• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Prevention of early child disruptive behavior: A latent growth curve analysis

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Amanda M Pearl
Abstract:
Parent training (PT) is one of the most efficacious methods to prevent and treat disruptive behavior problems in children (Serketich & Dumas, 1996).Unfortunately, the exact mechanisms are not thoroughly understood and maintenance of gains following PT has been found to be lacking. The current study investigated the trajectory of child disruptive behavior and dysfunctional parental disciplinary styles following PT, as well as how these trajectories vary by child gender, ethnicity, and socioeconomic (SES) status. The results indicated that child disruptive behavior and dysfunctional parenting styles decreased over time. More specifically, child anger/aggression and dysfunctional parenting styles were found to decrease more rapidly for females, minorities, and high SES children. Implications for PT as prevention of child disruptive behavior are discussed.

TABLE OF CONTENTS

Page

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . iv

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . v

ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . vii

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1

Parent Training (PT) . . . . . . . . . . . . . . . . . . . 2 Factors Associated With Efficacy of PT . . . . . . . . . . . . 5 Child Gender . . . . . . . . . . . . . . . . . . . 5 Child Ethnicity and SES. . . . . . . . . . . . . . . . 7 Present Study . . . . . . . . . . . . . . . . . . . . . 9

METHOD . . . . . . . . . . . . . . . . . . . . . . . . 14

Participants . . . . . . . . . . . . . . . . . . . . . . 14 Procedures . . . . . . . . . . . . . . . . . . . . . . 15 Measures. . . . . . . . . . . . . . . . . . . . . . . 15 Child Disruptive Behavior . . . . . . . . . . . . . . . 15 Parent Disciplinary Style . . . . . . . . . . . . . . . 16 Attendance . . . . . . . . . . . . . . . . . . . . 17 SES Index . . . . . . . . . . . . . . . . . . . . 17

RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 18

Preliminary Analyses. . . . . . . . . . . . . . . . . . . 22 Latent Growth Analyses . . . . . . . . . . . . . . . . . 24 Two-Factor LGM . . . . . . . . . . . . . . . . . . 24 Multivariate or Associative LGM . . . . . . . . . . . . 26 Multiple-Sample LGM . . . . . . . . . . . . . . . . 27 Post-Hoc Analyses . . . . . . . . . . . . . . . . . . . 32

DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . 38

iii Page

Limitations . . . . . . . . . . . . . . . . . . . . . . 43 Implications . . . . . . . . . . . . . . . . . . . . . 45

LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . 47

APPENDICES

Appendix A. . . . . . . . . . . . . . . . . . . . . . 63 Appendix B. . . . . . . . . . . . . . . . . . . . . . 70

VITA . . . . . . . . . . . . . . . . . . . . . . . . . . 85

iv

LIST OF TABLES

Appendix Table Page

1. Descriptive Statistics . . . . . . . . . . . . . . . . . 63

2. Cronbach’s Alphas. . . . . . . . . . . . . . . . . . 63

3. Pearson Product Moment Correlations, Means, and Standard Deviations Among Predictor and Outcome Variables . . . . . . 64

4. Parameter Estimates and Fit Indices for Two-Factor LGMs . . . . 65

5. Parameter Estimates and Fit Indices for Associative LGM. . . . . 65

6. Fit Indices for Multiple-Sample LGMs . . . . . . . . . . . 66

7. Parameter Estimates for Multiple-Sample LGMs (ECBI – I) . . . . 67

8. Chi-Square Difference Test for Multiple-Sample LGMs (ECBI – I). . . . . . . . . . . . . . . . . . . . . 67

9. Parameter Estimates for Multiple-Sample LGMs (SCBE – AA) . . . 68

10. Chi-Square Difference Test for Multiple-Sample LGMS (SCBE – AA) . . . . . . . . . . . . . . . . . . . 68

11. Parameter Estimates for Multiple-Sample LGMs (PS – TOT). . . . 69

12. Chi-Square Difference Test for Multiple-Sample LGMS (PS – TOT) . . . . . . . . . . . . . . . . . . . . 69

v

LIST OF FIGURES

Appendix Figure Page

1. Univariate, 2-factor LGM . . . . . . . . . . . . . . . 70

2. Univariate, 3-factor LGM . . . . . . . . . . . . . . . 71

3. Multivariate or associative LGM . . . . . . . . . . . . . 72

4. Multiple-sample LGM . . . . . . . . . . . . . . . . 73

5. Trajectory of intensity of child disruptive behavior (ECBI – I) over time . . . . . . . . . . . . . . . . . . . . . 74

6. Trajectory of child anger/aggression (SCBE – AA) over time. . . . 75

7. Trajectory of dysfunctional parenting styles (PS) over time . . . . 76

8. Simultaneous trajectories of intensity of child disruptive behavior (ECBI – I) and child anger/aggression (SCBE – AA) over time . . . 77

9. Trajectories of intensity of child disruptive behavior (ECBI – I) over time by gender . . . . . . . . . . . . . . . . . 78

10. Trajectories of intensity of child disruptive behavior (ECBI – I) over time by ethnicity. . . . . . . . . . . . . . . . . 79

11. Trajectories of intensity of child disruptive behavior (ECBI – I) over time by SES level . . . . . . . . . . . . . . . . 80

12. Trajectories of intensity of child disruptive behavior (ECBI – I) over time by attendance . . . . . . . . . . . . . . . . 81

13. Trajectories of dysfunctional parenting styles (PS) over tim e by ethnicity . . . . . . . . . . . . . . . . . . . . 82

vi Appendix Figure Page

14. Trajectories of dysfunctional parenting styles (PS) over tim e by SES level. . . . . . . . . . . . . . . . . . . . . 83

15. Trajectories of dysfunctional parenting styles (PS) over tim e by attendance . . . . . . . . . . . . . . . . . . . . 84

vii

ABSTRACT

Pearl, Amanda M. Ph.D., Purdue University, May 2009. Prevention of Early Child Disruptive Behavior: A Latent Growth Curve Analysis. Major Professor: Judith C. Conger, Ph.D.

Parent training (PT) is o ne of the most efficacious m ethods to prevent and treat disruptive behavior problems in children (Serketich & Dumas, 1996).Unfortunately, the exact mechanisms are not thoroughly understood and maintenance of gains following PT has been found to be lacking. The current study investigated the trajectory of child disruptive behavior and dysfunctional parental disciplinary styles following PT, as well as how these trajectories vary by child gender, ethnicity, and socioeconomic (SES) status. The results indicated that child disruptive behavior and dysfunctional parenting styles decreased over time. More specifically, child anger/aggression and dysfunctional parenting styles were found to decrease more rapidly for females, minorities, and high SES children. Implications for PT as prevention of child disruptive behavior are discussed.

1

INTRODUCTION

Disruptive behavior problems, such as aggressive and oppositional behavior, are perhaps the mo st salient issue in child mental health today (Dishion & Patterson, 2006; Kazdin, 1987; Shaw, Dishion, Supplee, Gardner, & Arnds, 2006). Normative developmental pathways of disruptive behavior in children peak at approximately 2- years-old and thereafter show a decline as children mature and develop social and social-cognitive competence (Brownell & Hazen, 1999; Hartup, 1974; Tremblay, 2000). On the other hand, about 4% to 6% of preschoolers who present with significant behavior problems continue to have clinically significant behavior problems that remain stable or increase into the school-age years (Briggs-Gowan, Carter, Skuban, & Horwitz, 2001; Raver & Knitze, 2002). This is especially true for children who display more extreme levels of disruptive behavior (i.e., these children’s behavior problems are more likely to remain moderately stable over time; Campbell, March, Pierce, Ewing, & Szumowski, 1991; Fox, Henderson, Rubin, Calkins, & Schmidt, 2001). The majority of these children who display stable or increasin g significant levels of disruptive behavior into the school-age years also continue to have problems in adolescence and adulthood (Bub, McCartney, & Willet, 2007; Campbell, 2002; Caspi, Elder, & Bem, 1987; Farrington, 1983; Loeber, Lahey, & Thomas, 1991; Moffitt, 1993; Patterson, 1993; Robins & Ratcliff, 1979; Tremblay & Craig, 1995).

2 These problems include drug and alcohol abuse, family violence, crime, as well as comorbid psychological disorders. Given this information, identification of the factors that contribute to these differential trajectories of disruptive behavior in children is paramount (Hill, Degnan, Calkins, & Keane, 2006). Research examining predictors of disruptive behavior problems finds that parental quality, specifically pa rental disciplinary practice, may be the most important proximal factor influencing the development and maintenance of disruptive behaviors in children (Dadds, Heard & Rapee, 1992; Patterson, Reid, & Dishion, 1992). Nevertheless, the exact association is not understood at this time. Parents’ use of specific types of discipline, particularly inconsistent, harsh, or extremely lax discipline, is proposed to be associated with negative outcomes in children such as delinquency and aggression (Arnold, O’Leary, Wolff, & Acker, 1993; Deater-Deckard, Dodge, Bates, & Pettit, 1996; Maccoby & Martin, 1983; Patterson et al., 1992). In contrast, there is evidence that authoritarian discipline (i.e., firm, consistent, warm) is related to positive child outcomes (Dodge, Bates, & Pettit, 1990; Farrington & Hawkins, 1991; Patterson et al., 1992). Parent Training (PT) Given the important role of parenting in relation to disruptive behavior problems in child ren, parent training (PT) is one of the most efficacious methods to prevent, as well as treat disruptive behavior problems in children (Brestan & Eyberg, 1998; Hood & Eyberg, 2003; Reid, Webster-Stratton, & Hammond, 2003; Serketich & Dumas, 1996). One of the main tenets of PT is that it assumes all behavior is a function of reinforcements or punishments incurred from the environment. PT focuses on

3 parenting practices that directly influence or reinforce children’s disruptive behavior, such as parenting practices that are inconsistent, harsh, or permissive (Forehand, Wells, & Griest, 1980; Webster-Stratton, Kolpacoff, & Hollinsworth, 1988). Patterson (1997) maintains that the reinforcement of disruptive behaviors is the direct determinant for children’s aggression, and most determinants for aggression are found in the reactions provided by the social environment, rather than within the child (Patterson, 1982; 1997). In order to decrease child disruptive behavior PT focuses on substituting dysfunctional parenting styles with clear, firm, consistent and appropriate practices. In other words, the overall goal of PT is to modify a caregiver’s behavior, specifically by increasing positive reinforcement for desired behaviors and punishment or ignoring of undesired behaviors, so that this will in turn effectively increase his/her child’s level of compliance and decrease noncompliance (Serketich & Dumas, 1996). While the effects of PT are well supported by empirical and theoretical evidence, as prevention or an intervention for child disruptive behavior PT it is still far from completely effective. For some children, improvements in behavior will be documented directly following treatment, but will be lost quickly following the end of the intervention (Bank, Marlowe, Reid, Patterson, & Weinrott, 1991; Baum & Forehand, 1981; Patterson & Fleischman, 1979). In other cases, child disruptive behavior does not decrease at all in response to PT. This is particularly true for those children with more significant levels of disruptive behavior. Individual studies have found that at best the level of effectiveness of PT has positive ef fects on approximately two-thirds of participants (Webster-Stratton & Hammond, 1997). In a meta-analysis evaluating the effectiveness of behavioral parent

4 training, Serketich and Dumas (1996) found that approximately 80% of children whose parents participated in PT had more positive outcomes following treatment. While establishing the efficacy of PT is important, it is arguably more important to identify the factors that predict a decrease in dysfunctional parenting styles, as well as a decrease in child disruptive behavior, as a result of PT in specific groups of caregivers and children (Beauchaine, Webster-Stratton, & Reid, 2005; Brestan & Eyberg, 1998; Nock, 2003; Owens et al., 2003). Moreover, the salient issue then is to determine if those who do not respond to treatment have similar characteristics that predict resistance to change. As Brestan and Eyberg (1998) posited, for whom does PT work and when is PT not enough to lead to significant behavioral change? Given that not all families respond uniformly to parent training several studies have identified predictors present prior to the beginning of treatment that effect a participant’s response to PT (Beauchaine, W ebster-Stratton, & Reid, 2005; Kazdin, 1997; Sanders, 1992; Webster-Stratton & Hammond, 1990). These predictors have included children with comorbid anxiety/depression, parental psychopathology such as maternal depression, parental relationship satisfaction, life stress, and substance abuse (Baydar, Reid, & Webster-Stratton, 2003; Beauchaine et al., 2000; DeGarmo, Patterson, & Forgatch, 2004; Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000; Webster-Stratton & Hammond, 1999). Treatment effects of PT on parental disciplinary strategies also have been found to vary by child and family characteristics more traditionally associated with disruptive behavior, including child gender, ethnicity, and socioeconomic status (SES) (Deater-Deckard & Dodge, 1997; Dodge, Pettit, & Bates,

5 1994; Parke & Buriel, 1998; Rothbaum & Weisz, 1994; Webster-Stratton, Hollingsworth, & Kolpacoff, 1989). Currently there is considerable correlational and cross-section al research examining parental disciplinary strategies in relation to gender, ethnicity, and socioeconomic (SES) status (Deater-Deckard & Dodge, 1997; Dodge et al., 1994; Parke & Buriel, 1998; Rothbaum & Weisz, 1994; Webster-Stratton et al., 1989). However, there is uncertainty about the exact mechanisms and what specifically distinguishes these groups from one another. By examining the processes by which PT effects parental discipline and how parental discipline varies depending on factors such as child gender, ethnicity, and SES level, improvements may be made on efforts to improve prevention of child disruptive behavior. Factors Associated With Efficacy of PT Child Gender Males are at greater risk for and tend to display more disruptive behavior as compared to females (Bates, Bayles, Bennett, Ridge, & Brown, 1991; Keenan & Shaw, 1994; Maccoby & Jacklin, 1980; Mahoney, Donnelly, Lewis, & Maynard, 2000; Moffitt, Caspi, Rutter, & Silva, 2001; Zahn-Waxler, 2003). This gender difference tends to be minor before the age of 4-years-old and then appears to grow increasingly disparate in early childhood (Achenbach, McConaughy, & Howell, 1987; Keenan & Shaw, 1997; Maccoby, 1998). Approximately 5% to 10% of males, while only 1% to 3% of females, display clinically significant behavior problems during the late preschool and school-age years (Lavigne et al., 1996; Verhulst, 1995).

6 Even though differing prevalence rates of disruptive behavior by gender are well documented in the literature, research examining the influence of gender on parenting practices is limited (Kim, Arnold, Fisher, & Zeljo, 2005; Rothbaum & Weisz, 1994). Correlations between parenting disciplinary strategies and disruptive behavior problems differ depending on child gender. Specifically, the correlation between parental disciplinary style and disruptive behavior are stronger for males than for females (Deater-Deckard & Dodge, 1997; Leve & Fagot, 1997; Rothbaum & Weisz, 1994). In a meta-analysis examining the effects of physical discipline on child behavior Gershoff (2002) reported that the greater the proportion of girls in study samples, the weaker the association between harsh physical discipline and child aggression. It is possible that discipline varies by gender because of differing socialization and parental disciplinary practices (McLoyd et al., 2000). Leve and Fagot (1997) found that parents used different di sciplinary strategies depending on the gender of the child. Parents tend to overreact to behavior in children that is incongruent to gender stereotypes (i.e., aggressive behavior in females, shyness and dependency in males; Fagot, 1984). On the other hand, lax parenting is found to be associated with aggressive behavior in males and shy/dependent behavior in females (Kim et al., 2005). As a result aggressive behavior may be more acceptable and tolerated to a greater extent in males than females. This discrepancy is parenting styles may partially explain the higher incidence rate of aggression and general disruptive behavior as parents ma y be more permissive and lax of the behavior in males. It is possible that outcomes following PT, which aims

7 to make changes in parental disciplinary styles, may differ depending on the child’s gender. Caregivers with male children may not respond as rapidly and have poorer outcomes directly after and following PT than caregivers with female children. Child Ethnicity and SES Several studies have found ethnic and SES differences in children with disruptive behavior problems (Deater-Deckard & Dodge, 1997; Deater-Deckard et al., 1996; Skiba, Michael, Nardo, & Peterson, 2002). The highest reported rates of early- onset conduct problems and teacher-rated externalizing behavior are found in ethnic minority and low-SES families (Dodge et al., 1994; Parke & Buriel, 1998; Webster- Stratton et al., 1989). These increased rates of disruptive behavior in ethnic minority/low SES children have been posited to be due to differences in parental disciplinary styles (Parke & Buriel, 1998). African American parents endorse higher levels of physical or harsh discipline as compared to European Am erican parents, which is related to increased levels of disruptive behavior problems (Deater-Deckard & Dodge, 1997; Hill, Bush, & Roosa, 2003). However, some researchers have found that parental physical discipline is associated with disruptive behavior problems in European American children, but not in African American children (Deater-Deckard et al., 1996; Larzelere, Cox, Danelia, & Mandara, 2008). Similarly, spanking frequency predicted significantly more teacher- reported disruptive behavior, fighting, and antisocial behavior for European American children, but less problems for African American children (Gunnoe & Mariner, 1997; McLeod, Kruttschnitt, & Dornfeld, 1994).

8 This difference in outcomes can be explained by the group difference hypothesis in that it may be the acceptability of or normality of physical discipline in an in-group (e.g., minority status) that influences child outcome. In other words, the use of harsh physical discipline in African American children may not have as detrimental an effect or as great an influence on level of disruptive behavior problems as it does for European American in children (Boivin, Dodge, & Coie, 1995; Wright, Giammarino, & Parad, 1986). Even though these associations have been found, the specific effects of types of parental discipline on child outcome for minority/non- minority groups, including the effects of PT on these groups, it is not understood entirely at this time (Deater-Deckard & Dodge, 1997). Therefore, while ethnic non- minorities may have been found to have higher initial rates of disruptive behavior, this may not be accurate as it was confounded by SES. It is also possible that non-minority caregivers reporting harsh, more dysfunctional parenting styles may not have respond as rapidly and have poorer child outcomes (i.e., higher levels of child disruptive behavior) directly after and following PT. SES, or socioeconomic status, has been defined by several researchers as including factors su ch as marital status, income, and level of education (Webster- Stratton & Hammond, 1990). SES as defined by the combination of these factors has been associated with poorer treatment response, dropped out early from treatment, and had difficulty sustaining treatment effects over time following the end of PT (Dumas & Wahler, 1983; Kazdin, 1995; McMahon, Forehand, Griest, & Wells, 1981; Wahler, 1980; Webster-Stratton & Hammond, 1990). Therefore, low SES families may not

9 respond as well to PT as would those families from higher SES brackets (Serketich & Dumas, 1996). Given this information, it is important to examine the effect that ethnicity and SES level has on the outcom e of PT. Unfortunately, the effects of both ethnicity and SES are often difficult to disentangle as ethnic minorities are frequently overrepresented in low SES groups. It has been hypothesized that minority and/or low- SES parents may not fare well in PT programs that were originally validated using predominantly middle- or upper-class and/or European American samples as the parenting characteristics within these populations may vary (Hill, Soriano, Chen, & LaFromboise, 1995; Wood & Baker, 1999). In other words, given the possible difference between the effects of dysfunctional or harsh parental disciplinary styles between these groups, PT which aims to decrease harsh disciplinary styles with the goal of decreasing child disruptive behavior may not be as effective for one group as compared to another. Present Study Given that parent disciplinary style predicts disruptive behavior in children, and that researchers have found differences in levels of disruptive behavior problems in children depending on gender, ethnicity, and SE S level, the question remains as to what the specific relationship is between child disruptive behavior and these variables. It would be erroneous to assume that disciplinary strategies, as well as PT, have the same universal effects on caregivers with children of different genders, ethnicities, and of different SES groups (Deater-Deckard & Dodge, 1997; Rowe, Vazsonyi & Flannery, 1994). To the author’s knowledge although child disruptive behavior has

10 been examined at length in school-age children, only recently have researchers began to examine the trajectory, or change over time, of child disruptive behavior during the preschool age years (NICHD, 2004; Shaw, Gilliom, Ingoldsby, & Nagin, 2003; Gilliom & Shaw, 2004; Shaw, Lacourse, & Nagin, 2005; van Aken, Junger, Verhoeven, van Aken, & Deković, 2007). Within these studies few have examined any factors predicting differential outcomes for prevention and/or intervention programs for child disruptive behavior, specifically in preschool age children (Bub et al., 2007; Brestan & Eyberg, 1998; Nock, 2003). Latent growth models (LGMs) are an excellent way to examine research questions regarding change over time in that they com bine both group and individual level analyses, as well as individual variation in groups over time (Nagin, 2005; Ram & Grimm, 2007). Even more important, given significant variation in change over time LGM can test whether there are multiple trajectories for specific groups of individuals, rather than a single underlying trajectory (Duncan, Duncan, & Stryker, 2006). Often, researchers assume that data from different individuals provide comparable information, but it is often the case that individuals may be identified as belonging to certain groups (e.g., males/females, ethnicities, socioeconomic levels) where it may be appropriate to investigate whether multiple developmental pathways exist. Beauchaine et al. (2005) took advantage of the benefits of LGM to examine treatment outcom es for PT. They found that marital adjustment, maternal depression, paternal substance abuse, and child comorbid anxiety/depression influenced PT treatment response. It also was noted that critical, harsh, and ineffective parenting predicted treatment outcome. Specifically, parents who reported lower levels initially

11 and improved during treatment were found to respond best to PT. Similarly, DeGarmo et al. (2004) examined PT intervention effects on male internalizing and externalizing symptoms and maternal depression using LGM. They found that increases in effective parenting led to decreases in child behavior problems, as well as decreases in maternal depression. Finally, most recently and more generally, van Aken et al. (2007) used LGM to examine the trajectories of five parenting dimensions associated with externalizing behaviors (i.e., support, positive discipline, psychological control, lack of structure, physical punishment) in children under the age of 4-years-old. This study investigated the effects of parent training on child disruptive behavior problems and dysfunctional parenting style using LGM to examine change over time, as well as differences in change over time vary by child gender, ethnicity, and SES. The study hypothesizes that:

1. Over time, it is expected that there will be syst ematic change in the average or mean rates of disruptive behavior as a result of attendance to PT sessions. Specifically, it is expected that the rates of disruptive behavior will decrease over time. a. Both linear and curvilinear or quadr atic models will be tested to determine which trajectory best fits the data. b. A multivariate model of child disruptive behavior, operationalized by intensity of child disruptive behavior and child anger/aggress ion will be tested to determine whether change in one behavior covaries with the other over time.

12 2. Over time, it is expected that there will be systematic change in the average or mean rates of dysfunctional parenting styles as a result of attendance to PT sessions. It is expected that the rates of dysfunctional parenting styles will decrease over time. a. Both linear and curvilinear or quadr atic models will be tested to determine which trajectory best fits the data. 3. At Time 1, or baseline, it is expected that, on average, children with higher rates of disruptive behavior will be m ore likely to be male, African American, and from low SES families. 4. At Time 1, or baseline, it is expected that, on average, caregivers who report a m ore dysfunctional parenting style will have children who are more likely to be male, African American and from low SES families. 5. Directly following treatment and at a year following termination of treatm ent, it is expected that groups of children may show different trajectories over time. Different groups (e.g., child gender, ethnicity, SES level) will be examined to determine if there groups of children show different rates of change over time in response to PT. That is, are there groups of children/caregivers whose behavior decreases, remains stable, or show an increase in disruptive behavior/dysfunctional parenting style after parent training. a. Number of PT session attended will also be examined to determ ine if the trajectories of parents who attended a lower

13 number of sessions may have a different trajectory of change in child disruptive behavior and/or dysfunctional parenting style compared to parents who attended a high number of sessions.

Full document contains 106 pages
Abstract: Parent training (PT) is one of the most efficacious methods to prevent and treat disruptive behavior problems in children (Serketich & Dumas, 1996).Unfortunately, the exact mechanisms are not thoroughly understood and maintenance of gains following PT has been found to be lacking. The current study investigated the trajectory of child disruptive behavior and dysfunctional parental disciplinary styles following PT, as well as how these trajectories vary by child gender, ethnicity, and socioeconomic (SES) status. The results indicated that child disruptive behavior and dysfunctional parenting styles decreased over time. More specifically, child anger/aggression and dysfunctional parenting styles were found to decrease more rapidly for females, minorities, and high SES children. Implications for PT as prevention of child disruptive behavior are discussed.