Cognitive Impulsivity as a Mediator of the Parental Knowledge-Childhood Aggression Relationship

Abstract

In a previous study, reactive criminal thinking or cognitive impulsivity mediated the relationship between parental knowledge and delinquency. This study sought to determine whether cognitive impulsivity also mediated the relationship between parental knowledge and childhood aggression. A path analysis was performed on a sample of 438 early adolescent boys (n = 206) and girls (n = 232) from the Illinois Study of Bullying and Sexual Violence using three waves of non-overlapping data. As predicted, cognitive impulsivity mediated the relationship between parental knowledge and childhood aggression, but cognitive insensitivity did not. The results of this study provide ongoing support for the general conceptual argument that childhood aggression parallels delinquency in certain respects and that parental knowledge deters both future delinquency and childhood aggression by reducing the cognitive impulsivity that is central to the behavioral patterns of delinquency and childhood aggression.

Keywords

parental knowledge cognitive impulsivity childhood aggression

It has become evident that childhood aggression and other early externalizing behaviors, like animal cruelty, may parallel or overlap delinquency in both form and process (Longobardi & Badenes-Ribera, 2019; Morcillo et al., 2015). Similarity in form was noted by Baldry and Farrington (2000) who discovered that childhood aggression and delinquency were moderately correlated and shared some of the same risk factors in common. Similarity in process, however, occurs when delinquency and aggression serve analogous functions as part of a larger group of variables. An example would be mediation of crime and aggression continuity. Walters (2016), seeking insight into the nature of crime continuity, discovered that while reactive (impulsive, reckless, emotional) criminal thinking mediated crime continuity (past crime → future crime), proactive (planned, calculated, amoral) criminal thinking did not. In a follow-up to this study, Walters and Espelage (2018) ascertained that cognitive impulsivity, which is marked by weak effortful control and can be considered the developmental antecedent to reactive criminal thinking, mediated aggression continuity (past aggression → future aggression), whereas cognitive insensitivity, which is marked by a clear absence of interpersonal concern and can be considered the developmental antecedent to proactive criminal thinking, did not.

A further example of similarity in process would be mediation of the parental knowledge-delinquency/aggression relationship. Research indicates that parental control, of which parental knowledge is a facet, predicts future delinquency (Hoeve et al., 2009) and childhood aggression (Lösel & Bender, 2014). Walters (2020a) proposed that reactive criminal thinking/cognitive impulsivity mediated the relationship between parental knowledge and delinquency and uncovered support for this supposition in a study of early adolescent male and female schoolchildren. The reason why cognitive impulsivity was proposed as a link between parental knowledge and delinquency is that parental control, knowledge, and monitoring can be considered constraints on a child’s natural tendency to act impulsivity and seek immediate gratification, which, if left unchecked, can give rise to delinquency. The same process could occur with aggression, which, like delinquency, can be viewed as a partially impulsive behavioral pattern. Although aggression perpetration would seem to be largely proactive, a recent sequential process growth mixture modeling analysis revealed that pure aggression perpetration was rare and that most aggressive children had a history of aggression victimization (Walters, 2020b). Hence, impulsivity may be just as much a part of aggression as it is of delinquency.

The purpose of the current investigation was to extend the results of the Walters (2020a) investigation to aggression perpetration in an effort to determine whether a process similar to the one found for parental knowledge, reactive criminal thinking, and delinquency generalized to the relationship between parental knowledge, cognitive impulsivity, and aggression perpetration. Most of the control variables from the previous Walters (2020a) probe were included in the current investigation (age, sex, race, family structure, and parent education). One difference was that there was no measure of unsupervised routine activities, which was included as a control variable in the Walters (2020a) study based on its well-established association with parental knowledge and delinquency (Walters, 2018), in the data base for this study. As such, the unsupervised routine activities variable was replaced by aggression victimization, a well-known correlate of parental knowledge and aggression perpetration (Stavrinides, Nikiforou, & Georgiou, 2015). Following from the results of the previous Walters (2020a) investigation, it was hypothesized that cognitive impulsivity, the developmental antecedent to reactive criminal thinking, would mediate the parental knowledge-childhood aggression relationship after age, sex, race, family structure, parent education, and aggression victimization were controlled, but that cognitive insensitivity, the developmental antecedent to proactive criminal thinking, would not.

Method

Participants

Participants for this study were 438 members of the Illinois Study of Bullying and Sexual Violence (ISBSV: Espelage, Low, Anderson, & De La Rue, 2014), a seven-wave longitudinal study that ran from Spring 2008 until Spring 2013. As a result of study attrition and a rolling matriculation approach, there was a moderately high degree of missing data. The current sample was constructed by eliminating all cases at and above the missing data peak. Because there were 2 to 15 times more cases with three out of 13 missing items than at any other point on the missing items plot, the cutoff for inclusion in this study was two or fewer missing items. The average age of participants at Wave 1 of the ISBSV was 12.33 years (SD = 0.87, Range = 10-15) and the sample was composed of 206 boys and 232 girls. The racial/ethnic breakdown was 31.5% White, 51.4% Black, 3.2% Hispanic, and 13.9% other. An early adolescent sample was employed because this is the developmental period during which inhibitory control is most clearly connected to future delinquency (Fosco, Hawk, Colder, Meisel, & Lengua, 2019), yet it is still several years before the graduated transition from cognitive impulsivity/insensitivity to reactive/proactive criminal thinking takes place (Walters, 2017).

Measures

Parental knowledge and monitoring was assessed with eight items rated by participants (“parents ask about homework”; “parents know if you come home on time”; “parents know where you are and who you are with”; “ rules are clear”; “clear rules about alcohol and drug use”; “parents know about alcohol use”; “ parents know about handgun possession”; “parents know if you skip school”) on a 4-point scale (0 = never, 1 = seldom, 2 = often, 3 = always). When summed, these items produced a total score that could range from 0 to 24. The perceived parental knowledge scale was designed specifically for the ISBSV and possessed good internal consistency in the current sample of participants (α = .86).

Cognitive impulsivity was evaluated with four self-report items (“I have a hard time sitting still”; “I start things but have a hard time finishing them”; “I do things without thinking”; “I need to use a lot of self-control to keep out of trouble”), with each item rated on a 5-point scale (0 = never, 1 = seldom, 2 = sometimes, 3 = often, 4 = always) to yield a score that could range from 0 to 16. The internal consistency of this scale at Waves 1 and 2 of the ISBSV was adequate (α = .75-.77). This scale was first used to evaluate impulsivity in a study by Bosworth, Espelage, and Simon (1999).

Cognitive insensitivity was appraised with the University of Illinois Positive Attitudes toward Bullying Scale (POSATT: Espelage & Asidao, 2001). The POSATT consists of three self-report items (“a little teasing doesn’t hurt anyone”; “I don’t care what mean things people say, as long as it’s not about me”; “If other students are being teased too much, it’s not my problem”) rated on a 4-point Likert-type scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree) and summed to produce a total score with a range of 3 to 12. The internal consistency of this three-item scale was modest at Waves 1 and 2 of the ISBSV (α = .61-.68).

Aggression perpetration was measured using the Illinois Bully Scale (IBS: Espelage & Holt, 2001), a self-report inventory that asked participants to indicate the frequency with which they engaged in the following nine behaviors over the past 30 days (“upset other students”; “teased other students in a group”; “spread rumors”; “started arguments”; “harassed other students”; “threatened another student”; “encouraged people to fight”; “teased other students”; “mean when angry”) using a 5-point frequency scale (0 = never, 1 = one or two times, 2 = three or four times, 3 = five or six times, 4 = seven or more times). Item scores are combined to produce a total score that can range from 0 to 36. This score achieved good internal consistency at Waves 1 and 3 of the ISBSV (α = .86-.88). Bullying has traditionally been defined by a power imbalance between the victim and perpetrator (Olweus, 1993), but this is not how the IBS measures bullying, partly because of problems associated with operationalizing the term power imbalance and partly because doing so would require multiple follow-up questions. As such, behaviors measured by the IBS are referred to in this article as aggression perpetration and victimization.

Six control variables were included in this study: age (in years), sex (1 = male, 2 = female), race (1 = white, 2 = non-white), family structure (1 = live in a two-parent home, 0 = do not live in a two-parent home), mean parent education (1 = less than high school, 2 = high school diploma or GED, 3 = some college, 4 = graduated from college, 5 = some graduate school, 6 = graduate or professional degree), and aggression victimization. Aggression victimization was assessed with four items from the ISB (“other students picked on me”; “other students called me gay”; “other students called me names”; “I got hit and pushed by other students”) considered over the past 30 days. Each item was rated on a 5-point scale (0 = never, 1 = one or two times, 2 = three or four times, 3 = five or six times, 4 = seven or more times) and summed to produce a total score that could range from 0 to 16 (α = .78). It should also be noted that each outcome variable was lagged by including its precursor measure (Cog Impulsivity-1, Cog Insensitivity-1, Aggression Perpetration-1) in the equation predicting that particular outcome.

Research Design and Data Analysis

This study employed a three-wave fixed-sample longitudinal panel design. The independent (parental knowledge), control (age, sex, race, family structure, parent education, and aggression victimization), and precursor (Cog Impulsivity-1, Cog Insensitivity-1, Aggression Perpetration-1) variables were measured at Wave 1; the mediator variables (Cog Impulsivity-2, Cog Insensitivity-2) were measured at Wave 2; and the dependent variable (Aggression Perpetration-3) was measured at Wave 3. The waves were administered 6 months apart and there was no overlap between them, thus preserving the prospective status of the data.

A path analysis was performed in MPlus 8.3 (Muthén & Muthén, 2017) using a maximum likelihood with robust parameters and standard errors (MLR) estimator. The MLR estimator was required to allow for the inclusion of auxiliary variables in the analysis. Because bootstrapping is not possible with an MLR estimator, the total indirect effects in this study were tested against a 95% confidence interval (CI) created using Preacher and Selig’s (2012) Monte Carlo Method for Assessing Mediation (MCMAM). MCMAM was performed with 20,000 repetitions and the CI was considered significant if it did not include zero.

Sensitivity testing was performed in order to rule out omitted variable and endogenous selection bias. Kenny’s (2013) “failsafe ef” procedure, (r_my.x) × (sd_m.x) × (sd_y.x)/(sd_m) × (sd_y), was used to test for omitted variable bias. The “failsafe ef” coefficient indicates how well an unobserved covariate confounder would need to correlate with the mediating and dependent variables to eliminate the b path of a significant indirect effect. Endogenous selection bias, which can occur when an outcome measure is conditioned on a precursor measure (Elwert & Winship, 2014), was evaluated by removing all precursor measures and redoing the analyses.

Missing Data

Approximately three quarters of the sample had complete data on all 13 variables (75.3%), whereas 15.5% of participants were missing data on one variable and 9.1% were missing data on two variables. The proportion of missing data points across the entire sample was 2.6% and four variables had more than 5% missing data: Parental Monitoring (9.1%), Cognitive Impulsivity-2 (9.6%), Cognitive Insensitivity-2 (8.2%), and Aggression Perpetration-3 (14.2%). Missing data were handled with full information maximum likelihood (FIML), a procedure that computes model parameters and standard errors by maximizing the likelihood function of all non-missing data. To further enhance the precision of FIML (Collins, Schafer, & Kam, 2001), 25 auxiliary variables were added to the analysis (Social Support-1/2/3, Aggression Perpetration-2, Aggression Victimization-2/3, Anger-1/2/3, Cog Impulsivity-3, Cog Insensitivity-3, Depression-1/2/3, Participant Delinquency-1/2/3, Peer Delinquency-1/2/3, Parental Knowledge-2/3, Hostility-1/2/3). It should be noted that auxiliary variables are only used to improve FIML and are not part of the actual analysis.

Results

Descriptive statistics for the 13 variables included in this study, along with an inter-variable correlational matrix, can be found in Table 1. Collinearity diagnostics uncovered no evidence of multicollinearity between predictor variables in any of the three regression equations: tolerance = .797 to .975; variance inflation factor = 1.026 to 1.255. Table 2 and Figure 1 summarize the results of the three-equation path analysis. Consistent with the research hypothesis tested in this study, Wave 2 cognitive impulsivity mediated the Wave 1 parental knowledge-Wave 3 aggression perpetration relationship (MCMAM 95% CI = [−0.6509, −0.0042]), whereas Wave 2 cognitive insensitivity did not (MCMAM 95% CI = [−0.1653, 0.1107]).

Table 1.

Descriptive Statistics and Correlations for the 13 Variables Included in This Study.

Variable	n	X–	SD	Range	2	3	4	5	6	7	8	9	10	11	12	13
1. Age	438	12.33	0.87	10-15	−.08	.06	−.09	−.10	.04	−.07	.12	.06	.23†	.11	.20†	.10
2. Sex	438	0.53	—	0-1	—	−.03	−.03	−.02	−.04	.11	.05	.06	−.15	−.12	−.04	−.07
3. Race	438	1.68	—	1-2	—	—	−.14	−.14	−.06	−.13	−.02	−.05	.21†	.18†	.16	.05
4. Family Structure	438	0.40	—	0-1	—	—	—	.12	−.07	.07	−.16	−.12	−.07	−.01	−.08	−.12
5. Parent Education	437	3.06	1.18	1-6	—	—	—	—	.03	.04	−.06	−.08	−.07	−.07	−.02	−.01
6. Aggression Victimization	435	2.35	3.26	0-16	—	—	—	—	—	−.06	.31†	.27†	−.06	−.08	.22†	.17
7. Parental Knowledge	437	18.66	6.02	0-24	—	—	—	—	—	—	−.08	−.14	−.12	−.10	−.22†	−.21†
8. Cognitive Impulsivity-1	437	5.11	3.87	0-16	—	—	—	—	—	—	—	.58†	.13	.14	.29†	.24†
9. Cognitive Impulsivity-2	396	4.73	3.65	0-16	—	—	—	—	—	—	—	—	.11	.08	.25†	.42†
10. Cognitive Insensitivity-1	438	6.98	2.19	3-12	—	—	—	—	—	—	—	—	—	.46†	.21†	.17
11. Cognitive Insensitivity-2	402	6.45	2.32	3-12	—	—	—	—	—	—	—	—	—	—	.18†	.23 †
12. Aggression Perpetration-1	436	4.01	5.51	0-31	—	—	—	—	—	—	—	—	—	—	—	.46†
13. Aggression Perpetration-3	376	4.14	5.87	0-36	—	—	—	—	—	—	—	—	—	—	—	—

Note. Variable = variable name; Age = chronological age in years; Sex = male (0) vs. female (1); Race = White (1) vs. non-White (2); Family Structure = 2-level scale of family structure (1 = live in a two-parent home, 0 = do not live in a two-parent home); Parent Education = parent education rated on a 6-point scale and averaged across both parents; Aggression Victimization = aggression victimization at Wave 1; Parental Knowledge = parental knowledge at Wave 1; Cognitive Impulsivity-1 = cognitive impulsivity at Wave 1; Cognitive Impulsivity-2 = cognitive impulsivity at Wave 2; Cognitive Insensitivity-1 = cognitive insensitivity at Wave 1; Cognitive Insensitivity-2 = cognitive insensitivity at Wave 2; Aggression Perpetration-1 = aggression perpetration at Wave 1; Aggression Perpetration-3 = aggression perpetration at Wave 3; Range = range of scores in current sample.

†

p < .00064 (Bonferroni-corrected alpha: .05/.78 correlations).

Table 2.

Three-Equation Path Analysis of Cognitive Impulsivity and Cognitive Insensitivity as Mediators of the Parental Knowledge-Child Delinquency Relationship.

Variables	b [95% CI]	β	Z	p
Outcome = Cog Impulsivity-2
Age	−0.067 [–0.408, 0.274]	−0.016	−0.38	.701
Sex	0.233 [–0.356, 0.821]	0.032	0.78	.439
Race	−0.578 [–1.177, 0.021]	−0.073	−1.89	.059
Family Structure	−0.169 [–0.776, 0.439]	−0.023	−0.54	.586
Parent Education	−0.169 [–0.433, 0.096]	−0.054	−1.25	.211
Aggression Victimization	0.101 [0.000, 0.201]	0.090	1.97	.049
Parental Knowledge	−0.596 [–1.174, –0.017]	−0.098	−2.02	.044
Cog Impulsivity-1	0.492 [0.407, 0.577]	0.520	11.33	<.001
Outcome = Cog Insensitivity-2
Age	−0.016 [–0.270, 0.238]	−0.006	−0.12	.901
Sex	−0.237 [–0.648, 0.174]	−0.051	−1.13	.259
Race	0.439 [0.007, 0.872]	0.088	1.99	.046
Family Structure	0.087 [–0.323, 0.498]	0.018	0.42	.677
Parent Education	−0.047 [–0.222,0.127]	−0.024	−0.53	.594
Aggression Victimization	−0.039 [–0.097, 0.019]	−0.055	−1.33	.183
Parental Knowledge	−0.069 [–0.472, 0.334]	−0.018	−0.34	.737
Cog Insensitivity-1	0.447 [0.346, 0.548]	0.419	8.67	<.001
Outcome = Aggression Perpetration-3
Age	−0.035 [–0.573, 0.503]	−0.005	−0.13	.898
Sex	−0.542 [–1.565, 0.480]	−0.047	−1.04	.298
Race	−0.292 [–1.389, 0.804]	−0.024	−0.52	.601
Family Structure	−0.674 [–1.642, 0.295]	−0.058	−1.36	.173
Parent Education	0.148 [–0.236, 0.532]	0.030	0.76	.450
Aggression Victimization	0.012 [–0.191, 0.214]	0.007	0.11	.910
Parental Knowledge	−0.616 [–1.639, 0.407]	−0.065	−1.18	.238
Cog Impulsivity-2	0.501 [0.294, 0.708]	0.319	4.74	<.001
Cog Insensitivity-2	0.309 [0.092, 0.526]	0.126	2.80	.005
Aggression Perpetration-1	0.354 [0.207, 0.501]	0.339	4.71	<.001
Cog Imp-2 with Cog Ins-2	0.048 [-0.566, 0.661]	0.008	0.15	.879

Note. Outcome = outcome measure; Age = chronological age in years; Sex = male (0) vs. female (1); Race = White (1) vs. non-White (2); Family Structure = 2-level scale of family structure (1 = live in a two-parent home, 0 = do not live in a two-parent home); Parent Education = parent education rated on a 6-point scale and averaged across both parents; Aggression Victimization = aggression victimization at Wave 1; Parental Knowledge = parental knowledge at Wave 1; Cog Impulsivity-1 = cognitive impulsivity at Wave 1; Cog Impulsivity-2 = cognitive impulsivity at Wave 2; Cog Insensitivity-1 = cognitive insensitivity at Wave 1; Cog Insensitivity-2 = cognitive insensitivity at Wave 2; Aggression Perpetration-1 = aggression perpetration at Wave 1; Aggression Perpetration-3 = aggression perpetration at Wave 3; with = covariance; b [95% CI] = unstandardized coefficient and 95% symmetric confidence interval (in square brackets); β = standardized coefficient; Z = Wald Z test statistic; p = significance level of the Wald Z test statistic, N = 438.

Figure 1.

Path analysis of the relationships between Wave 1 parental knowledge and Wave 3 aggression perpetration via the mediating effects of Wave 2 cognitive impulsivity and cognitive insensitivity.

The “failsafe ef” sensitivity test revealed that an unobserved covariate confounder would need to correlate .39 with Cog Impulsivity-2 and .39 with Aggression Perpetration-3, controlling for Parental Knowledge-1 and Cog Impulsivity-2 in the case of Aggression Perpetration-3, to completely neutralize the significant b path of the indirect effect. This indicates that the results were highly robust to the effects of omitted variable bias. Endogenous selection bias was also tested as a viable alternative explanation for the current results. When precursor measures were removed from all equations, the path coefficients increased rather than decreased, a finding inconsistent with endogenous selection bias or a collider effect.

Discussion

The research hypothesis tested in this study predicted that cognitive impulsivity, the developmental progenitor of reactive criminal thinking, but not cognitive insensitivity, the developmental progenitor of proactive criminal thinking, would mediate the relationship between parental knowledge and aggression perpetration. Not only did this hypothesis find support in this study, but the results were highly robust to omitted variable bias and there was no evidence of endogenous selection bias or a collider effect. What is more, the reactive effects surfaced even after aggression victimization was controlled. Hence, just as parental knowledge appears to protect children against future delinquency by reducing reactive criminal thinking, it also appears to protect a child against future aggression perpetration by reducing cognitive impulsivity. This particular outcome is consistent with the notion, first introduced in Walters (2020a), that parental control, as represented in the present study by perceived parental knowledge, protects the child from future conduct disordered behavior by countering the effects of a process, cognitive impulsivity, that is an inherent part of human nature and which children can learn to control through parental instruction and supervision (Gottfredson & Hirschi, 1990). Given the low amounts of aggression and cognitive impulsivity that characterized the current sample of participants, a replication study on participants with higher levels of aggressive behavior and aggression-supporting attitudes would seem to be advisable.

There are a number of practical implications to the current results. First, they suggest that parents play a vital role in the future aggressive activities of their offspring, just as they play a vital role in the future delinquent activities of their offspring. Parenting training has been found to be effective in alleviating, and, in some cases, eliminating delinquent behavior (Piquero et al., 2016). It has been used much less often with childhood aggression, but the results of at least one study suggest that it may be just as effective in reducing aggressive behavior as it is in reducing delinquent behavior (Burkhart, Knox, & Brockmyer, 2013). Most bullying/aggression programs are based in school and research indicates that both aggression perpetration and victimization drop by 20% to 23% in students enrolled in school-based antibullying programs (Farrington & Ttofi, 2009). Perhaps the most important practical benefit or policy implication of this study is that it illustrates how parental knowledge deters future aggression perpetration by reducing cognitive impulsivity in the child. Cognitive impulsivity, in fact, accounted for over 10% of the total amount of variance in subsequent aggression, after controlling for age, sex, race, family structure, parent education, and prior aggression victimization and perpetration. Future research is required to identify the child skills that parental knowledge and other facets of parental control promote in countering the natural state of cognitive impulsivity. This way, clinicians and educators should be able to find more effective ways of imparting these skills to children through both parenting training and school-based programming.

One potential limitation of this study is the use of parental knowledge to estimate parental control. Stattin and Kerr (2000) have argued that the counter-delinquency and likely counter-aggression effects of parental knowledge have less to do with active parental control (i.e., monitoring, supervision, and surveillance) and more to do with passive receipt of information derived from a child’s willingness to self-disclose personal information to their parents. If true, this suggests that parental knowledge has as much to do with parental support as it does with parental control. Other researchers, however, have demonstrated that parental knowledge predicts offspring misconduct, even when child self-disclosure is controlled (Fletcher, Steinberg, & Williams-Wheeler, 2004; Lahey, Van Hulle, D’Onofrio, Rodgers, & Waldman, 2008). There is also evidence that child self-disclosure is partially a function of active parenting (Tokić & Pećnik, 2010). A second limitation of this study is the small zero-order correlations that surfaced between the two cognitive measures, cognitive impulsivity and cognitive insensitivity (r = .08-.13). This is several times smaller than what has traditionally been observed when reactive criminal thinking/cognitive impulsivity and proactive criminal thinking/cognitive insensitivity are correlated (r = .35-.70: Walters, 2017). Because the cognitive impulsivity items were very similar to the items used in prior research, the problem may lie with the cognitive insensitivity measure which assessed attitudes specific to childhood aggression and was composed of just three items. A third limitation of this study is that all of the variables were measured with participant self-report which could have artificially inflated the path coefficients between the independent, mediating, and dependent variables through shared method variance (Shadish, Cook, & Campbell, 2002).

In conjunction with the previously mentioned Walters (2020a) investigation, this study demonstrates how delinquency and aggression perpetration follow a similar course in response to parental knowledge and reactive criminal thinking/cognitive impulsivity. Both delinquency and childhood aggression are the result of an indirect effect that runs from parental knowledge to reactive criminal thinking to delinquency or from parental knowledge to cognitive impulsivity to aggression. There are at least two ways to test whether childhood aggression is a developmental antecedent to delinquency. First, researchers can examine the possibility that childhood aggression acts as a developmental antecedent to delinquency by comparing risk factors for each. Second, researchers can explore how these two constructs function within a common set of variables. Hence, in Walters (2020a), parental knowledge was indirectly linked to delinquency through the mediating effect of reactive but not proactive criminal thinking. In this study, parental knowledge was indirectly linked to aggression perpetration through the mediating effect of cognitive impulsivity but not cognitive insensitivity. Additional research is required to identify the skills parental knowledge and other forms of parental control instill in a child, which then aid them in combating reactive criminal thinking and its developmental antecedent, cognitive impulsivity, and thus reducing the future likelihood of the child engaging in delinquency and aggressive behavior. Finally, it may be helpful to test whether these results generalize to physical bullying and aggression given that the dependent variable used in the current investigation (IBS) focused exclusively on verbal and relational aggression.

Footnotes

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research for the current study was supported by the Centers for Disease Control & Prevention (#1U01/CE001677) to D.L.E. (PI).

ORCID iD

Glenn D. Walters

Author Biographies

Glenn D. Walters, PhD, is a professor in the Department of Criminal Justice at Kutztown University where he teaches classes in corrections, criminology, and research methods. His principal research interests include offender management, mediation analysis, and the connection between bullying and delinquency.

Dorothy L. Espelage, PhD, is the William C. Friday Distinguished Professor of Education at the University of North Carolina at Chapel Hill in Chapel Hill, North Carolina. She has authored over 140 peer-reviewed journal articles, five edited books, and 30 chapters on bullying, homophobic teasing, sexual harassment, dating violence, and gang violence.

References

Baldry

A. C.

Farrington

D. P.

(2000). Bullies and delinquents: Personal characteristics and parental styles. Journal of Community and Applied Social Psychology, 10, 17-31.

Bosworth

Espelage

D. L.

Simon

T. R.

(1999). Factors associated with bullying behavior in middle school students. Journal of Early Adolescence, 19, 341-362.

Burkhart

K. M.

Knox

Brockmyer

(2013). Pilot evaluation of the ACT Raising Safe Kids program on children’s bullying behavior. Journal of Child and Family Studies, 22, 942-951.

Collins

L. M.

Schafer

J. L.

Kam

C. M.

(2001). A comparison of inclusive and restrictive strategies in modern missing data procedures. Psychological Methods, 6, 330-351.

Elwert

Winship

(2014). Endogenous selection bias: The problem of conditioning on a collider variable. Annual Review of Sociology, 40, 31-53.

Espelage

D. L.

Asidao

C. S.

(2001). Conversations with middle school students about bullying and victimization: Should we be concerned? In Geffner

Loring

Young

(Eds.), Bullying behaviors: Current issues, research, and interventions (pp. 123-142). Binghamton, NY: Haworth Press.

Espelage

D. L.

Holt

M. K.

(2001). Bullying and victimization during early adolescence: Peer influences and psychosocial correlates. In Geffner

Loring

Young

(Eds.), Bullying behaviors: Current issues, research, and interventions (pp. 49-62). Binghamton, NY: Haworth Press.

Espelage

D. L.

Low

S. K.

Anderson

De La Rue

(2014). Bullying, sexual, and dating violence trajectories from early to late adolescence (Report submitted to the National Institute of Justice Grant #2011-MU-FX-0022). Washington, DC: National Institute of Justice.

Farrington

D. P.

Ttofi

M. M.

(2009). Reducing school bullying: Evidence-based implications for policy. Crime and Justice, 38, 281-345.

10.

Fletcher

A. C.

Steinberg

Williams-Wheeler

(2004). Parental influences on adolescent problem behavior: Revisiting Stattin and Kerr. Child Development, 75, 781-796.

11.

Fosco

W. D.

Hawk

L. W.

Colder

C. R.

Meisel

S. N.

Lengua

L. J.

(2019). The development of inhibitory control in adolescence and prospective relations with delinquency. Journal of Adolescence, 76, 37-47.

12.

Gottfredson

M. R.

Hirschi

(1990). A general theory of crime. Stanford, CA: Stanford University Press.

13.

Hoeve

Dubas

J. S.

Eichelsheim

V. I.

van der Laan

P. H.

Smeenk

Gerris

J. R. M.

(2009). The relationship between parenting and delinquency: A meta-analysis. Journal of Abnormal Child Psychology, 37, 749-775.

14.

Kenny

D. A.

(2013). Mediation: Sensitivity analysis [webinar]. Retrieved from http://davidakenny.net/webinars/Mediation/Sensitivity/Sensitivity.html

15.

Lahey

B. B.

Van Hulle

C. A.

D’Onofrio

B. M.

Rodgers

J. L.

Waldman

I. D.

(2008). Is parental knowledge of their adolescent offspring’s whereabouts and peer associations spuriously associated with offspring delinquency? Journal of Abnormal Child Psychology, 36, 807-823.

16.

Longobardi

Badenes-Ribera

(2019). The relationship between animal cruelty in children and adolescent and interpersonal violence: A systematic review. Aggression and Violent Behavior, 46, 201-211.

17.

Lösel

Bender

(2014). Aggressive, delinquent, and violent outcomes of school bullying: Do family and individual factors have a protective function? Journal of School Violence, 13, 59-79.

18.

Morcillo

Ramos-Olazagasti

M. A.

Blanco

Sala

Canino

Bird

Duarte

C. S.

(2015). Socio-cultural context and bulling others in childhood. Journal of Child and Family Studies, 24, 2241-2249.

19.

Muthén

(2017). Mplus user’s guide (8th ed.). Los Angeles, CA: Author.

20.

Olweus

(1993). Bullying at school: What we know and what we can do. Malden, MA: Blackwell.

21.

Piquero

A. R.

Jennings

W. G.

Diamond

Farrington

D. P.

Tremblay

R. E.

Welsh

B. C.

Reingle Gonzalez

J. M.

(2016). A meta-analysis update on the effects of early family/parent training programs on antisocial behavior and delinquency. Journal of Experimental Criminology, 12, 229-248.

22.

Preacher

K. J.

Selig

J. P.

(2012). Advantages of Monte Carlo confidence intervals for indirect effects. Communication Methods and Measures, 6, 77-98.

23.

Shadish

W. R.

Cook

T. D.

Campbell

D. T.

(2002). Experimental and quasi-experimental designs for generalized causal inference. Boston, MA: Houghton Mifflin.

24.

Stattin

Kerr

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

25.

Stavrinides

Nikiforou

Georgiou

(2015). Do mothers know? Longitudinal associations between parental knowledge, bullying, and victimization. Journal of Social and Personal Relationships, 32, 180-196.

26.

Tokić

Pećnik

(2010). Parental behaviors related to adolescents’ self-disclosure: Adolescents’ views. Journal of Social and Personal Relationships, 28, 201-222.

27.

Walters

G. D.

(2016). Proactive and reactive criminal thinking, psychological inertia, and the crime continuity conundrum. Journal of Criminal Justice, 46, 45-51.

28.

Walters

G. D.

(2017). Modelling the criminal lifestyle: Theorizing at the edge of chaos. London, England: Palgrave Macmillan.

29.

Walters

G. D.

(2018). Unsupervised routine activities as a mediator of the parent knowledge–delinquency relationship. Justice Quarterly, 35, 871-894.

30.

Walters

G. D.

(2020a). Parental control as a stimulus for youth control: Mediating the parental knowledge–child delinquency relationship with reactive criminal thinking. Crime & Delinquency. Advance online publication. doi:10.1177/0011128720962449

31.

Walters

G. D.

(2020b). Trajectories of bullying victimization and perpetration in Australian school children and their relationship to future delinquency and conduct problems. Psychology of Violence. Advance online publication. doi:10.1037/vio0000322

32.

Walters

G. D.

Espelage

D. L.

(2018). Cognitive insensitivity and cognitive impulsivity as mediators of bullying continuity: Extending the psychological inertia construct to bullying behavior. School Psychology Quarterly, 33, 527-536.