739)
		n			120	96	121	88	121
Child Aggression (Teacher Report Form)	Time 1	r			0	0.473**	0.504**	0.387**	0.166
		95% CI				(0.298, 0.617)	(0.356, 0.628)	(0.191, 0.553)	(−0.015, 0.337)
		n				93	118	86	118
	Time 2	r				−	0.377**	0.486**	0.093
		95% CI					(.190,0 .538)	(0.291, 0.642)	(−0.111, 0.289)
		n					95	75	95
Child Aggression (Child Behavior Checklist)	Time 1	r					–	0.664**	0.259**
		95% CI						(0.527, 0.767)	(0.084, 0.418)
		n						87	121
	Time 2	r						–	0.124
		95% CI							(−0.089, 0.326)
		n							87
Parental Aggression	Time 1								−
Parental Aggression	Time 1								−

Note: Pairwise deletion used. *p < 0.05, **p < 0.01.

Aggressive Behavior and Hostile Attributions

To test the hypotheses, a model predicting hostile attribution and child aggression across time was analyzed using Mplus 6 software (Muthén & Muthén, 1998–2011; see Figure 1). The model used full information maximum likelihood estimation to deal with missing values due to attrition. Across all variables in the model, an average of 13.9% of data were missing with a range of 0.8% to 30.1% of cases missing depending upon the variables. In order to use full information maximum likelihood estimation, data must be missing completely at random (MCAR). Little’s MCAR test indicated that data are MCAR (C²(112) = 103.035, p = 0.716; Little, 1988).

Figure 1.

Model of Parental Aggression, Child Hostile Attribution and Child Aggression. Note: Standardized path coefficients are presented as measures of effect sizes of direct effects. N = 123; *p < 0.05, **p < 0.01 R² measures of effect size are presented for the two dependent variables. Overall test of model fit: χ²(55) = 70.840, p = 0.074, CFI = 0.921, TLI = 0.889, RMSEA = 0.048, 90% CI(0.000, .078), SRMR = 0.= 0.074. TRF = Teacher Report Form, CBCL = Child Behavior Checklist.

The model appears to be a good fit for the data, χ²(55) = 70.840, p = 0.074, CFI = 0.921, TLI = 0.889, RMSEA = 0.048, 90% CI(0.000, 0.078), SRMR = 0.074. Statistical literature recommends using a combination of the standardized root mean squared residual (SRMR) and another fit index, such as the RMSEA, to measure model fit (Hu & Bentler, 1999). Among samples less than 250, research suggests using a cutoff of 0.06 and lower for RMSEA and 0.08 and lower for SRMR to indicate good model fit (Hu & Bentler, 1999). Using these metrics, the current data fit the model well.

Direct effects in the model included paths predicting hostile attribution at Time 2 from hostile attribution, parental aggression, and child aggression at Time 1, each of which were significant. The R² provides an estimate of the effect size of the independent variables at Time 1 explaining about 36% of the variance in hostile attribution at Time 2 (R² = 0.361). The model also indicated significant direct paths from Time 1 child aggression and Time 1 hostile attribution to Time 2 child aggression. The paths predicting Time 2 child aggression from parental aggression and Time 2 child attribution were not significant. Overall, the amount of variability in Time 2 child aggression explained by the independent variables was 73% (R² = 0.359, 95% CI [−0.059, 0.777]).

Lastly, the half-mediational model of Time 1 parental aggression on Time 2 child aggression through Time 2 hostile attribution was tested by examining the indirect effect of Time 1 parental aggression on Time 2 child aggression. The 95% CI for this indirect effect included zero (95% CI [−0.415, 0.149]) and was nonsignificant (standardized indirect effect = −0.133, p = 0.345). This suggests child hostile attribution does not mediate the effect of parental aggression on child aggression.

Discussion

The results of the current brief report support a model in which parental aggression predicts increases in hostile attribution over time, and hostile attribution predicts increases in aggressive behavior in boys across the transition to middle school. The paths measure change from Time 1 to Time 2, while controlling for Time 1 levels. The findings of this study replicate previous research that has shown a link between parental aggression, social information-processing and child aggression (Weiss et al., 1992; see Orobio de Castro et al., 2002 for a meta-analysis) during a time when young adolescents’ relationships with their parents and peers are changing. Our findings suggest that the link between hostile attributions and child aggression remains strong during this developmental period given the medium to large effect size obtained by the model. This is particularly important given the changing role of both parents and peers in middle school (Eccles et al., 1993) and suggests that intervention during this time could be particularly impactful.

As expected, parental aggression predicted children’s later hostile attributions above and beyond the impact of previous levels of child hostile attributions. Literature suggests that children may learn hostile attributions through parental modeling of attributions as well as physical abuse (Dodge, 2006), but the results of the current study highlight the effect of broad-based parental aggressive behavior, including corporal punishment and aggression outside of discipline. Our results suggest that parental aggression has an impact on the way that children think about their social environment, including in peer situations. These parenting behaviors appear to serve as a risk factor for the development of hostile attribution biases, which may explain part of the reason why parental aggression is a risk factor for child aggression.

Given the strong direct effects between parental aggression and child hostile attribution, and child hostile attribution and child aggression over time, we tested a half-longitudinal (Fritz & MacKinnon, 2012) mediational model to determine if hostile attribution could serve as a mediator between parental and child aggression. The change in hostile attribution measured at Time 2 did not relate to that of child aggression measured at the same timepoint and the test of the indirect effect was not significant. It is possible, that if children were followed for another timepoint, a full mediational model could be tested of the effect of parental aggression on child aggression with hostile attribution as a mediator and find different results.

An important strength of this study is its use of parent and teacher report of child aggression. This means that across the model three different informants were used—child, parent, and teacher, which not only reduces informant bias, but also allows for measuring child aggressive behavior in two contexts—at home with family and at school with peers and teachers. The study may be further improved if we were to use peer nominations or observational measures as have been used in similar studies (MacKinnon-Lewis et al., 2001; Weiss et al., 1992).

It is important to note the limitations of this brief report. Notably, the sample includes only boys and therefore results are not able to be generalized to girls. Despite this, literature suggests that the link between hostile attributions and aggression in youth is similar in girls and boys (Orobio de Castro et al., 2002). Further research should include both male and female youth and should compare results between the two genders. Additional studies of parent and child hostile attributions and aggressive behavior are needed to examine these variables with more varied populations. Moreover, some additional variable, such as SES, parental stress, or child temperament could be an important piece that could explain the relation between parental aggression, hostile attribution, and child aggression. Some previous research indicates that cultural factors including differences among races and ethnicities in their use of corporal punishment could differentially affect outcomes for youth (Lansford, Deater-Deckard, Dodge, Bates, & Pettit, 2004). Further research with a larger sample and across a longer span of time could help to solve some of these questions.

Findings of this brief report could have implications in preventative interventions. Both parental aggressive behaviors and hostile attribution biases are often targeted by interventions to prevent youth behavior problems. In addition, both of these variables have been shown to mediate the impact of interventions on youth aggression (Dodge & Godwin, 2013; Lochman & Wells, 2002a). The results of this study provide some context about these two variables indicating that parental aggression is associated with changes in youth hostile attribution biases. Understanding this relation can help focus elements of preventive intervention on parents to maximize the effect of the intervention on youth aggression prevention.

Footnotes

Funding

This work was supported by the National Institute on Drug Abuse (grant number R01DA008453).

References

Achenbach

T. M.

(1991a). Manual for the child behavior checklist and revised child behavior Profile. Burlington, VT: University Associates in Psychiatry.

Achenbach

T. M.

(1991b). Manual for the teacher’s report form and 1991 profile. Burlington, VT: University of Vermont.

Achenbach

T. M.

Rescorla

L. A.

(2001). Manual for the ASEBA school-age forms & profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families.

Blyth

D. A.

Simmons

R. G.

Bush

(1978). The transition into early adolescence: A longitudinal comparison of youth in two educational contexts. Sociology of Education, 51, 149–162.

Conger

R. D.

Neppl

Kim

K. J.

Scaramella

(2003). Angry and aggressive behavior across three generations: A prospective, longitudinal study of parents and children. Journal of Abnormal Child Psychology, 31(2), 143–160.

Crick

N. R.

Dodge

K. A.

(1994). A review and reformulation of social information-processing mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74–101.

Dodge

K. A.

(2006) Translational science in action: Hostile attributional style and the development of aggressive behavior problems. Developmental Psychopathology, 18, 791–814.

Dodge

K. A.

Bates

J. E.

Pettit

G. S.

(1990). Mechanisms in the cycle of violence. Science, 250(4988), 1678–1683.

Dodge

K. A.

Godwin

(2013). Social-information-processing patterns mediate the impact of preventive intervention on adolescent antisocial behavior. Psychological Science, 24(4), 456–465.

10.

Dodge

K. A.

Pettit

G. S.

Bates

J. E.

Valente

(1995). Social information-processing patterns partially mediate the effect of early physical abuse on later conduct problems. Journal of Abnormal Psychology, 104, 632–643.

11.

Eccles

J. S.

Midgley

Wigfield

Buchanan

C. M.

Reuman

Flanagan

MacIver

(1993). Development during adolescence: The impact of stage-environment fit on young adolescents’ experiences in schools and in families. American Psychologist, 48, 90–101.

12.

Fite

P. J.

Colder

C. R.

Lochman

J. E.

Wells

K. C.

(2006). The mutual influence of parenting and boys’ externalizing behavior problems. Applied Developmental Psychology, 27, 151–164.

13.

Fritz

M. S.

MacKinnon

D. P.

(2012). Mediation models for developmental data. In Laursen

Little

P. T. D.

(Eds.), Handbook of Developmental Research Methods (pp. 291–310). New York, NY: Guilford Press.

14.

Gershoff

E. T.

(2002) Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin, 128, 539–579.

15.

Heidgerken

A. D.

Hughes

J. N.

Cavell

T. A.

Willson

V. L.

(2004). Direct and indirect effects of parenting and children’s goals on child aggression. Journal of Clinical Child and Adolescent Psychology, 33(4), 684–693.

16.

Bentler

P. M.

(1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55.

17.

Lansford

J. E.

Deater-Deckard

Dodge

K. A.

Bates

J. E.

Pettit

G. S.

(2004). Ethnic in the link between physical discipline and later adolescent externalizing behaviors. Journal of Child Psychology and Psychiatry, 45(4), 801–812.

18.

Lansford

J. E.

Malone

P. S.

Dodge

K. A.

Crozier

J. C.

Pettit

G. S.

Bates

J. E.

(2006). A 12-year prospective study of patterns of social information processing problems and externalizing behaviors. Journal of Abnormal Child Psychology, 34(5), 709–718.

19.

Little

R. J.

(1988). A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Association, 83(404), 1198–1202.

20.

Lochman

J. E.

(2006) Translation of research into intervention. International Journal of Behavioral Development, 30, 31–38.

21.

Lochman

J. E.

Dodge

K. A.

(1994). Social-cognitive processes of severely violent, moderately aggressive, and nonaggressive boys. Journal of Consulting and Clinical Psychology, 62, 366–374.

22.

Lochman

J. E.

Wells

K. C.

(2002a). Contextual social-cognitive mediators and child outcome: A test of the theoretical model in the Coping Power program. Development and Psychopathology, 14, 945–967.

23.

Lochman

J. E.

Wells

K. C.

(2002b). The Coping Power program at the middle-school transition: Universal and indicated prevention effects. Psychology of Addictive Behaviors, 16, S40–S54.

24.

MacKinnon-Lewis

Lamb

M. E.

Hattie

Baradaran

L. P.

(2001). A longitudinal examination of the associations between mothers’ and sons’ attributions and their aggression. Development and Psychopathology, 13, 69–81.

25.

MacKinnon-Lewis

Lindsey

E. W.

Frabutt

J. M.

Chambers

J. C.

(2014). Mother–Adolescent conflict in African American and European American families: The role of corporal punishment, adolescent aggression, and adolescents’ hostile attributions of mothers’ intent. Journal of Adolescence, 37(6), 851–861.

26.

Muthén

L. K.

Muthén

B. O.

(1998–2011). Mplus user’s guide (6th ed.). Los Angeles, CA: Muthén & Muthén.

27.

Nelson

D. A.

Coyne

S. M.

(2009). Children’s intent attributions and feelings of distress: Associations with maternal and paternal parenting practices. Journal of Abnormal Child Psychology, 37(2), 223–237.

28.

Orobio de Castro

Veerman

J. W.

Koops

Bosch

J. D.

Monshouwer

H. J.

(2002). Hostile attribution of intent and aggressive behavior: A meta-analysis. Child Development, 73, 916–934.

29.

Shelton

K. K.

Frick

P. J.

Wootton

(1996). Assessment of parenting practices in families of elementary school-age children. Journal of Clinical Child Psychology, 25, 317–329.

30.

Stattin

Kerr

(2000). Parental monitoring: A reinterpretation. Child Development, 71(4), 1072–1085.

31.

Steinberg

Morris

A.S.

(2001) Adolescent development. Annual Review of Psychology, 52, 83–110.

32.

Stormshak

E. A.

Bierman

K. L.

McMahon

R. J.

Lengua

L. J.

(2000). Parenting practices and child disruptive behavior problems in early elementary school. Journal of Clinical Child Psychology, 29(1), 17–29.

33.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage & the Family, 41, 75–88.

34.

Weiss

Dodge

K. A.

Bates

J. E.

Pettit

G. S.

(1992). Some consequences of early harsh discipline: Child aggression and a maladaptive social information processing style. Child Development, 63(6), 1321–1335.