Associations between adolescent aggressive behavior problems with-9-71

1

Chapter I: Introduction

During adolescence, children change dramatically in psychology and physiology. For

example, they are more likely to become rebellious (McDermott, 1988). The relationship

between parents and children during adolescence has increasing conflict and decreasing

closeness (Baer, 2002; De Goede et al., 2009; McGue et al., 2005). Moreover, a growing sense

of independence and autonomy occurs during adolescence (Steinberg, 2013). Adolescents are

more than one in three adolescents in high school report they have been in a physical fight in the

past year (Centers for Disease Control and Prevention, 2012). In addition, adolescent aggressive

behavior typically has stronger negative consequences than those of children’s aggressive

behavior. For example, adolescent aggression can be associated with violent crimes such as

robbery, rape and homicide (Loeber & Hay, 1997; Yonas et al., 2005), and during adolescence

the rate of children arrested for violence dramatically and stably increases (Loeber & Stallings,

2011). According to the dual taxonomy model, the evocative, reactive, and proactive interactions

between children and their environment can produce a pathological personality for some

children, which manifests in the persistence and stability of antisocial behavior (for a review, see

Moffitt, 1993). Based on this model, aggressive behavior problems for a small group of

2

justifies the attention of researchers.

Externalizing behavior problems is often defined as consisting of two facets: aggressive

behaviors (e.g., physically attacking people, teasing a lot) and delinquent behaviors (e.g.,

breaking the evening curfew; Fergusson et al., 1994; Lahey & Loeber, 1994). Parents’ behaviors

parental knowledge, most of these studies investigated delinquency or externalizing behavior

problems that include both delinquency and aggression. Only a few studies examined the nature

of the relationship between adolescent aggression specifically with parental monitoring and

parental knowledge. For example, a study by Racz and McMahon (2011) reviewed 47 studies on

the relationship between parental knowledge and monitoring and child and adolescent conduct

problems from 2000 to 2010. Among these 47 studies, only one study exclusively examined

adolescent aggressive behaviors. Although aggressive behaviors and delinquent behaviors are

highly associated with each other, research suggested that these conduct problems indeed

represent two distinct entities (Achenbach et al., 1995; Barnow et al., 2005; Bongers et al., 2004;

Stanger et al., 1997). For instance, aggressive behaviors on average decrease over time, while

3

developmental research has been recommended to distinguish aggressive behaviors and

al., 2004; Tremblay, 2000). Therefore, research is needed that specifically focuses on adolescent

extend the literature in this field, the current dissertation focuses on adolescent aggressive

dissertation further investigates three possible moderators on these associations.

Statement of the Problem

Distinguishing Parental Monitoring and Parental Knowledge

Parental monitoring involves a set of parental behaviors intended to obtain information

about the child’s whereabouts, activities, and adaptations (Dishion & McMahon, 1998). Parental

knowledge is what parents know about their child’s whereabouts, activities, and others (Crouter

& Head, 2002). Research has found that children perceiving more parental monitoring and

parental knowledge during adolescence had fewer aggressive behavior problems (Buehler et al.,

2006; Simons-Morton et al., 2004). For example, Merrin et al. (2019) discovered that adolescents

who reported higher average rates of parental monitoring had lower rates of deviant and violent

behaviors during middle and high school. Brody (2003) also showed that when mothers knew

more about their child (11 to 13-year-old African American adolescents), their child had fewer

externalizing behavior problems as indicated by teachers’ report.

However, the measures of parental monitoring and parental knowledge have been mixed in

4

Stattin, 2000; Stattin & Kerr, 2000; Racz & McMahon, 2011). Up to today, this phenomenon still

monitoring assesses the degree to which parents actively make efforts to obtain their child’s

child and child’s friends for information (i.e., parental solicitation) as well as establishing the

research measures parental monitoring by assessing parental knowledge of their child’s

al. (1994) to assess parental monitoring. A closer scrutiny on the content of these five items, such

being assessed is not parental monitoring but parental knowledge. Other studies used similar

your friends are? Where you go at night? Where you are most afternoons after school?” (Fletcher

Stattin & Kerr, 2000), rather than about how parents actively make efforts to know what their

Other researchers have measured parental monitoring by assessing 1) parental control, 2)

instance, Merrin et al. (2019) combined parental solicitation, parental control, and child active

5

meta-analysis study also used the term parental monitoring in a way that combined parental

in a way that combines parental solicitation, parental control, and child active disclosure hides

information. While child active disclosure is a significant source of parental knowledge (Kerr et

The term parental monitoring is used in the following sections to specifically represent parents

parental control.

Different Effects of Parental Monitoring and Parental Knowledge on Adolescents Behavior

Problems

Although there are some overlaps between parental monitoring and parental knowledge

(e.g., the more parents monitor their child, the more they are likely to know their child, Hirschi,

1969; Liu et al., 2020), parental monitoring and parental knowledge should be considered as two

different dimensions of parenting (Kerr & Stattin, 2000; Stattin & Kerr, 2000). Several empirical

studies showed the different functions of parental monitoring and parental knowledge on

adolescent behavior problems when distinguishing these measures (e.g., Laird et al., 2010;

Lippold et al., 2014; Kerr et al., 2010). A study by Kerr et al. (2010), for instance, examined the

associations between adolescent delinquent behaviors with parental monitoring and parental

knowledge from a longitudinal perspective. Parental knowledge in this study was measured by

6

parental solicitation (i.e., only a focus on parent-driven behaviors). Interestingly, they found that

parental monitoring does not predict the changes of adolescent delinquent behavior over time.

parental monitoring do not predict fewer adolescent antisocial behavior problems using the

leads to the increases in parental knowledge (Lippold et al., 2014).

make the findings on adolescent aggressive behavior problems unclear. Indeed when

antisocial behavior problems (Kerr et al., 2010; Laird et al., 2010; Lippold et al., 2014; Wang et

distinguished the measures between parental monitoring and parental knowledge, let alone

behavior problems as two different dimensions of parenting (for more information, see the

associations between adolescent aggressive behavior problems and parental monitoring as well

monitoring (i.e., parent-driven behaviors) and parental knowledge (see the hypothesized

7

desirable implications of such a study are discussed in the following section.

The Hypothesized Structural Model between Parental Monitoring and Parental Knowledge

Note. The effects of mothering and fathering would be tested separately in this model. Intervention status would be

controlled as a covariate in all model.

8

Different Roles of Mothers and Fathers in Parenting

Several studies have indicated that maternal parenting and paternal parenting have unique

contributions on adolescent development (Hoeve et al., 2011; Jeynes, 2016; Murray et al., 2014;

Van Heel et al., 2019). For example, Jeynes (2016) conducted a meta-analysis that included 34

studies and demonstrated that fathers play a unique role in parenting that is different from

mothers’ roles in children’s development controlling for psychological, sociological, and

Viewing the influences of mothers and fathers on children as the same and investigating mother

and father as a whole – “parent” – may lose some important information (Jeynes, 2016). It is

worthwhile to examine the effects of mothers and fathers separately.

However, except one study on delinquency (Keijsers et al., 2010) that found the majority

associations between parental knowledge and adolescent behaviors occurred to mothers, to my

best knowledge, no prior research examined the associations between adolescent aggression with

parental monitoring and parental knowledge considering the potential different roles of mothers

and fathers. Therefore, the current study adds to literature by examining parental monitoring and

parental knowledge with mothers’ and fathers’ perspectives separately (i.e., maternal monitoring,

paternal monitoring, maternal knowledge, and paternal knowledge) in the association with

9

Three Potential Moderators on the Associations between Parental Monitoring and Parental

Knowledge and Adolescent Aggressive Behaviors

Parent-Adolescent Relationship Quality. An extensive body of studies have illustrated the

effects of parent-child relationship quality on adolescent behavior problems (for more details, see

the Literature Review section). Importantly, several studies demonstrated the moderation effect

of parent-adolescent relationship quality on the associations between adolescent delinquency

relationship quality, the associations between adolescent delinquency and parental knowledge as

al., 2009; Lippold et al., 2014). However, no existing studies show the moderating effect of

problems with parental monitoring and parental knowledge.

The integrative model of parenting style proposed by Darling and Steinberg (1993) indicates

that parenting style as an overarching characteristic (e.g., parent-child interactions, including

emotion, relationship) impacts the associations between specific parenting practices and youth

outcomes. In accordance with the integrative model of parenting style (Darling & Steinberg,

1993), several empirical studies have demonstrated the moderation effect of parenting style on

the association between adolescent risk behaviors (e.g., sexual, substance use, delinquency) and

specific parenting practices (e.g., Abar et al., 2014; Mounts, 2002; Zhengyan et al., 2007). Thus,

it is possible that parent-adolescent relationship quality may influence the associations of

parental monitoring and parental knowledge with adolescent aggressive behavior problems, and

the current study provides insight on this issue.

16

2004). Therefore, parents generally have more knowledge about their girls than boys (Pettit et

daughter relationship (e.g., more closeness and less confliction), it may be easier for their parents

boys have more aggressive behavior problems than girls and the aggressive behavior problems in

of more aggressive behavior problems, less parental knowledge and less parental monitoring in

girls (Rothbaum & Weisz, 1994). So the associations between adolescent aggressive behavior

This study can provide some important implications for researchers and parents who care

about adolescent aggressive behavior problems. For example, if the effects of parental

monitoring and parental knowledge on adolescent aggressive behavior problems are

demonstrated to be different, such as parental knowledge rather than parental monitoring (i.e.,

parent-driven behaviors) associated with adolescent aggression, it will suggest that parents

should be encouraged to get more knowledge about their adolescents. The existing empirical

research has been consistently demonstrated that child active disclosure is the main source of

parental knowledge (e.g., Kerr et al., 2010; Stattin & Kerr, 2000). Thus parents are

17

active efforts to monitor their adolescents may be ineffective in terms of reducing their

Moreover, if reciprocal relationships between adolescent aggressive behavior with parental

not only focus on the effects of parental monitoring and parental knowledge on adolescent

educators should not only consider the negative effect of aggressive behavior problems on

proposing intervention programs, researchers should build in ways that support those parents

bidirectional relationships between parental monitoring and parental knowledge with adolescent

aggression differ from the effects of paternal monitoring and knowledge (e.g., the degree or even

information to mothers and fathers. More importantly, this study could provide some robust

most of the research in this field does not explore the function of parental monitoring on

18

research. Although this issue has been discussed by Stattin, Kerr and colleagues since 2000 (Kerr

2016; see Racz & McMahon, 2011), this problem still exists in the research on adolescent

literature on adolescent aggression. To review the literature in this field, five broad aspects are

problems, 3) the effect of parent-adolescent relationship quality on adolescent aggressive

aggressive behavior problems, and 5) youth gender differences on adolescent aggressive

When focusing on parental monitoring that refers to parent-driven behaviors, knowledge on

the associations between parental monitoring and adolescent behavior problems is unclear. Some

studies have demonstrated that adolescents who perceived more parental monitoring reported

fewer behavior problems (e.g., Stavrinides et al. 2015; Wang et al., 2011; Wertz et al., 2016;

Willoughby & Hamza, 2011). Whereas several studies did not find any association between

parental monitoring and adolescent behavior problems (e.g., Keijsers, 2016; Keijsers et al., 2010;

Gault-Sherman, 2012). Additionally, others indeed found that more parental monitoring was

associated with more adolescent behavior problems (e.g., Abar et al., 2014; Kerr et al., 2010;

Kiesner et al., 2009; Kapetanovic et al., 2020; Lippold et al., 2014). When particularly focusing

19

behavior problems (for a review, see Racz & McMahon, 2011). Most of the existing studies

risk behavior, and antisocial behavior problems. To review this domain, two examples on

reference for research on aggression.

wave longitudinal data (baseline modal age = 14). They used the Dutch translation of scales

adolescent or friends about unsupervised time), parental control (i.e., the way in which parents

the negative bidirectional relationship between adolescent disclosure and adolescent

on adolescent delinquency was not found (Keijsers et al., 2010). Furthermore, one contribution

informant sources. In other words, they examined parental solicitation, parental control, and

adolescent self-report. The result was consistently found in father, mother, and adolescent reports

they found that maternal solicitation predicted adolescent disclosure. This suggests that although

20

behaviors.

parental monitoring was operationalized by parental active monitoring efforts to solicit

youth perceptions of maternal knowledge about how often their mother knows where they are

him/her to do. Lippold et al. (2014) separately measured parental monitoring and parental

and delinquent behavior problems. This research illustrates that parental knowledge predicts

They suggested that the beneficial effect of parental monitoring on adolescent behavior problems

partly consistent with Keijsers and her colleagues’ work (2010) that implies the importance of

In sum, the picture on the association between parental monitoring and adolescent behavior

knowledge, examined the association of parental monitoring with adolescent aggression, let

al., 2014; Keijsers et al., 2010; Lippold et al., 2014), the current study adds to literature by

21

aggressive behavior should be considered as a distinct entity from delinquent behavior

should investigate aggression and delinquency separately to reveal meaningful, developmental

longitudinally examine whether the association between parental monitoring and adolescent

perceived more parental knowledge reported themselves having fewer norm-breaking behaviors.

Subsequently, Kerr et al. (2010) conducted a two-year longitudinal study among 938 seventh and

eighth graders and their parents. This study also showed that parental knowledge predicted

decreases in adolescent delinquent behavior problems over time, and that adolescent delinquency

predicted decreases in parental knowledge over time. Research involving adolescent aggressive

behavior problems showed similar results to Kerr et al. (2010) (Brody, 2003; Buehler et al.,

2006; Gaertner et al., 2010; Merrin et al., 2019; Reitz et al., 2007; Simons-Morton et al., 2004).

Reitz et al. (2007), for instance, examined the effects of parental knowledge on adolescent

aggressive and delinquent behavior problems by using latent growth curve modeling in the

longitudinal data and illustrated that parental knowledge negatively predicted adolescent

22

Adolescents perceiving more parental knowledge reported fewer aggressive and delinquent

association between “parental monitoring” and early adolescent aggressive behavior. However,

children in school and free time. The measurement of “parental monitoring” in the study by

subsequent aggressive behavior problems (Simons-Morton et al., 2004). Last example by Merrin

behavior and the developmental changes of adolescent aggression from Grade 5 to Grade 11.

attendance, peer relationships, alcohol or drug use, and weapon possession) combined. They

deviant and violent behaviors during middle and high school. Moreover, when controlling for the

perceived more parental behavior than their typical level reported fewer deviant and violent

level, their child may engage in fewer deviant and violent behaviors.

aggression, the study by Stavrinides et al. (2015) investigated this focusing on bullying and

23

parental solicitation, parental control, and child disclosure. Three-hundred and forty-eight early

behavior (Olweus, 1993). Results related to bullying are presented here. They found that parental

parental solicitation on bullying is that parental solicitation (e.g., explicating asking their

et al. (2015) uncovered that parental control and child disclosure predicted more bullying and

In summary, the negative association between parental knowledge and adolescent aggressive

association between parental knowledge and bullying, the findings in their study are inconsistent

between parental knowledge and adolescent delinquency (e.g., Gaertner et al., 2010; Keijsers et

Stavrinides et al. (2015), no previous studies longitudinally test the possible reciprocal

while simultaneously accounting for parent-driven monitoring behaviors. The current study

negative relationship between parental knowledge and adolescent aggressive behavior problems

24

Effect of Parent-Adolescent Relationship Quality on Adolescent Aggressive Behavior

Problems

The quality of parent-adolescent relationship is associated with adolescent externalizing and

risk behavior problems (Pinquart, 2017). For example, parent-child conflict predicts increases in

adolescent externalizing behavior problems (Klahr et al., 2011), while parent-child warm

relationship quality is associated with fewer adolescent externalizing behavior problems

(Pinquart, 2017). In particular, Fosco et al. (2014) explored the predictive function of parent-

adolescent hostile relationships on adolescent aggressive behavior. This research demonstrated

that parent-adolescent hostile relationships in seventh grade positively predicted adolescent

aggressive behaviors in ninth grade (Fosco et al., 2014). Moreover, the meta-analysis study by

Pinquart (2017) indicated that parent-child relationship was associated with children’s

externalizing behavior problems. Specifically, warm parent-child relationships had a small

body of research has demonstrated the indirect effects of the quality of parent-child relationship

on adolescent behavior problems via parental monitoring and parental knowledge. For example,

Ary et al. (1999) illustrated that families with high levels of conflict and low positive family

25

increases of adolescent problem behaviors, such as antisocial behavior, high risk sex, and

included every family member (e.g., inter-parent relationship, inter-sibling relationship, and

Kapetanovic et al. (2019) illustrated the indirect effect of adolescent connectedness to parent on

relationship between parents and adolescents was associated with more parental knowledge at

point.

Moderation Effect of Parent-Adolescent Relationship Quality on the Associations between

Adolescent Aggressive Behavior Problems and Parenting

No previous evidence shows the moderation effects of parent-adolescent relationship quality

on the associations between adolescent aggressive behavior and parental monitoring or parental

knowledge. Few studies investigated the moderation effect of parent-child relationship quality on

the associations between monitoring/knowledge and youth behavior problems (e.g., delinquency,

antisociality) (e.g., Abar et al., 2014; Lippold et al., 2014; Keijsers et al., 2009). Keijsers et al.

(2009), for instance, found when having a high level of supportive relationship between parents

and adolescents, the association between adolescent delinquent behavior and monitoring

(defined in this study as solicitation) became stronger when having a low level of supportive

relationship. Keijsers et al. (2009) discovered that a greater decrease in parental control was

32

study adds to literature by examining mothers and fathers separately.

Participants

Data for the current study come from a subset of the PROSPER project (Promoting School-

Community-University Partnerships to Enhance Resilience; Spoth et al. 2004). The PROSPER

delivering evidence-based interventions (Spoth et al., 2017) and includes 28 rural communities

sequence of evidence based preventive interventions was implemented in PROSPER intervention

family-focused and school-based interventions. All teams chose the Strengthening Families

Program: For Parents and Youth 10-14 (SFP: 10-14) as the family-focused intervention. SFP

targets adolescent skills training (e.g., teaching effective communication and problem solving

between parents and children) to reduce adolescent aggressive behavior problems (Spoth et al.,

2000). For the school-based program six teams chose the All Stars program (McNeal et al.,

2004), and four each chose Life Skills Training (Botvin & Kantor, 2000) and Project Alert

(Ellickson et al., 2003). All three programs target decision making, social norms regrading

substance misuse and personal goal setting. A random sample of 2,267 families of PROSPER

youth were invited to participate in an in-home data collection, and 977 families at wave 1

completed the intensive in-home interviews. Wave 1 was finished in the fall semester of 6th

grade. Wave 2 was finished in the spring semester of 6th grade and Wave 3 to Wave 5 were

33

Fifty-two percent of current analytic sample at wave 1 were female; 87% of them were

missing data at wave 5 did not differ on any variables at wave 1 used in this study (see Measures

University at Albany, SUNY, on March 6, 2020.

Measures

Parental Monitoring

Based on the research by Kerr et al. (2010), the operational definition of parental monitoring

in the current study is that parents actively make efforts to gain the knowledge of their child’s

whereabouts and activities. Parental monitoring was measured by assessing the degree of

parental active monitoring efforts to solicit information from them and set up monitoring

relatedrules about their activities (Lippold et al., 2014). In total, five Likert-scaled items

included (1 = Almost always true to 5 = Almost always false), such as “I expect my child to let

me know in advance who will be driving for my child and his/her friends when they go out (to

parties, motives, etc.).” Items were recoded so that higher scores mean higher parental

monitoring. The same five items were reported by mothers and fathers, separately. The average

scores of these five items were used to create maternal monitoring and paternal monitoring

variables. Alpha reliabilities of parental monitoring ranged from .63 to .76 across five waves.

34

Parental Knowledge

According to research by Crouter and Head (2002), parental knowledge was

operationalized as how much information parents know about their child’s whereabouts and

activities. In total, five Likert-scaled items were included (1 = Almost always to 5 = Almost

never). These five items were used in prior studies to measure parental knowledge (e.g., Lippold

et al., 2014; Lippold et al., 2015), indicating good validity. An example item is “How often do

Adolescent Aggressive Behavior

Aggressive behavior among adolescents was measured using the Child Behavior Check

List (CBCL), which is a widely used parent-reported measure for assessing children’s

internalizing and externalizing behavior problems (Achenbach, 1991; Achenbach & Rescorla,

2001). The CBCL externalizing behavior problems subscale includes a general measure of

aggressive behavior problems (Achenbach, 1991; Achenbach & Rescorla, 2001). In this study,

the 17-item aggressive behavior subscale was used to measure adolescent aggressive behavior

problems. Mothers and fathers were asked to rate how true each of the 17 items was for their

adolescents within the past 6 months, such as physically attacking people, destroying things, and

teasing a lot. Items were scored on a 0-2 scale (0 = Never true, 1 = Sometimes true, 2 = Always

true) and were averaged to create an aggressive behavior problems measure. Alpha reliabilities

35

Parent-Adolescent Affective Quality

Always to 7 = Never). Items were designed to measure adolescents’ perceptions of parental

hostility and warmth toward them (Conger et al., 1994; Lippold et al., 2014), such as how often

adolescents think their mothers/fathers get angry at them, and how often adolescents think their

mothers/fathers act loving and affectionate toward them. Items were recoded so that higher

al., 2014). To test Hypothesis 2 - whether the associations of parental monitoring and parental

knowledge with adolescent aggression differ between high and low positive affective quality

circumstances - a median split approach was then applied to create two parent-adolescent

positive affective quality groups: low positive affective quality group (< median) and high

positive affective quality group (

median) (Frazier et al., 2004; Lippold et al., 2011).

communication quality from their parents including negative and positive communication, such

as how often adolescents think their mothers/fathers refuse, even after discussion, to work out a

36

ideas for solving the problem. All items were recoded so that higher values mean higher levels of

create the composites of mother-adolescent communication quality and father-adolescent

from .77 to .90 across five waves. Similar to parent-adolescent positive affective quality, given

(Abar et al., 2014). To test the Hypothesis 3 that is whether the associations of parental

communication quality circumstances, a median split approach was then applied to create two

and high communication quality group (

median) (Frazier et al., 2004; Lippold et al., 2011).

Intervention Status

Intervention status (0 = Control, 41%, N = 401; 1 = Intervention, 59%, N = 574) that may be

associated with adolescent aggressive behavior problems (Spoth et a., 2000) was controlled in all

models as a covariate.

(Jorgensen et al. 2018; Mastrotheodoros et al., 2020), in order to validate whether the constructs

across waves were measured the same. Measurement invariance involves three components,

37

configural invariance is satisfied, this indicates that the same basic organization of one construct

item contributes to the latent construct to a similar degree across time. If the longitudinal scalar

capture all mean differences in the shared variance of the items. Together, if the longitudinal

Following measurement invariance tests, all four hypotheses were tested using Mplus

associations between adolescent aggression and parental monitoring as well as parental

associations between two or three variables (Kearney, 2003). The cross-lagged panel model

between two variables. However, during early- to mid-adolescence (aga 11.5 to age 15), parental

reported adolescent aggressive behavior problems (e.g., De Haan et al., 2012; Fowler et al.,

an Autoregressive Latent Trajectory (ALT) model (Bollen & Curran, 2004; Abar et al., 2014;

adolescent aggression with parental monitoring and parental knowledge, while simultaneously

38

Curran, 2004). Specifically, the ALT approach combines growth curve modeling with cross-

examined while controlling for change (or stability if there is not significant change) in the

model tests whether parental monitoring/parental knowledge at age 11.5 predicts adolescent

model for adolescent aggression is a latent-variable model, the ALT approach disentangles the

growth. The lagged association is actually a residual regression coefficient after accounting for

this model examines whether adolescent aggression at age 11.5 predicts parental

controlled by the growth models of parental monitoring/parental knowledge from age 11.5 to age

The hypothesized ALT model with a linear developmental trend for all constructs is depicted

slope, and quadratic terms were estimated to model the developmental trend of parental

age 15, separately. All factors were allowed to freely covary. The best growth model was

39

adolescent aggression and between parental knowledge and adolescent aggression were

constructs. Note the intervention status was controlled as a covariate in all models.

The Hypothesized Autoregressive Latent Trajectory Model with Linear Developmental Trend for

Note. The effects of mothering and fathering would be tested separately in this model. Intervention status would be

controlled as a covariate in all models. Bolded lines indicate the associations of interest in current study. I_PM =

intercept of parental monitoring, S_PM = linear slope of parental monitoring, I_Agg = intercept of adolescent

aggression, S_Agg = linear slope of adolescent aggression, I_PK = intercept of parental knowledge, S_PK = linear

equality constraints (i.e., only constraining cross-lagged paths to be equal between the high

40

between the free and constrained models indicated a moderation effect. In all the models,

CHAPTER IV: RESULTS

Preliminary Analyses

Descriptive Statistics and Comparisons between Mothering and Fathering

parental knowledge showed that the differences of parental monitoring and parental knowledge

between mother report and father report were significant across ages (ps < .001), such that

mothers always reported more parental monitoring and parental knowledge compared to fathers

from age 11.5 to age 15. Moreover, though the means of mother-reported adolescent aggression

were higher than the means of father report across ages (see Table 1), the differences between

mother report and father report were not statistically significant. The concurrent correlations

between parental monitoring and parental knowledge at each wave ranged from .16 to .40. The

Variance Inflation Factor (VIF) test (i.e., multiple regression models that using parental

monitoring and parental knowledge predicted subsequent adolescent aggression) indicates that

there was no collinearity issue including parental monitoring and parental knowledge in one

41

48

and adolescents is bad, maternal monitoring and knowledge are actually associated with worse

adolescent outcomes (i.e., higher levels of adolescent aggression). High positive affective quality

Table 6

The Standardized Beta Coefficients of Cross-Lagged Associations from a Multiple-Group

Autoregressive Latent Trajectory Model for Mothering

Mother-Adolescent Positive Affective Quality

49

Note. The solid lines indicate positive significant associations for a low level of mother-adolescent positive affective

quality group; the bolded and dashed line indicates a negative significant association for a high level of mother-

adolescent positive affective quality. Intervention status is controlled as a covariate in all models. Growth

parameters are not depicted for ease of interpretation (see Figure 2).

Father Model

For fathering, no difference between the high father-adolescent positive affective quality

group and the low father-adolescent positive affective quality group was found on the

group and the low parent-adolescent communication quality group was found on the associations

Moderation Effect of Adolescent Gender

50

associations between adolescent aggression with maternal monitoring and maternal knowledge

associations between adolescent aggression and maternal knowledge were found only for

associated with more maternal knowledge and in turn more maternal knowledge was associated

monitoring at age 14 was found to predict more adolescent girls’ aggression at age 15 (β = .11, p

Table 7

51

adolescent girls was found on the associations between fathering and adolescent aggression

CHAPTER V: DISCUSSION

Aggressive behavior problems during adolescence can lead to seriously adverse

consequences for both adolescents themselves and society (Loeber & Stallings, 2011; Kokko et

aggressive behavior problems with distinct parental monitoring and parental knowledge while

simultaneously controlling for the two dimensions of parenting. In addition, following the

contextual model of parenting style (Darling & Steinberg, 1993), the moderation effects of

parent-adolescent affective quality and parent-adolescent communication quality on these

associations were investigated. Last, given the mixed findings in adolescent gender differences in

the extant literature, the current study also tested whether these associations have adolescent

gender differences. Importantly, to provide a comprehensive picture on the interaction between

parents and adolescents and propose more nuanced suggestions on intervention programs, all

these foresaid associations and effects were examined by taking mothers and fathers separately

into consideration.

52

aggression was found, while no association between paternal knowledge and adolescent

have a positively predictive effect on adolescent aggression (i.e., more parental monitoring

aggression on parental monitoring was found (RQ1). Mother-adolescent positive affective

affective quality on the associations between adolescent aggression and fathering was discovered

monitoring predicted less adolescent subsequent aggression; whereas under a low level of

The quality of parent-adolescent communication, however, was not found to have any

adolescent aggression and mothering such that positive, bidirectional associations between

adolescent gender differences existed in the associations of adolescent aggression with fathering

After controlling for the development in marental knowledge and adolescent aggression

over time, a bidirectional association between maternal knowledge and adolescent aggression

53

associations between maternal knowledge and adolescent aggression were positive after

predicted more subsequent adolescent aggression and adolescent aggression also predicted more

methodologies used between current study and prior studies (e.g., Gaertner et al., 2010; Keijsers

panel models control for the stabilities of parental knowledge and adolescent behavior problems

knowledge (see Lionetti et al., 2019; Laird et al., 2003) and parent-reported adolescent

autoregressive latent trajectory model to test the bidirectional associations while simultaneously

Indeed, when accounting for the developmental course of parental knowledge and adolescent

bidirectional associations between parental knowledge and adolescent outcomes, which is in

may result in contradictory findings in the correlations’ direction (e.g., negative, bidirectional

autoregressive latent trajectory models), additional methodological research is needed.

Associations between Parental Monitoring and Adolescent Aggression

54

over time, only the effect of parental monitoring on adolescent aggression was found. This

predictive effective of parental monitoring on adolescent delinquency. However, inconsistent

found that both maternal monitoring and paternal monitoring predicted more adolescent

behaviors that may impede adolescent autonomy development (less parental monitoring). In

over time. In this process, more parental monitoring, which manifests as actively soliciting

adolescents may negatively respond to parental monitoring such that perceived more monitoring

McLeod et al., 2007). Parents more actively monitoring their adolescents may see an increase in

problems.

This study supports the integrative model of parenting style (Darling & Steinberg, 1993)

that mother-adolescent affective quality could moderate the relationship between parental

monitoring and parental knowledge with adolescent aggression. Under a high level of mother-

55

fewer subsequent adolescent aggressive behavior problems after controlling for the development

associations between adolescent aggression and maternal monitoring as well as maternal

These findings indicate that when the relationship between mothers and adolescents is bad,

quality, adolescents are more likely to rebel against parents and negatively respond to parental

with more adolescent aggression when the relationship quality between mothers and adolescents

parental behaviors (Abar et al., 2014; Lipplod et al., 2014; Keijsers et al., 2009). As a result,

study uncovers the importance of mother-adolescent affective quality in the interaction between

suggested to make the protective effectiveness of maternal monitoring valid.

knowledge, and adolescent aggression were found after controlling for the development in

56

mothering and adolescent delinquency (Keijsers et al., 2010). In the current study, the

bidirectional associations between maternal knowledge and adolescent aggression were only

knowledge and adolescent boys’ aggression were positive, such that adolescent boys’ aggressive

was associated with more subsequent adolescent boys’ aggressive behavior. A cross-gender

effect between mothering and adolescent boys’ behaviors emerged (Stolz et al., 2005; Rothbart

& Maccoby, 1966). During adolescence, boys generally have more aggressive behavior problems

adolescent active disclosure (Kerr et al., 2010). Through active disclosure, adolescent boys may

get some degree of relief from performing aggressive behavior problems. Thus maternal

knowledge about their adolescent boys may increase as boys perform more aggression. This

implies that mothers obtaining more information from adolescent disclosure may be a sign for

more their boys’ aggression. In addition, mothers may be stricter to their daughters compared to

sons (Radziszewska et al., 1996). Sons may feel less punitive than daughters if their mothers

know more about their negative information. Thus, more maternal knowledge may be associated

with more adolescent boys’ aggression, given the possibility of less fear of being discovered and

being punished. Additional research, such as interview studies, is needed to test these

possibilities.

aggression. No effect of adolescent girls’ aggression on maternal monitoring was discovered.

57

This implies that, during adolescence, parent-driven behaviors that are intended to obtain

information concerning adolescent girls are more likely to result in adverse behaviors such as

Girls may perceive higher levels of maternal monitoring compared to boys. As discussed before

et al., 2014; Grusec & Davidov, 2007), girls who perceive more maternal monitoring may

negatively react to their mothers’ behaviors (more aggressive behavior problems). However,

because boys may perceive lower levels of maternal monitoring compared to girls, maternal

1977). Thus there may be no responses from boys to their mothers’ monitoring.

adolescence. This reflects the fact that fathers are less involved in parenting compared to mothers

(Harris & Morgan, 1991). Partly due to less involvement of fathers in adolescent development,

the moderation effects of parent-adolescent affective quality and adolescent gender failed to be

model. That is partly due to that mothering and fathering in terms of the same parenting are