By-gender risk paths of parental psychological control effects on emerging adult overt and relational aggression

Abstract

We modeled by-gender risk paths of maternal, paternal, and parental psychological control effects on emerging adult overt and peer relational aggression using structural equation models. The effect of adolescent conduct problems on emerging adults’ (aged 18–24) aggression was examined through its effect on parents’ psychological control. Additionally, a social–cognitive mediation model of aggression was tested linking conduct problems and parents’ psychological control with hostility and low guilt and concern for harming others. Results highlighted the salience of the social–cognitive model for female emerging adults’ aggression risk. For example, maternal/paternal psychological control and hostility mediated effects of conduct problems on female aggression types. Additionally, combined parental psychological control and hostility mediated effects of conduct problems on female aggression types.

Keywords

Emerging adulthood gender overt aggression parenting relational aggression structural equation modeling

As socioeconomic demands push back the clock on educational and social role transitions, the typical transition to adulthood has been delayed as well. As Arnett first documented, a majority of 18- to 25-year-old emerging adults do not consider themselves to be adults (Arnett, 2000) and their parents agree (Nelson et al., 2007). As a result, parents face a challenge in maintaining close, supportive ties while granting their emerging adult children increasing autonomy within the parent–adult relationship (Aquilino, 2006). Not surprisingly, therefore, parents’ chosen methods of support and control continue to affect their sons’ and daughters’ emerging adult adjustment.

During emerging adulthood, parents’ use of psychological control may impede on their sons’ and daughters’ growing sense of autonomy and lead to behavioral maladjustment. Psychological control is comprised of controlling and hostile behaviors including love withdrawal, invalidating feelings, and personal attack (Barber & Harmon, 2002). Unlike behavioral control that is focused on regulating and supervising specific behaviors and effective in moderation, psychological control is designed to regulate and negate emotions and opinions of the adult child (Barber & Harmon, 2002). The intrusive and hostile nature of psychological control not only interferes with youth’s development of autonomy (Allen, Hauser, Eickholt, Bell, & O′Connor, 1994) but also promotes youthful externalizing problems in response to intrusion on their autonomy (i.e., aggression and rule breaking; Hoeve, Dubas, Eichelsheim, Van der Laan, Smeenk & Gerris, 2009). In fact, meta-analyses report a moderate effect size between parents’ psychological control and delinquency (Hoeve et al., 2009), and a small effect size between parents’ psychological control and peer relational aggression (Kuppens, Laurent, Heyvaert, & Onghena, 2012). However, less is known about parental psychological control effects on behavioral adjustment in emerging adulthood, when youth’s quest for autonomy and egalitarian relationships with parents peak (Nelson et al., 2007; Nelson, Padilla-Walker, Christensen, Evans, & Carroll, 2011).

The current study builds on prior child–adolescent research by delineating gender differences in maternal and paternal psychological control effects on emerging adult aggression (Kuppens, Grietens, Onghena, & Michiels, 2009a; Rogers, Buchanan, & Winchell, 2003). Current research acknowledges that parenting influences may vary across the four parent–youth dyads (mother–daughter, mother–son, father–daughter, and father–son) because of differences in the nature of dyad relationships. For example, mothers’ relationships with adolescent daughters are higher in closeness and conflict than their relationships with sons, and their closeness to daughters increases during the emerging adult transition (Rice & Mulkeen, 1995; Steinberg & Silk, 2002). In contrast, the adolescent father–daughter relationship is particularly distant during adolescence, whereas adolescent sons garner more support from their fathers (Steinberg & Silk, 2002). These differences in specific parent–youth relationships potentially portend to differential parental psychological control effects across youth gender. Specifically, female youth may be more sensitive to maternal parenting than male youth because of the more intense, conflicted nature of the mother–daughter relationship. Additionally, adolescent daughters may be more vulnerable to paternal psychological control, given the distant nature of the father–daughter relationship (Rogers et al., 2003). In line with these theories, the current study considered whether maternal and paternal psychological control is more likely to be associated with increases in female versus male emerging adult aggression.

Two forms of aggression that are known to be associated with parents’ use of psychological control were selected for the current study: overt and peer relational aggression. Peer relational aggression includes behaviors that reflect use and manipulation of peer relationships including the spread of malicious rumor and gossip, social exclusion, and threats of affiliation (Crick, 1995). In contrast, overt aggression comprises direct verbal and physical acts against peers. Both forms of aggression are a liability in emerging adulthood when peer relationships are increasingly salient for adaptation (Arnett, 2000). Although male and female emerging adults report comparable rates of peer relational aggression, relational aggression is the aggression of choice for females, as they are far less likely to use overt aggression than males (Eagly & Steffen, 1986; Murray-Close, Ostrov, Nelson, Crick, & Coccaro, 2010). The current study’s focus on emerging adult gender differences in parenting influences therefore contributes to understanding of gender-linked relational aggression risk paths.

Prior studies that examine gender moderation of paternal and maternal psychological control effects on child/adolescent aggression find that effects are conditioned on the gender mix of the parent–child dyad. Both maternal and paternal psychological control were related to girls’ relational aggression in multiple studies of children (Casas et al., 2006; Kuppens et al., 2009a; Nelson, Hart, Yang, Olsen, & Jin, 2006) and one study of adolescents (Soensens, Vansteenkiste, Goossens, Duriez, & Niemiec, 2008). However, two child studies reported significant associations between paternal, but not maternal, psychological control and girls’ relational aggression (Nelson & Crick, 2002; Nelson et al., 2006). In contrast, only one study of children and one study of adolescents linked maternal and paternal psychological control with increases in boys’ relational aggression (Kuppens et al., 2009a; Soensens et al., 2008). Gender differences in the effects of maternal psychological control on child overt aggression are also evident. Two studies found that maternal, but not paternal, psychological control was related to girls’ overt aggression (Casas et al., 2006; Hart, Nelson, Robinson, Olsen, & McNeilly-Choque, 1998), whereas only one study related maternal psychological control with young boys’ overt aggression (Hart et al., 1998). All in all, maternal psychological control has been more consistently associated with girls’, versus boys’, increased overt and relational aggression, and paternal psychological control is shown to be particularly important for the development of daughters’ relational aggression.

Less is known about emerging adult gender differences in the effects of maternal and paternal psychological control on overt and peer relational aggression. The current study addressed this gap in the literature by examining gender moderation of maternal and paternal psychological control effects on emerging adult aggression (see Figure 1, Panel 1). Additionally, the combined parental psychological control effect on emerging adult aggression was examined to determine whether the influence of two parents’ psychological control would contribute to emerging adult aggression across gender (Kuppens et al., 2009a). Research is scarce on this point; combined parental control has been associated with increases in children’s relational aggression (Kuppens et al., 2009a) and girls’ overt aggression (Nelson et al., 2006).

Figure 1.

Theoretical model.

Beyond the salience of parental psychological control effects, prior child/adolescent misbehavior may raise risk for emerging adult aggression. Child–adolescent conduct problems, characterized by overt aggression and rule breaking, have been associated with elevated overt aggression (Achenbach, Howell, McConaughy, & Stanger, 1995) and with elevated relational aggression (Murray-Close et al., 2010; Ostrov & Houston, 2008) in emerging adulthood. Therefore, to specify both parental and adolescent effects on emerging adult aggression, the current study examined the risk for emerging adult aggression from adolescent conduct problems (see Figure 1, Panel 1).

An equally plausible explanation of emerging adult aggression is that adolescent conduct problems contribute indirectly to emerging adult aggression by affecting parental psychological control processes (see Figure 1, Panel 1). Accordingly, Albrecht and Galambos (2007) found that the “child effect” of prior physical aggression predicted adolescent perceptions of maternal and paternal psychological control as well as physical aggression prospectively. Their findings suggested that parents may respond to serious adolescent misbehavior with increased psychological control. In turn, parents’ increase in psychological control may promote continuity in youths’ aggression. For example, Kuppens et al. found prospective links between child relational aggression, maternal psychological control, and subsequent relational aggression (Kuppens, Grietens, Onghena, & Michiels, 2009b). The current study therefore examined the potential that prior conduct problems affect emerging adult aggression through associations with maternal, paternal, and combined parental psychological control. In addition, we examined whether mediation via maternal psychological control on overt aggression and via paternal psychological control on relational aggression would be greater for females, given that the literature supports stronger corresponding direct effects of these variables for females (Casas et al., 2006; Nelson et al., 2006).

Potential explanations for links between parental psychological control and emerging adult aggression

Beyond evaluating competing risk paths in emerging adult aggression, a second goal of the current study was to test potential social–cognitive explanations of parental psychological control effects on emerging adult aggression. Both social learning and information processing theories have been advanced to explain the effects of psychological control on the development of overt and relational aggression (Kuppens et al., 2009a; Reed, Goldstein, Morris, & Keyes, 2008). From the social learning perspective, youth learn relational aggression from their direct experience of their parents’ psychological control, given the similarity of these strategies. Both relational aggression and psychological control involve relationship manipulation to control emotional and psychological responses of others (Reed et al., 2008). Youth may also learn overt aggression from the hostile punitive components of psychological control; personal attack serves as a model of verbal insult, directly evokes anger, and may lead to use of overt peer aggression (Casas et al., 2006). Further, social information processing theory suggests that youth’s personal experience of intrusive and hostile psychological control promotes tendencies to attribute hostile intent in ambiguous situations and therefore select overt and relationally aggressive conflict responses (Gomez, Gomez, DeMello, & Tallent, 2001; Nelson & Coyne, 2009).

In line with social learning and information processing theories, the current study model proposed that hostility mediates the effects of parents’ psychological control on emerging adult aggression (see Figure 1, Panel 2). Hostility, which is characterized by suspicion, resentment, and negative views of others, has been associated with perceiving aggressive intent in others in ambiguous situations and readiness for aggression (Kazdin, Rodgers, Colbus, & Siegel, 1987; Petzel & Michaels, 1973). Not surprisingly, therefore, hostility has been specifically linked with young adult overt and peer relational aggression (Loeber, 1990; Murray-Close et al., 2010; Verona, Patrick, & Lang, 2002). From a social learning perspective, parental psychological controlling behaviors, like personal attack, directly model hostility to youth. Alternatively, youth’s experience of being the target of psychological control may promote suspicious, negative views of others’ intentions and aggressive responding in peer conflict situations (Nelson & Coyne, 2009). Consistent with these theories, maternal psychological control was related to boys’ hostile intent attributions and overt aggression (Gomez et al., 2001) and paternal psychological control was linked with boys’ hostile intent attributions (Nelson & Coyne, 2009). It is also noteworthy that hostile bias to relationally provocative events predicts relational aggression among females who have been relationally victimized (Mathieson et al., 2011; Murray-Close, 2011). In a similar manner, parental psychological control may particularly relate to increased female relational aggression via the development of hostile tendencies.

Another way that parental psychological control may affect youth’s aggression is through its influence on socialization of guilt, particularly among females. From a social learning perspective (Bandura, 1979), the experience of guilt serves as a negative expectancy that dampens motive to reaggress. Consistent with this notion, Hoffman (1977) found that mature, humanistic guilt, stemming from concern for the effect of one’s behavior on others, engenders children’s reparation and inhibits their aggressive transgressions. Similarly, guilt is related to lower child aggression (Kochanska, Barry, Jimenez, Hollatz, & Woodard, 2009) and lower adolescent peer relational aggression (Onishi, Kawabata, Kurokawa, & Yoshida, 2012). Further, impoverished guilt, or lack of guilt, is an integral component of callous traits that are associated with emerging adult overt and relational aggression (Czar, Dahlen, Bullock, & Nicholson, 2010; Pardini & Loeber, 2008). Moreover, moral development research and theory suggest that parental psychological control dampens healthy guilt development. Multiple studies show that parental use of love withdrawal relates to reduced humanistic guilt and moral responding in children (Brody & Shaffer, 1982; Hoffman, 1977). Additionally, parents’ use of invalidating feelings and personal attack likely generates angry resentment toward parents rather than promoting adult children’s guilt and concern for others (Kochanska & Aksan, 2006). In fact, parental psychological control was linked with lower guilt among female emerging adults (Woien, Emst, Patock-Peckham, & Nagoshi, 2003). In turn, guilt may explain women’s reduced aggressive tendencies compared to men. For example, Eagly and Steffen’s (1986) meta-analysis showed that adult women’s greater tendency to anticipate guilt as a consequence of overt and psychological aggression mediated their reduced propensity to physically aggress compared to men. The current study, therefore, considered whether female gender moderated the proposed mediation effect of reduced guilt and concern for harming others on the prediction of aggression from parental psychological control.

Explicating adolescent effects

Not only parenting, but youthful conduct problems may directly and indirectly affect expression of hostility, impoverished guilt, and resulting aggressive tendencies in emerging adulthood (see Figure 1, Panel 2). As early as childhood, conduct problems are associated with expression of hostility (Kazdin et al., 1987). Moreover, child aggression, a component of conduct problems, has been linked with hostile attributional tendencies prospectively (Godleski & Ostrov, 2010) and adolescent and adult hostility has been associated with emerging adult aggression (Loeber, 1990; Murray-Close et al., 2010). Just as adolescents may learn hostile responding from family relations, antisocial participation may contribute to the tendency to formulate hostile intent attributions. So, too, Pardini and Loeber (2008) found that adolescent conduct problems were associated with impoverished guilt prospectively, which in turn was associated with emerging adult aggression. Beyond direct effects, adolescent conduct problems may contribute indirectly to emerging adult hostility, lack of guilt, and aggression by affecting parents’ use of psychological control (Figure 1, Panel 2). Given that prior research affirms links between prior aggression and youths’ perceptions of the psychological control of their parents (Albrecht & Galambos, 2007), we tested the potential that psychological control of parents of disruptive adolescents would relate to a negative cycle of hostility, low guilt, and high aggression in emerging adulthood. Additionally, we tested the hypothesis that those mediation effects would be greater for females than males, given that we expected that mediation effects of maternal, paternal, and parental psychological control on aggression types via hostility and low guilt would be greater for females.

Study goals and hypotheses

The current study’s first goal was to extend prior child/adolescent research by examining gender differences in maternal, paternal, and parental psychological control influences on emerging adult overt and peer relational aggression. Additionally, we simultaneously tested the potential that the “adolescent effect” of prior conduct problems would contribute to high levels of emerging adult aggression by increasing parents’ psychological control. Several interrelated hypotheses were therefore tested in by-gender models (see Figure 1, Panel 1). The first hypothesis was that prior adolescent conduct problems would be associated with high levels of emerging adult overt and peer relational aggression. Second, it was hypothesized that the positive effects of (1) paternal psychological control on relational aggression and (2) maternal psychological control on overt aggression would be greater for female versus male emerging adults. Third, it was hypothesized that combined parental psychological control would relate to overt and peer relational aggression across gender. Finally, we tested mediation of adolescent conduct problems on overt and peer relational aggression via maternal, paternal, and parental psychological control. We hypothesized that mediation effects via maternal psychological control on overt aggression and via paternal psychological control on relational aggression would be greater for females.

The second study goal was to examine gender differences in social–cognitive mediation of adolescent conduct problems and parental psychological control effects on emerging adult aggression (see Figure 1, Panel 2). First, we expected that hostility and guilt and concern for harming others would mediate effects of maternal, paternal, and combined parental psychological control on overt and peer relational aggression. Further, it was hypothesized that the mediation effects would be greater for female than male emerging adults. Second, it was expected that adolescent conduct problems would relate to emerging adult aggression through association with hostility and low levels of guilt and concern for harming others for both males and females. Finally, we tested the hypothesis that conduct problems would relate to emerging adult aggression by associations with high maternal, paternal, and parental psychological control, high hostility, and impoverished guilt and concern for harming others. Additionally, we hypothesized that these mediation path effects would be greater for females than for males.

Method

Sample

The sample consisted of 379 college students between the age of 18 and 24 (M = 19 years). The sample was balanced on gender and was diverse with respect to race and ethnicity; 32% of participants described themselves as White, non-Hispanic, 45% described themselves as Hispanic/Latino, and 22% were of another racial/ethnic minority. More than half (61%) of participants reported that one or more of their parents had graduated from college. Sixty-five percent of participants reported an intact family structure characterized by two married biological parents living together. Only 5% of the sample reported that they did not have either a male or a female caregiver.

Recruitment

Participants were recruited from introductory psychology courses at a large Southwestern state university serving a diverse student body. Students between the age of 17 and 24 were invited to participate in the study and signed up for participation via a web-based subject pool service. The study was described as a study of gender differences in aggression.

Measurement

The current study employed by-gender latent variable models in which observed variables were parceled. Parceling optimized power by reducing the number of observed variables used to estimate latent constructs. To facilitate valid parceling, exploratory factor analyses in structural equation modeling (SEM) were used to affirm the dimensionality of study constructs (Kishton & Widaman, 1994). Additionally, preliminary nested measurement invariance (MI) tests were used to ensure that the configural factor structure and factor loadings of latent variables were invariant across gender (Cheung & Rensvold, 2002). All preliminary dimensionality and MI tests are summarized in Appendix 1. Each latent construct was measured by three parcels computed from the average of multiple items. Item to construct balancing was used to select items for parcels of unidimensional constructs so that low, moderate, and high factor loadings were well represented across parcels (Little, Cunningham, & Shahar, 2002). For multidimensional constructs, domain representative parceling was used to ensure that all domains of the constructs were represented across parcels (Kishton & Widaman, 1994).

Demographics

Participants’ racial/ethnic status was ascertained from two questions with forced choice responses; one focused on racial background and the other focused on Hispanic/Latino ethnicity. Participants were instructed to provide the educational status of their current caregivers, whether biological or stepparents or another situation. Maternal/paternal acquired education was measured from two parallel questions using a 9-point scale (1 = less than grade school to 9 = professional or graduate school degree). Parental education was then coded as the highest level of education accrued by either parent. Given that the resulting variable was nonnormally distributed, the variable was censored to a dummy variable indicating whether accrued parent education was college degree (1) or less than college degree (0). A dummy code indicating intact family structure was coded from a question with 10 response choices. Response choices documented whether participants’ families were characterized by two biological parents living together (i.e., ‘intact’; coded as ‘1’), divorced single parents, divorced remarried parents, one or more deceased parents, or never married single parents.

Aggression

Overt aggression was measured from six physical aggression items from the social relational aggression behavior measure (SRASBM; Linder, Crick, & Collins, 2002) and four verbal aggression items from Little et al.’s measure of aggression (Little, Henrich, Jones, & Hawley, 2003). Peer relational aggression was measured from 10 SRASBM items. A 7-point Likert-type response scale was used for all items (1 = strongly disagree to 7 = strongly agree). Reliability of overt items (α = .82, M interitem r = .34) and relational items (α = .80, M interitem r = .27) was good in this sample.

Parental psychological control

Maternal, paternal, and parental psychological control was measured by the widely used 8-item Psychological Control Scale–Youth Self-Report (PCS-YSR; Barber, 1996; Barber & Harmon, 2002). Parallel forms were used to measure maternal and paternal psychological control. Participants were asked to think about their current male and female caregivers (e.g., biological parent or stepparent, etc) when responding to these questions. Item responses were measured using a 4-point Likert-type agreement scale (1 = never to 4 = always). Items comprise multiple components of psychological control including personal attack (“Blame you for other family members’ problems?”), love withdrawal (“Less friendly with you if you do not see things her way?”), invalidating feelings (“Try to change how you feel or think about things?”), and verbally constraining behaviors (“Change the subject, whenever you have something to say?”). Given that the maternal and paternal psychological control scales were highly correlated, r = .48, p < .01, a latent composite measure of parental psychological control was calculated using weighted sums of corresponding paternal and maternal items within each parcel. α Reliabilities of the maternal, paternal, and parental psychological control item sets were good (maternal scale α = .78, M interitem r = .31; paternal scale α = .85, M interitem r = .42; and combined parental scale α = .85, M inter-item r = .26).

Adolescent conduct problems

Adolescent conduct problems were measured from the Antisocial Personality Disorder section of the Personality Diagnostic Questionnaire (PDQ-4), which includes 15 items related to adolescent conduct problems (e.g., burglary, threatening someone with a weapon) prior to the age of 15 (Bagby & Farvolden, 2004). A three-point frequency response scale was used (0 = never, 1 = once, and 2 = two or more times), with wording adapted for minor acts (e.g., violating curfew, lying, and truancy) to (0 = never, 1 = sometimes, and 2 = often). Items responses were then coded as present (1) or absent (0) based on frequency responses. Specifically, serious acts were coded as present if participants responded with “once” or “two or more times.” Minor acts were coded as present if participants responded with “often.” Present codes were then summed for a total adolescent conduct problems score. The total score was censored to a range of 0–3 in accord with clinically significant criteria of conduct disorder. The resulting score was near normally distributed. α Reliability of the conduct problems items was acceptable (α = .73, M interitem r = .22). Slightly less than half (45%) of participants reported one or more conduct problems.

Guilt and concern for harming others

Seven items from the Inventory of Callous-Unemotional Traits (ICU; Essau, Sasagawa, & Frick, 2006) were used to measure guilt and concern for harming others (see Appendix 2). Items focused on guilt for transgressions (“I feel bad or guilty when I do something wrong.”) and associated concern for harming others (“I do not care who I hurt to get what I want,” reverse scored). Item responses were measured on a 4-point Likert-type scale (1 = not at all true to 4 = definitely true). Internal consistency was not optimal (α = .68, M interitem r = .25).

Hostility

A 7-item Hostility scale from the Buss-Perry Anger Questionnaire (BPAQ; Buss & Perry, 1992) was used to measure hostility. Items focus on suspicion, jealousy, and beliefs that others are acting in a hostile way (e.g., “I know that friends talk about me behind my back.”). Item responses were measured using a 5-point Likert-type agreement scale (1 = extremely uncharacteristic of me to 5 = extremely characteristic of me). α was .70 and the M interitem r was .26.

Procedure

After reading and signing internal review board approved consent forms, participants completed self-report measures on laptop computers using survey software in small private rooms. Study personnel provided participants general instruction on completing the computer-based survey measures. Current study measures were embedded in a larger survey in the following nonvarying construct order: aggression, hostility, guilt, conduct problems, and parents’ psychological control. Participants completed the full survey in 120–190 min. Participants received assignment credit in exchange for study participation. Data were collected over four semesters.

Analyses

All study analyses were completed as multigroup SEMs in MPLUS v. 5 software using a robust maximum likelihood estimator, which provides chi-square fit statistics and standard errors that are robust to violations of multivariate normality (Muthén & Muthén, 2008). Full information maximum likelihood estimation was used to manage missing data; 11% of participants had missing data on one or more observed variables. Parameters that did not differ across gender and were not necessary for estimating structural effects (e.g., factor loadings) were held equal across gender. Multimediator models began with an estimation of the initial c, or total, effects of the independent variable (IV) on the dependent variables (DVs). Mediation effects were estimated from models that included (1) the “a” direct effects of the IVs on the mediators, (2) the “b” direct effects of the mediators on the DVs, and (3) the final direct c′ effects of the IVs on the DVs. Significance of the indirect effects (a × b) was used to determine whether mediation was true. Nonsignificant final direct c′ effects support full mediation. Bias-corrected bootstrapping was used to establish 95% confidence intervals of indirect effects (MacKinnon, 2008). Standardized effect sizes of indirect effects were calculated from a × b/SD_Y (MacKinnon, 2008). Although initial significance of the total effect is of interest for causal tests of mediation, mediation effects may exist when the total effect is nonsignificant, particularly when using multimediator models (MacKinnon, 2008, p. 50). Indirect effects were therefore calculated regardless of the significance of the initial total effect. Wald difference tests were used to test gender moderation of the magnitude of direct and indirect effects.

Results

Preliminary tests of univariate normality showed that the distribution of study variables was within acceptable limits (skew < 3 and kurtosis < 8; Kline, 2005). Additionally, preliminary by-gender confirmatory factor analysis models allowing factor loadings of all latent constructs to vary showed that standardized factor loadings were acceptable (>.40 across gender). Further, regression in SEM revealed that females reported fewer adolescent conduct problems, less overt aggression, and more guilt than males (all ps < .05; see Table 1). Means of adolescent conduct problems, overt aggression, and guilt were therefore allowed to vary across gender in SEMs. Finally, regression in SEM did not suggest that parental education, intact two-parent intact family structure, Latino ethnicity, or other non-White racial–ethnic status was associated with overt or peer relational aggression (all ps > .05).

Table 1.

Means of study constructs by gender.

Construct	Females	Males
	M (SD)	M (SD)
Overt aggression	1.63 (.70)	1.96 (.87)
Relational aggression	2.28 (.82)	2.17 (.75)
Maternal psychological control	1.49 (.42)	1.42 (.26)
Paternal psychological control	1.42 (.45)	1.42 (.41)
Parental psychological control	1.51 (.40)	1.48 (.33)
Guilt and concern for harming others	3.42 (.37)	3.31 (.38)
Hostility	2.64 (.57)	2.82 (.57)
Adolescent conduct problems	.50 (.81)	.98 (1.09)

Note. Boldface means are significantly different across gender (p < .05).

Adolescent conduct problems, parents’ psychological control, and emerging adult aggression

The effect of prior adolescent conduct problems on emerging adult aggression was tested first in a by-gender multigroup SEM. Both male and female emerging adults who reported more adolescent conduct problems reported higher levels of overt and relational aggression (see the total c effect in Table 2). Two additional by-gender multigroup SEMs were used to examine gender differences in the effects of maternal, paternal, and parental psychological control on emerging adult aggression with effects of conduct problems on parents’ psychological control and aggression controlled (see Table 2, b effect column). Results revealed that paternal psychological control was significantly associated with relational aggression across gender and with female overt aggression and showed marginal nonsignificant associations with male overt aggression. Also, size of paternal effects did not vary across gender; thus disconfirming our hypothesis (Wald dif test, ps > .05). Contrary to hypotheses, maternal psychological control showed a marginally nonsignificant relation with female overt aggression and nonsignificant relations with female relational aggression and male aggression subtypes. Additionally, as expected, a third by-gender model revealed that combined parental psychological control was significantly associated with male and female overt and relational aggression and the size of effects did not vary by gender (Wald dif tests, ps > .05).

Table 2.

By-gender mediation of adolescent conduct problems on aggression by maternal, paternal, and combined parental psychological control.

Independent variable (IV)	Mediator	Dependent variable (DV)	Total or c effect Initial IV > DV b	a Effect IV > mediator b	b Effect Mediator > DV b	c ′ Effect Final IV> DV b	Indirect effect a × b b	95% CI		Effect size a × b/ SD _Y
Independent variable (IV)	Mediator	Dependent variable (DV)	Total or c effect Initial IV > DV b	a Effect IV > mediator b	b Effect Mediator > DV b	c ′ Effect Final IV> DV b	Indirect effect a × b b	LL	UL	Effect size a × b/ SD _Y
Model 2: Males
Adolescent conduct problems	Maternal psychological control	Overt aggression	.16*	.05	.49	.12†	.02	−.015	.099	.02
Adolescent conduct problems	Paternal psychological control	Overt aggression	.16*	.02	.52^†	.12^†	.01	−.012	.075	.01
Adolescent conduct problems	Maternal psychological control	Relational aggression	.09*	.05	.13	.07†	.01	−.028	.052	.00
Adolescent conduct problems	Paternal psychological control	Relational aggression	.09*	.02	.83*	.07†	.02	−.028	.083	.03
Model 2: Females
Adolescent conduct problems	Maternal psychological control	Overt aggression	.39*	.16*	.24†	.30*	.04	.000	.114	.05
Adolescent conduct problems	Paternal psychological control	Overt aggression	.39*	.12*	.40*	.30*	.05*	.006	.123	.06
Adolescent conduct problems	Maternal psychological control	Relational aggression	.21*	.16*	.14	.13	.02	−.010	.088	.03
Adolescent conduct problems	Paternal psychological control	Relational aggression	.21*	.12*	.53*	.13	.06*	.014	.143	.08
Model 3: Males
Adolescent conduct problems	Parental psychological control	Overt aggression	.16*	.04	.87*	.12†	.03	−.005	.100	.04
Adolescent conduct problems	Parental psychological control	Relational aggression	.09*	.04	1.02*	.06	.04	−.008	.101	.05
Model 3: Females
Adolescent conduct problems	Parental psychological control	Overt aggression	.39*	.15*	.52*	.30*	.08*	.031	.153	.10
Adolescent conduct problems	Parental psychological control	Relational aggression	.21*	.15*	.57*	.12	.09*	.032	.183	.12

Note. Initial total effect was derived from Model 1. Model fit indices were the χ², CFI, and RMSEA. All models showed good fit: Model 1: χ²(df = 29) = 39.1, CFI = .99, RMSEA = .04; Model 2: χ²(df = 123) = 179.21*, CFI = .97, RMSEA = .05; Model 3: χ²(df = 68) = 88.39*, CFI = .99, RMSEA = .04. CFI = comparative fit index; RMSEA = root mean square error of approximation.

†p < .10; *p < .05.

By-gender mediation of adolescent conduct problems on aggression via maternal, paternal, and parental psychological control was evaluated next by inspecting indirect effects in the second and third SEMs (see Table 2). Contrary to hypotheses, maternal psychological control did not mediate the conduct problem effects on aggression types across gender (all ps > .05). However, consistent with expectation, mediation effects of conduct problems on female, but not male, overt and peer relational aggression by paternal psychological control were significant, although the magnitude of indirect effects did not vary by gender (Wald dif tests, ps > .05). Unexpectedly, mediation effects of conduct problems on aggression types via parental psychological control were also significant for females only. Further, the effect of conduct problems on combined parental psychological control was significantly greater for females than males (Wald dif test, p < .05). Additionally, final c′ direct effects suggested that paternal and combined parental psychological control fully mediated the conduct problem effect on female peer relational aggression but partially mediated the conduct problem effect on female overt aggression.

Gender differences in social–cognitive mediation of parental psychological control on aggression

By-gender mediation of maternal, paternal, and parental psychological control on aggression via hostility and guilt and concern for harming others was examined in the fourth and fifth SEMs. Figures 2 and 3 summarize unstandardized direct effects, and Table 3 summarizes total and indirect effects of the models. As hypothesized, hostility mediated female, but not male, effects of paternal psychological control on overt and peer relational aggression, and effect sizes of indirect effects were larger for females than males. Although the magnitude of indirect effects did not vary by gender (Wald dif tests, ps > .05), the direct effect of hostility on relational aggression was greater for females than males (Wald dif test, p < .05). In contrast, paternal psychological effects on overt and relational aggression were not mediated by guilt and concern for harming others across gender. Finally, nonsignificant direct c′ effects suggested full mediation of paternal psychological control on female overt and peer relational aggression by hostility. Results also revealed significant indirect effects of maternal psychological control on overt and relational aggression via increased hostility across gender, although the corresponding initial total effects were nonsignificant. Unexpectedly, the direct effect of maternal psychological control on hostility was greater for males than females (Wald dif test p < .05). Consistent with the hypothesis, maternal psychological control was indirectly associated with female, but not male, overt aggression via low guilt and concern for harming others, but magnitude of the indirect effect did not vary by gender (Wald dif test p > .05).

Figure 2.

Male and female direct effects in the mediation of adolescent conduct problems and maternal/paternal psychological control on overt and relational aggression via guilt and concern for harming others and hostility. Underlined parameters are significantly different at p < .05; †p < .10; *p < .05.

Figure 3.

Male and female direct effects in the mediation of adolescent conduct problems and parental psychological control on overt and relational aggression via guilt and concern for harming others and hostility. Underlined parameters are significantly different at p < .05; †p < .10; *p < .05.

Table 3.

By-gender mediation of maternal, paternal, and parental psychological control on overt and peer relational aggression via hostility and guilt and concern for harming others.

IV	Mediator	DV	Females						Males
			Total c Effect	c ′ Effect	Indirect effect	95% CI		Effect size	Total c effect	c ′ Effect	Indirect effect	95% CI		Effect size
			b	b	b	LL	UL	B	b	b	b	LL	UL	B
Model 4
Maternal psychological control	Hostility	Overt aggression	.24†	−.06	.21*	.067	.491	.26	.49	.10	.32*	.045	1.03	.40
	Guilt		.24^†	−.06	.07*	.001	.308	.09	.49	.10	.07	−.334	.400	.09
	Hostility	Relational aggression	.14	−.20	.26*	.086	.569	.33	.13	−.23	.34*	.080	.935	.44
	Guilt		.14	−.20	.07	−.003	.283	.10	.13	−.23	.03	−.154	.197	.04
Paternal psychological control	Hostility	Overt aggression	.40*	.20	.20*	.036	.485	.25	.52^†	.37	.04	−.073	.277	.05
	Guilt		.40*	.20	−.00	−.174	.077	−.00	.52†	.37	.10	−.097	.562	.12
	Hostility	Relational aggression	.53*	.28	.26*	.034	.651	.33	.83*	.75*	.04	−.107	.264	.05
	Guilt		.53*	.28	−.00	−.131	.077	−.00	.83*	.75*	.05	−.053	.227	.06
Model 5
Parental psychological control	Hostility	Overt aggression	.52*	.10	.36*	.160	.702	.44	.87*	.47	.28*	.060	.693	.35
	Guilt		.52*	.10	.06	.000	.245	.07	.87*	.47	.14	−.017	.458	.17
	Hostility	Relational aggression	.57*	.08	.44*	.200	.807	.56	1.02*	.71*	.25*	.027	.538	.32
	Guilt		.57*	.08	.05	−.001	.200	.06	1.02*	.71*	.06	−.005	.210	.08

Note. Guilt refers to guilt and concern for harming others. Indirect effect is a × b. Effect size of the indirect effect is a × b/SD _Y. Overall model fit for all models was good (Model 4: χ²(df = 275) = 354.22*, CFI = .97, RMSEA= .04; Model 5: χ²(df = 193) = 250.79*, CFI = .97, RMSEA= .04). IV = independent variable; DV = dependent variable; c effect = the effect of the IV on the DV; c′ = the effect of the IV on the DV with the mediator in the model; CFI = comparative fit index; RMSEA = root mean square error of approximation; df = degree of freedom.

†p < .10; *p < .05.

Consistent with expectation, hostility mediated the effect of parental psychological control on overt and peer relational aggression across gender. Although effect sizes of corresponding indirect effects were larger for females than males, the magnitude of indirect effects did not vary across gender by Wald dif tests (ps > .05). In contrast, guilt and concern for harming others did not mediate effects of parental psychological control on aggression types. Final direct c′ effects of parental psychological control on overt aggression across gender and on female relational aggression were nonsignificant, thus suggesting full mediation. However, the final direct effect of parental psychological control on male relational aggression was significant, thus indicating partial mediation.

Social–cognitive mediation of adolescent conduct problems on emerging adult aggression across gender

Social–cognitive mediation of adolescent conduct problem effects on overt and relational aggression was examined in the fourth and fifth SEMs (see Figures 2 and 3 and Table 4). In accord with hypotheses, effects of adolescent conduct problems on female overt and peer relational aggression were mediated by hostility and low levels of guilt and concern for harming others. In contrast, neither hostility nor guilt mediated the effects of adolescent conduct problems on male overt and peer relational aggression. Wald dif tests underscored gender differences; the direct effect of adolescent conduct problems on hostility, and the indirect effects of conduct problems on overt and peer relational aggression via hostility were significantly greater for females than males (Wald dif tests, ps < .05). Results also affirmed that maternal, paternal, and combined parental psychological control contributed to social–cognitive mediation of adolescent conduct problem effects on female, but not male, overt and peer relational aggression. Specifically, the effects of conduct problems on female overt and peer relational aggression were mediated by paths from maternal psychological control to hostility and to low guilt. Additionally, conduct problem effects on female overt and peer relational aggression were mediated by paths from paternal psychological control to hostility. In the fifth model, mediation paths between conduct problems, parental psychological control, hostility and both forms of aggression were significant for females. Further, the indirect effect on peer relational aggression was significantly greater for females than males (Wald dif test, p < .05). Also, a parental psychological control to low guilt path mediated the effect of conduct problems on female overt aggression. Finally, with all mediators in the model, associations between adolescent conduct problems and maternal, paternal, and parental psychological control were reduced to nonsignificance for females, thus supporting full mediation.

Table 4.

By-gender mediation of adolescent problems on aggression by parental psychological control, hostility, and guilt and concern for harming others.

	Females						Males
	c Effect	c ′ Effect	Indirect effect	95% CI		Effect size	c Effect	c ′ Effect	Indirect effect	95% CI		Effect size
	b	b	b	LL	UL		b	b	b	LL	UL	a × b/SD_Y
Model 4:
CP > hostility > overt aggression	.39*	.09	.15*	.073	.284	.19	.16*	.09	−.00	−.042	.035	.00
CP > guilt > overt aggression	.39*	.09	.06*	.012	.222	.07	.16*	.09	.03	−.008	.117	.04
CP > hostility > relational aggression	.21*	−.12	.19*	.081	.367	.24	.09*	.05	−.00	−.041	.037	.00
CP > guilt > relational aggression	.21*	−.12	.06*	.015	.184	.08	.09*	.05	.02	−.003	.057	.03
CP > maternal psychological control > hostility > overt aggression	.39*	.09	.03*	.010	.087	.04	.16*	.09	.02	.000	.063	.02
CP > maternal psychological control > guilt > overt aggression	.39*	.09	.01*	.001	.060	.01	.16*	.09	.00	−.011	.029	.00
CP > maternal psychological control > hostility > relational aggression	.21*	−.12	.04*	.013	.111	.05	.09*	.05	.02	.000	.068	.03
CP > maternal psychological control > guilt > relational aggression	.21*	−.12	.01*	.001	.054	.01	.09*	.05	.00	−.005	.014	.00
CP > paternal psychological control > hostility > overt aggression	.39*	.09	.03*	.003	.081	.04	.16*	.09	.00	−.001	.015	.00
CP > paternal psychological control > guilt > overt aggression	.39*	.09	.00	−.027	.009	.00	.16*	.09	.00	−.003	.027	.00
CP > paternal psychological control > hostility > relational aggression	.21*	−.12	.03*	.003	.113	.04	.09*	.05	.00	−.002	.015	.00
CP > paternal psychological control > guilt > relational aggression	.21*	−.12	.00	−.020	.010	.00	.09*	.05	.00	−.001	.014	.00
Model 5
CP > hostility > overt aggression	.39*	.08	.16*	.084	.297	.20	.16*	.09	.00	−.026	.044	.00
CP > guilt > overt aggression	.39*	.08	.07*	.014	.222	.09	.16*	.09	.03	−.001	.027	.04
CP > hostility > relational aggression	.21*	−.14	.20*	.085	.370	.26	.09*	.04	.00	−.021	.037	.00
CP > guilt > relational aggression	.21*	−.14	.06*	.018	.205	.08	.09*	.04	.02	−.003	.053	.03
CP > parental psychological control > hostility > overt aggression	.39*	.08	.06*	.025	.117	.07	.16*	.09	.01	−.001	.043	.01
CP > parental psychological control > guilt > overt aggression	.39*	.08	.01*	.001	.055	.01	.16*	.09	.01	−.001	.032	.01
CP > parental psychological control > hostility > relational aggression	.21*	−.14	.07*	.027	.143	.09	.09*	.04	.01	−.001	.034	.01
CP > parental psychological control > guilt > relational aggression	.21*	−.14	.01	.000	.035	.01	.09*	.04	.00	.000	.012	.00

Note. Guilt refers to guilt and concern for harming others. Underlined parameters are significantly different at p < .05 by Wald difference tests. CP = adolescent conduct problems; c effect = the effect of the IV on the DV; c ′ = the effect of the IV on the DV with the mediator in the model.

* p < .05.

Discussion

We found that paternal, as opposed to maternal, psychological control was uniquely associated with emerging adult peer relational aggression across gender. Therefore, paternal influence on daughters’ relational aggression potentially persists into emerging adulthood (Casas et al., 2006; Nelson & Crick, 2002; Nelson et al., 2006). Additionally, psychological control of the same gender parent (father) may influence sons’ use of peer relational aggression (Deater-Deckard & Dodge, 1997), thus affirming findings of a recent study of youth (Soensens et al., 2008). Surprisingly, too, paternal psychological control was significantly associated with female overt aggression, whereas maternal psychological control showed only a marginally nonsignificant relation with female overt aggression, thus contradicting hypotheses. Nonetheless, these results further underscore importance of the father–daughter relationship on youth adjustment (Steinberg & Silk, 2002). Potentially, the relative distance and alienation of the father–daughter relationship potentiates greater negative effects of psychological control on aggression, whereas the closeness of the mother–daughter dyad may cushion the impact of psychological control (Rogers et al., 2003). Finally, the lack of a direct relation between maternal/paternal psychological control and male overt aggression is not necessarily surprising with respect to prior child research. Results highlight the potential salience of other parenting variables (i.e., parental physical coercion) in the prediction of male overt aggression (Nelson et al., 2006).

Beyond specific paternal effects, our findings suggest that emerging adult sons and daughters will be more likely to use overt and relational aggression when both parents use parental psychological control. The strength of combined parental psychological control effects supports previous contention that psychological control has an additive or cumulative effect on youth adjustment, potentially through the emerging adult transition (Kuppens et al., 2009a). This finding points to the potential that when both parents use psychological control, they may be more likely to model controlling and relationally aggressive behaviors in their own peer and marital relationships, thus further promoting their adult children’s use of relational aggression (Reed et al., 2008).

Another key contribution of the current study is consideration of gender differences in the simultaneous effect of prior adolescent conduct problems on parenting and emerging adult aggression. In contrast to Albrecht and Galambos’ (2007) findings, we found that prior adolescent conduct problems were associated with daughters’, but not sons’, perceptions of their parents’ psychological control. In turn, increased paternal and combined parental psychological control mediated higher female overt and peer relational aggression. Potentially, parents may be more likely to respond in a psychologically controlling manner to their late adolescent daughters’ conduct problems, given that female conduct problems represent gender nonnormative behaviors (Eagly & Steffen, 1986). Alternatively, disruptive female youth may be likely to evoke parental psychological control in response to their own use of relational aggression, given that female disruptive behavior is commonly characterized by relational aggression (Loeber et al., 2009). Future research that accounts for the prospective effects of both prior adolescent conduct problems and prior parental psychological control on emerging adult aggression will aid in affirming these relations.

Social–cognitive mediation of parent and conduct problem effects on emerging adult aggression

Surprisingly, maternal and combined parental, but not paternal, psychological control was associated with hostility across gender. Although prior research suggests that paternal psychological control predicts boys’ hostile attributions (Nelson & Coyne, 2009), this study’s findings highlight the salience of the mother–child relationship on late adolescent adjustment (Steinberg & Silk, 2002). Potentially, increasing intimacy in late adolescent mother–youth relationships increases the emotional impact of maternal psychological control on hostile tendencies (Rice & Mulkeen, 1995).

Results showed that paternal psychological control was associated with female, but not male, hostility. This gender difference suggests that adult daughters are more sensitive to their fathers’ psychological control than adult sons. In the context of an otherwise distant relationship, females may be more likely to view their fathers’ psychological control as a signal of rejection, rather than discipline, which would engender alienation and hostile attributions as rationale for overt and relational aggression (Nelson et al., 2006; Steinberg & Silk, 2002). Potentially, emerging adult females’ accrued experience of their fathers’ psychological control may increase their hostile attribution tendencies, thus sensitizing them to peer relational victimization and in sum, increasing their propensity for aggression. Supporting this notion, the current study’s model showed that the direct effect of hostility on relational aggression was significantly greater for female than for male emerging adults. Similarly, prior research suggests that hostile attribution biases for relationally provocative events are more likely to be associated with girls’, versus boys’, relational aggression when girls have a history of relational victimization (Crick, 1995; Mathieson et al., 2011).

Maternal psychological control was indirectly related to overt and relational aggression via increased hostility across gender, but it was not significantly associated with aggression in initial models. Given that tests of mediators often have more power than initial total effects, this set of contradictory findings may be a result of small sampling. Alternatively, the fact that the final direct effect of maternal psychological control on relational aggression was negative though the indirect effects were positive suggests potential inconsistent mediation (MacKinnon, 2008). Future research using a larger sample would aid in teasing apart these relations.

Contrary to prior findings, the effect of psychological control on reduced female guilt was limited to a marginally nonsignificant relation within the mother–daughter dyad (Woien et al., 2003). Following from this, maternal psychological control was indirectly associated with increased female overt aggression via reduced guilt. This result suggests that the psychological control strategies of the same gender parent may be important in female emerging adult guilt socialization. Within the more intense and close mother–daughter relationship (Steinberg & Silk, 2002), maternal use of psychological control strategies may serve as a social model of aversive control processes.

A noteworthy finding of our study was the strength of mediation effects in the prediction of female aggression from adolescent conduct problems. All mediation paths linking adolescent conduct problems, hostility/guilt, and overt/relational aggression were significant for females but nonsignificant for males. Moreover, hostility mediation of conduct problem associations with overt and peer relational aggression was fully moderated by gender. This gender difference suggests that the accrual of hostility and suppression of guilt in the context of the social combat typical of adolescent antisocial peer relationships plays a particularly important role in shaping female emerging adult overt and relational aggression (Czar et al., 2010; Murray-Close, 2011; Ostrov & Houston, 2008).

The current study’s mediation model also sheds light on the potential role of female adolescent conduct problems in adverse parenting during the emerging adult period. Results suggest that parents may respond with a greater shift in psychological control in response to their adolescent daughters’ versus their sons’ misbehaviors in effort to curb the behavior. In turn, this paternal response may exacerbate cycles of hostility and aggression, whereas maternal psychological control may be associated with hostility, reduced guilt, and increased aggression. Pronounced gender differences in the current study’s multimediator path further suggests that female emerging adults may be particularly sensitive to their parents’ cumulative use of psychological control view it as a hostile sign of rejection and, in turn, respond with greater levels of aggression (Nelson et al., 2006). However, the concurrent correlational design of this study prohibits confidence in causal interpretation. Further prospective research accounting for effects of prior parental psychological control as well as adolescent conduct problems on the proposed mediation model would provide a more rigorous test of the contribution of adolescent and parent effects on gender differences in emerging adult aggression.

Directions and limits

The current study’s mediation model represents a preliminary test of potential explanations of gender differences in parental psychological control links with overt and relational aggression. Further longitudinal research evaluating prospective links between parental psychological control, proposed mediators, and emerging adult aggression would permit more rigorous tests of proposed mediation paths. Beyond the cross-sectional design of the current study, a few limitations should be noted. First, reliance on self-report measures limited measurement validity by increasing shared method biases. Second, study results were based on a relatively small convenience sample, thus limiting generalizability of our findings. Future research using additional parent reports and a larger sample representative of the community would reduce shared method biases and external validity limits.

Footnotes

Appendix 1

Cross-gender dimensionality and measurement invariance (MI) tests of latent measures.

Construct	Factors	Configural model				Weak invariance (loadings constrained)			Invariant?
		χ2(df)	CFI	RMSEA	Cross-gender Factor Loadings	Δχ2(df)	CFI	RMSEA
Overt aggression	2: Verbal and physical	89.515 (60)	.95	.05	All ≥ .30	7.740 (8)	.96	.05	yes
Maternal psychological control	1	25.153 (25)	1.00	.01	All ≥ .30	6.393 (5)	1.00	.02	yes
Paternal psychological control	1	25.551 (21)	1.00	.04	All ≥ .30	3.651 (5)	1.00	.02	yes
Guilt and concern for harming others	1	23.782 (16)	.99	.05	All ≥ .30	9.684 (6)	0.99	.05	yes
Hostility	1	27.967 (19)	.98	.05	All ≥ .30	9.01 (5)	0.98	.05	yes

Note. EFAs in SEM were used to determine dimensionality based on model fit, eigenvalue, and factor loading criteria (standardized loadings ≥ .30). For all measures except overt aggression, items were treated as ordinal categorical for MI tests and model identification procedures recommended by Millsap and Tein (2004) were used. A robust maximum likelihood estimator was used for overt aggression MI tests, given that overt aggression items used a 7-point scale. Configural models permitted all item factor loadings, residuals, and thresholds to vary across gender. Model fit for initial configural models was evaluated on the basis of the chi-square measure of absolute fit and two practical fit indices: the CFI and RMSEA. Cutoffs of CFI ≥ .95 and RMSEA ≤ .05 suggest good fit in baseline configural models (Cheung & Rensvold, 2002). By-gender measurement invariance of the factor structure in the configural model is indicated by the same pattern of nonzero factor loadings across sex (standardized loadings ≥ .30). Measurement invariance of factor loadings was evaluated by constraining factor loadings across gender and evaluating change in overall model fit. Determination of measurement non-invariance was based on change in a majority of fit indices. Thresholds for non-invariance included (1) significant change in the chi-square between successive nested models, (2) change in the RMSEA indicating worse fit (i.e., a change from ≤ .05 to > .05), and (3) a change in the CFI of ≤ −.01. Information on dimensionality and MI tests of relational aggression items is available in Murray-Close et al, 2010. MI: measurement invariance; CFI: comparative fit index; RMSEA: root mean square error of approximation; SEM: structural equation modeling; EFAs: exploratory factor analyses.

Appendix 2

Guilt and concern for harming other scale items
1. I do not feel remorseful when I do something wrong.*
2. I do not care who I hurt to get what I want.*
3. I am concerned about the feelings of others.
4. I try not to hurt others' feelings.
5. The feelings of others are unimportant to me.*
6. I apologize (“say I am sorry”) to persons I hurt.
7. I feel bad or guilty when I do something wrong.

Note. The items are from the Inventory of Callous–Unemotional Traits (ICU; Essau, Sasagawa, & Frick, 2006).

*Reverse scored.

Authors’ Note

An earlier version of this study was presented at the 2013 Biennial Meeting of the Society for Research in Child Development, Seattle, WA, USA.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Achenbach

T. M.

Howell

C. T.

McConaughy

S. H.

Stanger

(1995). Six-year predictors of problems in a national sample: III. Transitions to young adult syndromes. Journal of the American Academy of Child & Adolescent Psychiatry, 34, 658–669.

Albrecht

A. K.

Galambos

N. L.

(2007). Adolescents’ internalizing and aggressive behaviors and perceptions of parents’ psychological control: A panel study examining direction of effects. Journal of Youth and Adolescence, 36, 673–684. doi:10.1007/s10964-007-9191-5

Allen

J. P.

Hauser

S. T.

Eickholt

Bell

K. L.

O′Connor

T. G

. (1994). Autonomy and relatedness in family interactions as predictors of expression of negative of adolescent affect. Journal of Research on Adolescence, 4, 535–552. doi:10.1207/s15327795jra0404_6

Arnett

J. J.

(2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469–480. doi:10.1037/0003-066X.55.5.469

Aquilino

W. S.

(2006). Family relationships and support systems in emerging adulthood. In Arnett

J. J.

Tanner

J. L.

(Eds.), Emerging adults in America: Coming of age in the 21st century (pp. 193–217). Washington, DC: American Psychological Association.

Bagby

R. M.

Farvolden

(2004). The personality diagnostic questionnaire-4 (PDQ-4). In Hilsenroth

M. J.

Segal

D. L.

(Eds.), Comprehensive handbook of psychological assessment (Vol. 2, pp. 122–133): Hoboken, NJ: John Wiley & Sons Inc.

Bandura

(1979). The social learning perspective: Mechanisms of aggression. In Toch

(Ed.), Psychology of crime and criminal justice (pp. 198–236). Prospect Heights, IL: Waveland Press.

Barber

B. K.

(1996). Parental psychological control: Revisiting a neglected construct. Child Development, 67, 3296–3319. doi:10.2307/1131780

Barber

B. K.

Harmon

E. L.

(2002). Violating the self: Parental psychological control of children and adolescents. In Barber

B. K.

(Ed.), Intrusive parenting: How psychological control affects children and adolescents (pp. 15–22). Washington, DC: American Psychological Association. doi:10.1037/10422-006

10.

Brody

G. H.

Shaffer

D. R.

(1982). Contribution of parents and peers to children’s moral socialization. Developmental Review, 2, 31–75. doi:10.1016/0273-2297(82)90003-X

11.

Buss

A. H.

Perry

M. P.

(1992). The aggression questionnaire. Journal of Personality and Social Psychology, 63, 452–459. doi:10.1037/0022-3514.63.3.452

12.

Casas

J. F.

Weigel

S. M.

Crick

N. R.

Ostrov

J. M.

Woods

K. E.

Jansen Yeh

E. A.

Huddleston-Casas

(2006). Early parenting and children’s relational and physical aggression in the preschool and home contexts. Journal of Applied Developmental Psychology, 27, 209–227. doi:10.1016/j.appdev.2006.02.003

13.

Cheung

G. W.

Rensvold

R. B

. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9, 233–255. doi:10.1207/S15328007SEM0902_5

14.

Crick

N. R.

(1995). Relational aggression: The role of intent attributions, feelings of distress, and provocation type. Development and Psychopathology, 7, 313–322. doi:10.1017/S0954579400006520

15.

Czar

K. A.

Dahlen

E. R.

Bullock

E. E.

Nicholson

B. C.

(2010). Psychopathic personality traits in relational aggression among young adults. Aggressive Behavior, 37, 207–214. doi:10.1002/ab.20381

16.

Deater-Deckard

Dodge

K. A

. (1997). Externalizing behavior problems and discipline revisited: Nonlinear effects and variation by culture, context, and gender. Psychological Inquiry, 8, 161–175. doi:10.1207/s15327965pli0803_1

17.

Eagly

A. H.

Steffen

V. J.

(1986). Gender and aggressive behavior: A meta-analytic review of the social psychological literature. Psychological Bulletin, 100, 309–330. doi:10.1037/0033-2909.100.3.309

18.

Essau

C. A.

Sasagawa

Frick

P. J.

(2006). Callous-unemotional traits in a community sample of adolescents. Assessment, 13, 454–469. doi:10.1177/1073191106287354

19.

Godleski

S. A.

Ostrov

J. M.

(2010). Relational aggression and hostile attribution biases: Testing multiple statistical methods and models. Journal of Abnormal Child Psychology 38, 447–458. doi:10.1007/s10802-010-9391-4

20.

Gomez

DeMello

Tallent

(2001). Perceived maternal control and support: Effects on hostile biased social information processing and aggression among clinic-referred children with high aggression. Journal of Child Psychology and Psychiatry, 42, 513–522. doi:1 0.1111/1469-7610.00745

21.

Hart

C. H.

Nelson

D. A.

Robinson

C. C.

Olsen

S. F.

McNeilly-Choque

(1998). Overt and relational aggression in Russian nursery-school-age children: Parenting style and marital linkages. Developmental Psychology, 34, 687–697. doi:10.1037/0012-1649.34.4.687

22.

Hoeve

Dubas

J. S.

Eichelsheim

V. I.

Van der Laan

P. H.

Smeenk

Gerris

J. R.

(2009). The relationship between parenting and delinquency: A meta-analysis. Journal of Abnormal Child Psychology, 37, 749–775. doi:10.1007/s10802-009-9310-8

23.

Hoffman

M. L.

(1977). Moral internalization: Current theory and research. In: Berkowitz

(Ed.), Advances in experimental social psychology (Vol. 10, pp. 85–133). New York, NY: Academic Press.

24.

Kazdin

A. E.

Rodgers

Colbus

Siegel

(1987). Children's hostility inventory: Measurement of aggression and hostility in psychiatric inpatient children. Journal of Clinical Child Psychology, 16, 320–328.

25.

Kishton

J. M.

Widaman

K. F.

(1994). Unidimensional versus domain representative parceling of questionnaire items: An empirical example. Educational and Psychological Measurement, 54, 757–765. doi:10.1177/0013164494054003022

26.

Kline

R. B.

(2005). Principles and practice of structural equation modeling (2nd ed.) New York, NY: Guilford Press.

27.

Kochanska

Aksan

(2006). Children’s conscience and self-regulation. Journal of Personality, 74, 1587–1618. doi:10.1111/j.1467-6494.200600421.x

28.

Kochanska

Barry

R. A.

Jimenez

N. B.

Hollatz

A. L.

Woodard

(2009). Guilt and effortful control: Two mechanisms that prevent disruptive developmental trajectories. Journal of Personality and Social Psychology, 97, 322–333. doi:10.1037/a0015471

29.

Kuppens

Grietens

Onghena

Michiels

(2009a). Associations between parental control and children’s overt and relational aggression. British Journal of Developmental Psychology, 27, 607–623. doi:10.1348/026151008X345591

30.

Kuppens

Grietens

Onghena

Michiels

(2009b). Relations between parental psychological control and childhood relational aggression: Reciprocal in nature? Journal of Clinical Child and Adolescent Psychology, 38, 117–131. doi:10.1080/15374410802575354

31.

Kuppens

Laurent

Heyvaert

Onghena

(2012). Associations between parental psychological control and relational aggression in children and adolescents: A multilevel and sequential meta-analysis. Developmental Psychology, 49, 1697–1712. doi:10.1037/a0030740

32.

Linder

J. R.

Crick

N. R.

Collins

W. A.

(2002). Relational aggression and victimization in young adults’ romantic relationships: Associations with perceptions of parent, peer, and romantic relationship quality. Social Development, 11, 69–86. doi:10.1111/1467-9507.00187

33.

Little

Henrich

Jones

Hawley

(2003). Disentangling the” whys” from the” whats” of aggressive behaviour. International Journal of Behavioral Development, 27, 122–133.

34.

Little

T. D.

Cunningham

W. A.

Shahar

(2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9, 153–173. doi:10.1.1.118.6289

35.

Loeber

(1990). Development and risk factors of juvenile antisocial behavior and delinquency. Clinical Psychology Review, 10, 1–41.

36.

Loeber

Pardini

D. A.

Hipwell

Stouthamer-Loeber

Keenan

Sembower

M. A

. (2009). Are there stable factors in preadolescent girls’ externalizing behaviors? Journal of Abnormal Child Psychology, 37, 777–791. doi:https://dx-doi-org.web.bisu.edu.cn/10.1007/s10802-009-9320-6

37.

MacKinnon

D.P.

(2008). Introduction to statistical mediation analysis. New York, NY: Lawrence Erlbaum.

38.

Mathieson

L. C.

Murray-Close

Crick

N. C.

Woods

K. E.

Zimmer-Gembeck

Geiger

T. C.

Morales

J. R.

(2011). Hostile intent attributions and relational aggression: The moderating roles of emotional sensitivity, gender, and victimization. Journal of Abnormal Child Psychology, 39, 977–987. doi:10.1007/s10802-011-9515-5

39.

Millsap

R. E.

Tein

J. Y

. (2004). Assessing factorial invariance in ordered-categorical measures. Multivariate Behavioral Research, 39, 479–515. doi:10.1207/S15327906MBR3903_4

40.

Murray-Close

(2011). Autonomic reactivity and romantic relational aggression among female emerging adults: Moderating roles of social and cognitive risk. International Journal of Psychophysiology, 80, 28–35. doi:10.1016/j.ijpsycho.2011.01.007

41.

Murray-Close

Ostrov

J. M.

Nelson

D. A.

Crick

N. R.

Coccaro

E. F.

(2010). Proactive, reactive, and romantic relational aggression in adulthood: Measurement, predictive validity, gender differences, and association with intermittent explosive disorder. Journal of Psychiatric Research, 44, 393–404. doi:10.1016/j.jpsychires.2009.09.005

42.

Muthén

(2008). MPLUS user’s guide. Los Angeles, CA: Authors.

43.

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 Psychology, 37, 223–237. doi:10.1007/s10802-008-9271-3

44.

Nelson

D. A.

Crick

N. R.

(2002). Parental psychological control: Implications for childhood physical and relational aggression. In Barber

B. K.

(Ed.), Intrusive parenting: How psychological control affects children and adolescents (pp. 161–189). Washington, DC: American Psychological Association. doi:10.1037/10422-006

45.

Nelson

D. A.

Hart

C. H.

Yang

Olsen

J. A.

Jin

(2006). Aversive parenting in China: Associations with child physical and relational aggression. Child Development, 77, 554–572. doi:10.1111/j.1467- 8624.2006.00890.x

46.

Nelson

Padilla-Walker

Carroll

Madsen

Barry

Badger

(2007). ‘If you want me to treat you like an adult, start acting like one!’ Comparing the criteria that emerging adults and their parents have for adulthood. Journal of Family Psychology, 21, 665–674.

47.

Nelson

L. J.

Padilla-Walker

L. M.

Christensen

K. J.

Evans

C. A.

Carroll

J. S.

(2011). Parenting in emerging adulthood: An examination of parenting clusters and correlates. Journal of Youth and Adolescence, 40, 730–743. doi:10.1007/s10964-010-9584-8

48.

Onishi

Kawabata

Kurokawa

Yoshida

(2012). A mediating model of relational aggression, narcissistic orientations, guilt feelings, and perceived classroom norms. School Psychology International, 33, 367–390. doi:10.1177/0143034311421433

49.

Ostrov

J. M.

Houston

(2008). The utility of forms and functions of aggression in emerging adulthood: Association with personality disorder symptomatology. Journal of Youth and Adolescence, 37, 1147–1158. doi:10.1007/s10964-008-9289-4

50.

Pardini

D. A.

Loeber

(2008). Interpersonal callousness trajectories across adolescence early social influences and adult outcomes. Criminal Justice and Behavior, 35, 173–196.

51.

Petzel

T. P.

Michaels

E. J.

(1973). Perception of violence as a function of levels of hostility. Journal of Consulting and Clinical Psychology, 41, 35–36. doi:10.1037/h0035585

52.

Reed

T. J.

Goldstein

S. E.

Morris

A. S.

Keyes

A. W.

(2008). Relational aggression in mothers and children: Links with psychological control and child adjustment. Sex Roles, 59, 39–48. doi:10.1007/s11199- 008- 9423-5

53.

Rice

K. G.

Mulkeen

(1995). Relationships with parents and peers: A longitudinal study of adolescent intimacy. Journal of Adolescent Research, 10, 338–357. doi:10.1177/0743554895103003

54.

Rogers

K. N.

Buchanan

C. M.

Winchell

M. E.

(2003). Psychological control during early adolescence: Links to adjustment in differing parent/adolescent dyads. The Journal of Early Adolescence, 23, 349–383. doi:10.1177/0272431603258344

55.

Soensens

Vansteenkiste

Goossens

Duriez

Niemiec

C. P.

(2008). The intervening role of relational aggression between psychological control and friendship quality. Social Development, 17, 661–681. doi:10.1111/j.1467-9507.2007.00454.x

56.

Steinberg

Silk

J. S.

(2002). Parenting adolescents. In Bornstein

M. H.

(Ed.), Handbook of parenting: Vol. 1: Children and parenting (2nd ed., pp. 103–133). Mahwah, NJ: Lawrence Erlbaum Associates.

57.

Verona

Patrick

C. J.

Lang

A. R.

(2002). A direct assessment of the role of state and trait negative emotion in aggressive behavior. Journal of Abnormal Psychology, 111, 249–258. doi:10.1037/0021-843X.111.2.249

58.

Woien

S. L.

Emst

H. A.

Patock-Peckham

J. A.

Nagoshi

C. T.

(2003). Validation of the TOSCA to measure shame and guilt. Personality and Individual Differences, 35, 313–326. doi:10.1016/S0191- 8869(02)00191-5